Managing AEC project organizations at the edge of chaos: An analysis of AEC projects' adaptive capacity from a living systems perspective more

Managing AEC Project Organizations at the Edge of Chaos: An Analysis of AEC Projects’ Adaptive Capacity from a Living Systems Perspective by Beatrice C. Benne Architecture Diploma (University of Geneva, Switzerland) 1995 M.S. (University of California, Berkeley) 1998 A dissertation submitted in partial satisfaction of the requirements for the degree of Doctor in Philosophy in Architecture in the GRADUATE DIVISION of the UNIVERSITY OF CALIFORNIA, BERKELEY Committee in charge: Professor Yehuda E. Kalay, Chair Professor Mike W. Martin Professor Iris D. Tommelein Fall 2005 Managing AEC Project Organizations at the Edge of Chaos: An Analysis of AEC Projects’ Adaptive Capacity from a Living Systems Perspective Copyright 2005 by Beatrice C. Benne Abstract Managing AEC Project Organizations at the Edge of Chaos: An Analysis of AEC Projects’ Adaptive Capacity from a Living Systems Perspective by Beatrice C. Benne Doctor of Philosophy in Architecture University of California, Berkeley Professor Yehuda E. Kalay, Chair Architectural, Engineering, and Construction (AEC) projects are complex organizations that operate within complex, dynamic and uncertain environments. Project performance is based on the organizations’ ability to manage efficiently and deliver innovative solutions, thereby delivering value to clients. Conventional project management is grounded on the old mechanistic and deterministic belief that an organization and its environment operate as linear, cause-andeffect systems that, therefore, are predictable. Monitoring performance consists of identifying disturbances in the environment so as to predict change and work out a plan to manage it, thereby maintaining organizational equilibrium. However, conventional management strategies have been rather limited in their effectiveness for managing today’s project organizations in complex and uncertain environments. Complexity theory has shown that the behavior of complex systems cannot be predicted, because even the tiniest disturbance may be amplified over time by a nonlinear chain of cause-and-effects that cannot be foreseen. 1 From this theoretical perspective, an AEC project organization is a complex adaptive system and more specifically, a living system. A living system has the adaptive capacity to self-organize when faced with uncertainties, i.e., it can modify the connections among its components so that its overall behavior is better adapted to the new situation. This exploratory study uses understanding derived from living systems theory as a means to develop a new AEC project performance model that provides a framework for assessing an organization’s adaptive capacity. To illustrate, it describes two case studies, the International Terminal at the San Francisco Airport and the San José Martin Luther King Library, which were investigated through interviews with project participants in different disciplines and different companies that made up the respective project organizations. The case studies demonstrate that different project organizations respond to their challenges with different adaptive capacities that affect their ability to learn and self-organize. The higher the adaptive capacity of a project organization, the more value is delivered throughout the project life cycle and less waste is generated. Based on these findings, it is suggested that new project management strategies emerging from a deeper understanding of organizations as living systems have the potential to produce improved project performance. 2 To my parents, Evely and Claude Benne i TABLE OF CONTENTS Managing AEC Project Organizations at the Edge of Chaos: An Analysis of AEC Projects’ Adaptive Capacity from a Living Systems Perspective Chapter 1: Creating Value in AEC Project Organizations…………………………... 1 1.1. A Complex and Dynamic AEC Project Environment……………………………… 7 1.1.1. Stability……………………………………………………………………… 10 1.1.2. Complexity…………………………………………………………………... 10 1.1.3. Diversity……………………………………………………………………... 12 1.1.4. Hostility………………………………………………………………………13 1.2. AEC Project Organizations’ Adaptive Challenges………………………………... 14 1.3. Outdated AEC Business Practices and Beliefs……………………………………. 21 1.4. The Inheritance of an Old Worldview……………………………………………...29 1.5. From the Perspective of a New Science…………………………………………... 34 1.6. A New Performance Model for AEC Project Organizations………………............ 38 ii Chapter 2: Complexity Theory and Living Systems………………………………… 41 2.1. Paradigm of Complexity Theory………………….................................................. 43 2.1.1. Discovery of Chaos………………………………………………………….. 43 2.1.2. Strange Attractors…………………………………………………………… 46 2.1.3. The Edge of Chaos…………………………………………………………... 48 2.1.4. Theory of Dissipative Structures……………………………………………. 50 2.1.5. What is Life?.................................................................................................... 52 2.1.6. Summary……………………………………………………………………...55 2.2. Living Systems Principles………………………………………………………….56 2.2.1. Complex Adaptive Systems…………………………………………………. 57 2.2.2. Self-Organization in Living Systems………………………………………... 61 2.2.3. Fitness Landscapes………………………………………………...................64 2.2.4. Autopoiesis Theory…………………………………….................................. 66 2.2.5. Summary……………………………………………………………………...77 2.3. Complexity Theory Applied to Social Systems……………………….................... 79 2.3.1. Self-Similar Nature of the Creative Process in the Mind, Groups, and Organizations……………………………………………………………………. 81 2.3.2. Social Autopoiesis…………………………………………………………... 83 2.4. Organizations as Living Systems………………………………………………….. 85 Chapter 3: Managing AEC Projects at the Edge of Chaos………………………….. 86 3.1. AEC Project Organizations as Complex Adaptive Systems………………………. 87 3.2. Self-Organization and Adaptation in AEC Project Organizations…………........... 91 iii 3.3. Influencing the Control Parameters………………………….................................. 94 3.3.1. Rate of Information Flow…………………………………………………….96 3.3.2. Diversity between AEC Professionals’ Schemas…………………………...100 3.3.3. Richness of Connections…………………………………………………… 106 3.3.4. Level of Contained Anxiety………………………………………………... 109 3.3.5. Level of Power Differentials……………………………………………….. 112 3.3.6. Summary…………………………………………………………………… 114 3.4. The Power of Strange Attractors in AEC Project Organizations………………… 120 3.4.1. Balancing Positive and Negative Feedback Processes……………………...121 3.4.2. Synergy between Competition and Cooperation……………………………124 3.4.3. Creative Tension between Dominant and Recessive Schemas…………….. 130 3.4.4. Attractors’ Meaning………………………………………………………... 134 3.5. Testing out the Model……………………………………………………………. 135 Chapter 4: International Terminal at the SFO Airport…………………...……...... 136 4.1. Terminal Project Case Study Introduction……………………………………….. 138 4.1.1. Case Study Methodology…………………………………………………... 138 4.1.2. Project Timeline and Organizational Structure……………………………..139 4.1.3. Project Specifics…………………………………………………………….143 4.2. Challenging Environment……………………………………………………... 143 4.3. Path Dependence: Coping with Project History…………………………………. 145 4.3.1. Uncertainty of a Convoluted Design Team Selection Process……………...146 4.3.1.1. Preliminary Design Firms Selection Process………………………… 146 iv 4.3.1.2. Different Perceptions of the Environment…………………………… 149 4.3.1.3. Integrating Three Agendas in One Single Design Team……………...152 4.3.2. Limits of Coordination without Self-Organization………………………… 156 4.3.3. Discontinuity Due to Change of Leadership……………………………….. 160 4.3.4. Unsuccessful Fast Track Process Due to Late Design Changes…………… 162 4.3.5. Reactive Budget Management Strategy……………………………………. 164 4.3.6. Unforeseen Consequences of Decisions Made…………………………….. 168 4.4. Coping with Changes and Unforeseen Events…………………………………… 169 4.4.1. Coping with Economic Changes…………………………………………… 169 4.4.2. Coping with Mother Nature………………………………………………... 170 4.4.3. Coping with Fast Tracking and Program Changes………………………… 171 4.4.4. Difficult Adaptation……………………………………………………....... 175 4.5. Unsuccessful Integration of the Project team……………………………………. 176 4.6. Missed Opportunities?............................................................................................ 183 Chapter 5: San José Martin Luther King Library…………………………………. 185 5.1. King Library Case Study Introduction…………………………………………… 188 5.1.1. Case Study Methodology…………………………………………………... 188 5.1.2. Project Timeline and Organizational Structure……………………………..190 5.1.3. Project Specifics…………………………………………………………….194 5.2. Miracle on 4th Street……………………………………………………………… 194 5.3. Client’s Transformative Challenge………………………………………………. 198 5.3.1. Attractor of Meaning………………………………………………………..198 v 5.3.2. Creating Stability by Focusing on the “How”……………………………... 201 5.3.3. Fostering a Collaborative Environment……………………………………. 203 5.3.4. Unleashing the Distributed Intelligence in the Client Team………………. 205 5.3.5. Documenting Everything…………………………………………………... 207 5.3.6. Embracing Complexity through a Systemic Process………………………. 209 5.4. Keeping the Momentum Going through Design and Construction……………… 213 5.4.1. Creative Dialogue between Programming and Design…………………….. 213 5.4.2. Benefits of Participants and Process Continuity…………………………… 221 5.4.3. Flexible, Decentralized Communication and Decision-Making…………… 222 5.4.4. Building the Right Project Team Chemistry………………………………..224 5.4.4.1. Cherry-Picking Teammates…………………………………………. 225 5.4.4.2. Making Partnering Work...................................................................... 227 5.4.4.3. Nurture the Team and Reinforce Good Behavior……………………. 228 5.4.5. Balancing the Degree of Power Differential Among Professionals………...229 5.5. Adaptive Capacity through Innovations…………………………………………. 234 Chapter 6: Adaptive AEC Project Management and Leadership………………….237 6.1. Assessing and Embracing Project Complexity…………………………………... 239 6.1.1. Using “What If” Scenarios and/or Storytelling……………………………. 240 6.1.2. Scanning the Environment for Discontinuities and Conflicts………………242 6.1.3. Deferring decision until the Last Responsible Moment…………………… 243 6.2. Dynamic Contingency and Planning…………………………………………….. 243 6.3. Developing the Adaptive Capacity of AEC project Organizations……………… 246 vi 6.3.1. Diagnosing the Adaptive Capacity of a Project Organization…………….. 248 6.3.2. Influencing Control Parameters……………………………………………. 250 6.4. Building the AEC Industry’s Adaptive Capacity…………………………………259 6.4.1. Adaptive Leadership……………………………………………………….. 260 6.4.2. Adaptive Skills for AEC Professionals and Firms………………………… 258 6.4.2.1. Systems Thinking……………………………………………………. 265 6.4.2.2. Facilitation and Process Management Skills………………………….266 6.4.2.3. Self-Reflection………………………………………………………...268 6.4.2.4. Emotional Intelligence……………………………………………….. 269 6.4.3. Implications for Educational and Professionals Institutions………………. 270 6.4.4. Aligning Contracting Models with Adaptive Management Practices……... 273 6.5. Living Systems Theory’s Implications for Further Research…………………… 276 6.6. Delivering Value through Adaptive Project Management………………………..277 Notes…………………………………………………………………………………… 281 References……………………………………………………………………………... 289 vii LIST OF FIGURES AND TABLES List of Figures: Figure 1.1a: Single-Loop Learning …………………………………………………….. 19 Figure 1.1b: Double-Loop Learning……………….…………………………………….19 Figure 2.1: Principles of Self-Organization in Living Systems……………………….. 79 Figure 3.1: Organizational Variables’ Influence on the Control Parameters………….. 96 Figure 3.2a: Variables Influencing the Rate of Information Flow…………………….. 115 Figure 3.2b: Variables Influencing Agents’ Schema Diversity………………………...116 Figure 3.2c: Variables Influencing the Richness of the Connections…………………. 117 Figure 3.2d: Variables Influencing the Level of Contained Anxiety…………………...118 Figure 3.2e: Variables Influencing the Level of Contained Anxiety………………….. 119 Figure 4.1: International Terminal Project Timeline…………………………………. 142 Figure 5.1: King Library Project Timeline…………………………………………… 193 Figure 5.2a: Traditional Linear Project Life Cycle……………………………………. 220 Figure 5.2b: King Library Project Life Cycle…………………………………………. 220 viii Figure 6.1a: Assessing the Rate of Information Flow…………………………………. 252 Figure 6.1b: Assessing Agents’ Schema Diversity……………………………………. 253 Figure 6.1c: Assessing the Richness of Connections…………………………………. 254 Figure 6.1d: Assessing the Level of Contained Anxiety………………………………. 255 Figure 6.1e: Assessing the Level of Power Differentials……………………………… 256 List of Tables: Table 1.1: Sources of Project Complexity and Uncertainty……….................................. 14 Table 4.1: Terminal Project Participants………………………………………………. 143 Table 5.1: King Library Project Participants……………………………….................. 194 ix ACKNOWLEDGEMENTS I owe the development of the theoretical foundation for this dissertation, as well as the completion of the work, to individuals who insightfully provided direction at critical times in the process, when I needed it the most. These mentors and friends shared their knowledge and kindly offered information – the name of a paper or book, a new contact, or a fresh perspective about the direction of my research – that often provided me with the missing link in the development of my ideas. I am deeply indebted to Michael Wakelin, Fellow emeritus at Bechtel Corp. Michael is one of those rare renaissance spirits who possess an unlimited capacity of the mind to build meaningful connections between any possible subjects. I have greatly enjoyed our intellectual exchanges, both face-to-face and on the phone, which always rejuvenated and inspired me. Michael was the first person who encouraged me to consider AEC project organizations as living systems, and for this, I will always be grateful. x Kyle Davy and Susan Harris, both Senior Fellows at the Advanced Management Institute for Architecture & Engineering in San Francisco, reemphasized the legitimacy of the living systems approach. Over a lunch in Berkeley during the summer of 2003, Kyle indicated that I could use the San José King Library project as a potential case study. The following week, he kindly introduced me to Anderson Brulé Architects (ABA) – one of the design firms that participated in the King Library project. This exciting case study supported the case that AEC project management can benefit from the living systems approach. I met Susan Harris months later at a discovery workshop hosted by ABA and facilitated by Susan. The topic of Susan’s presentation that day was “A Search for New Models of Practice” (now the title of a new book that she and Kyle co-authored). The workshop that brought together CEOs from a dozen architectural firms, who were all eager to discover new ways of delivering innovative architectural services, was incredibly inspiring. Later on, in a couple of follow-up dinners in Piedmont, Oakland, Susan was instrumental in defining key concepts related to living systems theory and in pointing out references whose readings anchored my research and provided structure to my writing. I am very grateful to the project case studies participants for kindly consenting to participate in my research; freeing up their busy schedule to meet with me; and candidly answering my questions. At the International Terminal at the San Francisco Airport: Peter Bittenbender from the Bureau of Design and Construction at the SFO Airport; Keith Boswell from Skidmore, Owings & Merril L.L.P.; Martin Del Campo from Del Campo & Maru; John Draguesku from Parsons; Michael Kaechner from Tutor-Sliba Corp.; and Michael Willis from Michael Willis & Architects. At the San José King xi Library: Pamela Anderson-Brulé from Anderson Brulé Architects; Patricia Breivik from the San José State University Library; Art Heinrich from San José State University; Jim Klingensmith from Gilbane Co.; Kevin Krumdieck from Carrier Johnson; Jane Light from San José Public Library; Sam McBane Mulford from Anderson Brulé Architects; Dolores Montenegro from the San José Redevelopment Agency; and Greg Overhage from Hensel Phelps. The many stories they recalled from their respective projects brought life to the writing and meaning to my personal analysis. I owe the quality of the writing of this manuscript to my friend Linda Aldridge who read numerous drafts and very carefully edited each one of them. With great humor and endless encouragement, she made the load of the task lighter. The final editing touches are from Kah Ying Choo who did a remarkable professional job in the finalization of this work. I want to thank my colleagues at Bechtel for their encouragements and Bechtel’s management in the Information Systems & Technology group for letting me take the time that I needed off from work to complete this dissertation. I would not have been able to complete my Ph.D. without their support and flexibility. My advisors, Yehuda Kalay, Mike Martin and Iris Tommelein have been very patient along the journey. I apologize to them if my exploration of topics has, at times, confused them, even though, intuitively, I always knew – not the details or the form – the general direction toward which I was headed. I wish to particularly thank Iris for her meticulous review of the final drafts. Her editing suggestions have much to do with the clarity of the final work. xii I also want to acknowledge my friends: Brinda Dalal, Lisa Esherick, Cate Gable, Barbara Phillips, Lisa Sullivan and many others for their encouragement throughout the journey. I want to express my deep gratitude to Didier Bouvet for “taking the heat” in the midst of my frustration and anguish and for always being there to comfort me and gently push me forward. Finally, I would like to recognize my parents in France who never doubt that I will complete this dissertation. Of the many things I have learned from them, the love of hard work and the uncompromising tenacity required to realize any vision, however daunting, have been instrumental to the completion of this dissertation. I love them dearly. xiii xiv CHAPTER 1 Creating Value in AEC Project Organizations Multi-million-dollar facilities are the product of complex architectural, engineering and construction (AEC) project organizations operating within a larger social, economical, political and technical context. For those of us who are somewhat familiar with design and construction processes, the delivery of a facility to a client may seem like a miracle, so great is the amount of coordination required among the members of a building enterprise and so overwhelming are the challenges faced by a project organization. When one considers the complexity of the average AEC project environment, one may wonder how any project manages to be built. At the same time, upon closer examination of the project delivery processes, one will also realize that, while most facilities have indeed been completed and delivered to their clients (except, perhaps, on rare occasions of aborted programs), the completion of many of those projects was very costly, not only in terms of money, but also in terms of waste, conflicts generated and often, the lack of overall value delivered to the clients. 1 Consider, for instance, the construction of the International Terminal at the San Francisco Airport. In the September 8, 2000 edition of the San Francisco Business Times, journalist Steve Ginsberg depicted the challenges faced by the Terminal project team in the following words: All major airport construction jobs are messy and complex. The construction problems at new airports in Denver, Hong Kong and Kuala Lumpur made international headlines with their delayed openings and lost luggage. The $3 billion International Terminal expansion at San Francisco International Airport is no exception. The opening has been delayed several times, there are cost overruns of about $100 million and the FBI is investigating the hiring of minority contractors who have just a small piece of the project. When SFO officials announced Aug. 11 that the Sept. 26 opening would be missed by at least six weeks, airport officials blamed two subcontractors who, in turn, blamed the master contractor. Everybody has blamed El Niño rains for part of the delays. SFO’s construction problems are unavoidable. The logistical nightmare of building a 2.5 million-square-foot international terminal along with new roadways in the midst of the world’s ninth-busiest airport is an immense construction job and among the largest in Bay Area history. The project’s numbers were intimidating. There were 18,000 piles driven with 40,000 tons of steel erected. An army of 3,000 workers was employed through 84 contract packages. They toiled in only 50 acres… In contrast, now consider this second example: the Martin Luther King, Jr. Library project in San José. This project’s program was to meet the needs of a “twoheaded” client: the San José University Library and the San José City Library – two completely different organizations in terms of cultural and operational needs. Becky Bartindale (1998), staff writer for the San José Mercury News, describes the early negotiation days of the design and construction of the library: The project’s chief planners presented their latest thoughts on combining the two library operations Monday afternoon and evening forums at the San José McEnery Convention Center… Many of the queries revealed anxiety about the proposal to put a $171 million library at the corner of San Fernando and Fourth streets…At 475,000 square feet and with 3,600 seats, it would be one of the largest public buildings downtown… 2 While there are several big joint city-university libraries in Europe, there are no comparable shared libraries in the United States… The project is unique, assistant city manager Darrell Dearborn told the audience. “There is not a cookbook based on successful experiences in other large cities,” he said. “But it’s clear the opportunities for collaboration, efficiencies and mutual benefit are very significant.” But there also are “challenges, problems and accommodations” that must be addressed, he said… Faculty and students are worried about having to compete for university materials in a city with more than 800,000 potential library users. They worry about erosion of the academic mission and the library’s studious atmosphere. City users…worry whether they and their children would be welcome on campus and where they would park. But without a collaboration between the city and the university, Dearborn said, ‘there is no prospect in the foreseeable future of the city having any significant improvement to its main library.’ The same is true, he said, for the university. While the Terminal project incurred many delays and overruns, the Library project was delivered below budget, on schedule and without a single claim. Nevertheless, as the article excerpt above demonstrates, the Library project organization faced multiple challenges from the project’s inception onwards: not only had the merger of two libraries never been attempted before in the United States, the two clients’ needs and requirements were also very disparate. These two projects, while very different in nature and scope, nevertheless, have similar characteristics. First, they are both products of the work of complex organizations, involving a large number of firms, professionals and agencies working with one another over many years to complete the job. Second, these project organizations operated in very uncertain, dynamic, if not turbulent, environments that disturbed their stability; thus, they both faced the challenge of having to adapt to discontinuities in their environments. Third, these project organizations were embedded in broader social-political-economic contexts that both influenced the projects’ 3 operations: each project organization had to take into account this added complexity in its management strategy. Large, complex, multi-million-dollar AEC projects such as the Library and the Terminal must deliver value to their clients within increasingly challenging business contexts. In order to deliver value in today’s 21st century environment, not only do AEC project organizations need to manage projects efficiently (i.e., deliver projects on budget and on schedule, while fulfilling their clients and end-users’ requirements), they also need to innovate, both in terms of process and products delivered. Process innovation is required for a project organization to successfully cope with project complexity and unexpected events. Product innovation is required for a project organization to deliver a creative solution that often goes beyond the traditional delivery of a facility. Indeed, because of the complexity of the social and political environment that surrounds many projects, it may occur that the most creative solution may be to not design a physical product; rather, a change in social behavior or values, or other forms of learning people acquire in the process, may create the highest form of value (Heifetz 1994). Thus, the question arises: what are the mechanisms by which AEC project organizations deliver great value to clients and achieve high performance levels within complex and dynamic environments? Answers are provided by diverse organizational theories that have evolved to respond to organizations’ needs to improve performance in a particular socio-economical context. Such theories attempt to make sense of the mechanisms by which organizations coordinate work and aim to understand the parameters that affect performance. Based on the theoretical knowledge of organizations, 4 management philosophies and techniques have been developed, which provide recipes for organizations to apply to their environments in order to improve productivity. Traditional organizational and economic theories more or less share the same paradigm: they build on the unchallenged assumptions that: (1) equilibrium is necessary for an organization to perform; and (2) an organization can devise strategies to control its destiny in complex and uncertain environments. These assumptions are grounded in the old mechanistic and deterministic belief that an organization and its environment operate as linear, cause-and-effect systems that are predictable. In order to preserve its equilibrium, therefore, an organization must identify disturbances in its environment so that it can predict change and work out a plan to manage it. Conventional project management consists of monitoring performance by assessing the gaps between the actual state of an organization at any given time and the desired state. When a discrepancy is discovered, actions are taken to bring the state of the organization back to the desired state and thus maintain equilibrium (Stacey 1996). This conventional project management strategy may be effective in relatively stable environments. However, this strategy has often culminated in the tardy delivery of projects that are over-budget. The effectiveness of standard management strategies seems rather limited for managing today’s project organizations in complex and uncertain environments. In contrast to mechanistic and rationalistic approaches, complexity theory that emerged in the mid-20th century has shown that the behavior of complex systems cannot be predicted, because even the tiniest disturbance may be amplified over time through a nonlinear chain of cause-and-effects that cannot be foreseen. Faced with uncertainties, a complex system must have the ability to self-organize, i.e., to modify the 5 connections among its components so that the system’s overall behavior is better adapted to the new situation. Self-organization requires a system to learn from its environment. Learning is an information-intensive process; a system uses information available in its environment as a driver of change. Thus, a self-organizing system is a creative system. The outcome of self-organization is the emergence of new system structures that are better adapted to the environment. Research in biology and evolution has shown that living systems exhibit the properties of complex adaptive systems. Living systems operate within a larger ecosystem and creatively adapt to environmental changes by modifying the way their own components interact with one another and with the whole ecosystem. While organizations in the industrial era have been modeled as machines of production operating in relatively stable environments, in today’s knowledge-based economy, this traditional model does not accurately depict organizations operating in complex and dynamic environments. Interestingly, organizations more closely resemble living systems operating within a larger ecosystem and interacting with it by exchanging information; in doing so, organizations create an “information ecology” (Davenport, cited in Schrage 1997). As we abandon the idea of a predictable, linear, cause-and-effect world and embrace the concepts of complexity and living systems theories, new opportunities arise with respect to understanding the behavior and management of complex project organizations. Within this new theoretical context, uncertainty is not a cause for fear, but instead, an opportunity for innovation. Self-organization provides the mechanism by 6 which an organization can become creative and develop new structures that will allow it to achieve higher-level performances. This chapter is organized as follows: first, we present the characteristics of today’s AEC project environment. Next, we discuss the adaptive challenges faced by AEC project organizations – the challenges of delivering innovative solutions efficiently within complex and dynamic environments. Third, we investigate the reasons why AEC project organizations have had so much difficulty in overcoming their adaptive challenges. Fourth, we explain in more detail the origins of the conventional mechanistic, rationalistic and deterministic worldview grounded in Cartesian and Newtonian thinking, which has influenced Western society since the 17th century. Fifth, we contrast the conventional worldview with complexity theory and argue that the “new science” provides a more appropriate approach for understanding the behavior of complex systems in today’s dynamic environment. Finally, we present the main questions addressed in this dissertation and provide the overall organizational structure of the remaining chapters. 1.1. A Complex and Dynamic AEC Project Environment AEC project organizations are complex social systems managing numerous, intricate processes in ever-changing environments. Their complexity is primarily due to the large number of professionals involved in the planning, design and construction of buildings. Given the fact that the AEC industry often involves the collaboration of disparate, specialized professions with complementary expertise, Cuff (1996) estimates that the number of professions involved in a project can range from 32 to 67. AEC professionals 7 typically work together in a large building enterprise that provides services as diverse as designing, building, financing, regulating and operating the built environment (The American Institute of Architects 1994). The discrete nature of the AEC industry is sustained by contractual structures. The many professionals involved in projects (owners, architects, engineers, contractors, suppliers, financial experts, regulatory agencies and so on) are linked by various types of contractual relationships (e.g., traditional design-bid-build, design-build, joint venture, alliances). These relationships determine the project phases in which each participant gets involved; the division of responsibilities; and the way in which each discipline performs work and coordinates work with others. Contracts are temporary and remain in place until each project team member has completed his/her tasks and the buildings are delivered to clients. Thus, a project organization can be considered to be a newly created entity, as the client, project scope, professionals involved and processes used – either alone or in combination – are always unique. Contracts, however, can add significantly to the cost of a project and do not guarantee value to the client. Indeed, it is very difficult to enforce contracts when project performance measures are based on subjective values or based on different interpretations of the content of the contract itself. Consequently, a complex project seldom gets delivered without the occurrence of claims and lawsuits, which, of course, damage the reputation of the firms involved in the project and cost clients even more money. Furthermore, the transience and heterogeneity of an AEC project organization affect the level of cooperation and types of social interactions among team members. They often are the source of friction for several reasons: First, project participants’ 8 identification with their project organization may not be complete. Second, each professional firm may try to satisfy its own agenda, often at the expense of the other firms or of the overall project. As a consequence, such behaviors create uncertainties, as the optimization of a given discipline’s performance does not necessarily lead to the optimization of the whole project. The separateness of the AEC industry is not only functional, but also geographical and temporal. Project participants are often located in offices in different cities and regions, if not countries and continents. Thus, a project team can be considered to be a virtual project organization or project network whose members need to exchange information, make decisions, perform and coordinate work, resolve conflicts and so on in a distributed environment. Organizational issues become even more complex when professionals get involved abroad where local business rules, cultures and practices may greatly differ from their own. In such a distributed and interdisciplinary environment, the coordination of activities is complex and vulnerable to errors because the information each discipline needs to make informed decisions is not always available. One may say that the professionals’ reasoning process is ‘bounded’ by the information available at a given time. The limit of “Technical Rationality” – i.e., the inability of professionals to have access to the information required to make rational decisions – is another source of project uncertainties (Schön 1983). Moreover, AEC project organizations cannot be considered in isolation from the larger business ecosystem in which they operate. Their complexity is not only influenced by their inherent internal structures, but also by their environment and society as a whole. 9 Four environmental characteristics impact an organization’s ability to perform: the market’s stability, complexity, diversity and hostility (Mintzberg 1979). These characteristics are briefly reviewed below within the context of the AEC industry. 1.1.1. Stability A business environment is considered stable when it is predictable. From this standpoint, AEC project organizations operate in an extremely unstable environment. The health of the AEC industry varies with the health of global economies and markets. In recent years, large client organizations have attempted to develop long-term relationships and agreements with design and construction firms. However, it remains difficult for AEC professionals to anticipate customer demand as it fluctuates with the overall health of regional and national economies and increasingly, the global economy as well.1 In addition, AEC projects are vulnerable to unexpected natural events that may delay construction or ruin previous work. For instance, the fluctuations in the price of construction materials may dramatically impact a project budget. 1.1.2. Complexity The AEC industry environment is extraordinarily complex in terms of clients’ expectations; society’s needs as a whole; building requirements; and technologies used (e.g., work processes, tools and materials). Clients are often complex organizations that are represented by committees involving occupants, users, consultants, financial experts and so on. They have high expectations in terms of the breadth of architectural services they require; the quality of services and products they need; and the type of delivery 10 methods used by the project team. They are becoming increasingly sophisticated and often have a great deal of experience with building projects to the point of getting intimately involved themselves with the building process. Society in general is also becoming more aware of design issues. Design is no longer a monopoly of the elite: everyone, today, is requiring higher quality products. Finally, new global issues such as sustainability are forcing AEC industry participants to design and construct facilities that consume fewer resources, generate less waste and minimize their negative impact on the environment. In fact, the complexity of the facilities’ requirements reflects that of the society in which we live. Their design must take into account many different aspects such as the economic context; needs of end-users and the general public as a whole; a myriad of regulations; sustainability issues and so on. The integration of multiple subsystems (e.g., financial, legal, cultural, architectural, structural, mechanical and technological) with one another requires innovative project organizations that are able to deliver innovative services efficiently. For instance, Norman Foster’s design for the Commerzbank in Frankfurt, Germany, dubbed the “first ecological skyscraper,” illustrates how architectural innovation may combine new design typology with new technologies and ecological considerations to create a completely new work environment experience for the building’s users. In his design, Foster integrated an advanced environmental control system; natural daylight and ventilation; an atrium reaching from the ground to the top floor; and multiple four-storied garden landscapes distributed throughout the building.2 Because of the complex interrelationships among building subsystems, a project such as the Commerzbank challenged AEC professionals to holistically consider all the phases of 11 a project (planning, program development, design, construction and operation) throughout all phases of the project. Information technology, of course, is taking an increasingly important role in the delivery of projects. It has the potential to substantially affect AEC industry business practices because efficient information sharing is critical to project performance. The use of 2D and 3D Computer Aided Design (CAD) is getting slowly generalized in most offices. Even 4D CAD (space plus time dimensions) is starting to be used in some projects to facilitate resource planning and task coordination on construction sites.3 In addition, the Internet facilitates the exchange of information across distributed project teams in real time, thus compressing time. The ubiquity of the Internet makes irrelevant the location of professionals involved in a project; however, while information technology facilitates coordination of distributed activities, it also amplifies the complexity of the work environment by providing professionals access to a huge amount of disciplinary information, which increases the level of interdependence among project subsystems. 1.1.3. Diversity The AEC market is also highly diversified. To stay competitive, AEC professionals must provide a broad range of services to disparate customers whose needs and objectives – as we have previously shown – are also diverse and complex. Therefore, professional practices are disparate. Firms’ sizes vary from small offices with only a few employees to large offices that have thousands of employees working in multiple locations around the world. Some professionals specialize in a particular type of work, which they 12 sometimes deliver like a commodity (e.g., retail or office work), while others remain generalists and provide high-end custom designs (e.g., museums or libraries). Some firms deliver a broad scope of integrated services (e.g., financing, pre-design consulting, design and architecture, construction services and post-design services), while other practices remain specialized (e.g., design and/or construction). 1.1.4. Hostility Because the AEC industry is so dependent on the economy’s health, it operates in a very competitive environment. AEC firms must search for clients, compete for work and bid for contracts. Limited client budgets and low fees often force professionals to take risks – for instance, bidding at a price lower than is possible for delivery. Those risks often jeopardize a project team’s ability to achieve a high-level project performance. Moreover, since professional firms are usually involved in multiple projects at any given time, they are forced to compete against one another for access to resources within the larger AEC industry ecosystem. These competitions constitute another source of AEC project organizations’ uncertainty. The above discussion has illustrated the complexity of the AEC project environment and the types of uncertainty faced by AEC project organizations and professionals. The sources of complexity and uncertainty are captured in Table 1.1. 13 Sources of Project Complexity • Ecosystem (e.g., diversified market, complex client organizations, complex requirements and myriad of regulations) Number of stakeholders involved Heterogeneity (e.g., professionals’ and other stakeholders’ culture, behavior, roles and responsibilities, expectations, performance metrics and so on) Fragmentation (geographical and temporal) Distributed coordination processes Diversified services Subsystems’ interdependence Information Technology • • • • • • • Sources of Project Uncertainty Transient nature of projects Diversity of culture and behaviors Distributed environment Limit of technical rationality Unpredictable business environment Unexpected events (natural, economic and process-oriented) Competitive market (low margins) • • • • • • • Table 1.1: Sources of Project Complexity and Uncertainty While the industry context is becoming more and more dynamic and uncertain, disparate customers are requesting that AEC professionals provide innovative ideas and integrated project solutions, rather than single services (e.g., design). These solutions, of course, must be provided at a low cost and within a very compressed timeframe. Thus, the challenge for AEC project organizations is threefold: (1) to deliver innovative solutions and high value to clients; (2) to achieve the previous goal efficiently in order to maximize the performance of the outcome of their projects; and (3) to manage the complexity and uncertainties in their environments. In the next section, we discuss what this three-tier challenge means for AEC project organizations. 1.2. AEC Project Organizations’ Adaptive Challenges The efficient delivery of innovative solutions within complex and uncertain environments is an intricate challenge. First, experts in the industry have suggested that AEC 14 professionals should aim to achieve 20 to 40 percent increases in overall project delivery efficiency (Greenway Consulting 2002). Waste in all its forms should be minimized (e.g., waste in time and money, or waste due to a lack of reuse of information). Fewer conflicts and claims should also become a goal, as they directly impact project performance; however, unexpected events always come to disturb the smooth management of project activities, thus creating discontinuities that are costly to project performance. Because AEC project organizations often operate on the assumption that they work within a linear cause-and-effect environment. In other words, they assume that contingencies can be anticipated through better planning. However, the rigid framework of traditional project management processes, as we argue below, impedes AEC project organizations’ flexible adaptation to discontinuities in the environment. Second, while uncertainty is usually perceived as something that needs to be minimized as much as possible, one should not underestimate its power as a source of creativity. Indeed, the pursuit of innovation may not always be a goal per se; on the contrary, it may arise as a by-product – for instance, when solving an unexpected problem that requires a project organization to devise innovative solutions and/or processes. In such a case, professionals must have the ability to flexibly adapt to contingencies by breaking out of routines and standard work processes imposed by standard project management and using inter-disciplinary learning as the source of creativity for solving the problem. This requires organizational agility and responsiveness to change, which are supported by strong communication and professional leadership. 15 Third, the processes supporting innovation and efficiency seem to be very much at odds. Innovation requires the integration of human minds, knowledge and technology. Indeed, innovation is a human activity that depends on out-of-the-box thinking. While innovation may traditionally have been defined as the outcome of the work of a single genius, the complexity of today’s problems requires multiple experts to collaborate with one another to integrate their disparate knowledge into innovative solutions. Moreover, innovation is not always technical; it often requires individuals and groups to re-assess their belief and value systems and embark on a transformative journey. Thus, the processes supporting innovation are not only interdisciplinary and non-linear, but also require flexible management structures that support learning and adaptation. In contrast, project efficiency revolves around the implementation of effective project management tools and methods to make a project environment predictable and controllable: that is, the role of project management is to keep a project within a stable or equilibrium state. Project management activities are very broad and have different facets such as organizing, planning, coordinating, managing and controlling: • The process of organizing includes defining a project’s organizational structure, the project delivery methods and professionals’ liabilities; facilitating teambuilding; and developing a project organizational culture. • Planning is the process that enables an organization to achieve the level of predictability required to control its destiny. The planning process is based on rational, analytical thinking. It is believed that by gaining more information about the environment, one may be able to foresee instabilities and subsequently, devise strategies that take these instabilities into account. Planning consists of defining 16 project goals; allocating project resources; sequencing the execution of design, procurement and construction operations; and anticipating activities related to the closeout of a project and the start-up of the facility. It also includes establishing the project performance criteria; and identifying potential risks. • The process of coordinating includes defining the framework for the exchange of information; the roles and responsibilities of the professionals involved; and the decision-making processes. • Finally, the processes of managing and controlling consist of monitoring activities throughout the project life cycle; controlling the use of resources; managing risks; and ensuring that the overall project performance meets the goals and performance measures established in the planning phase (Laufer & Hoffman 2000; Pinto 2003). Therefore, traditional project management tools that support linear processes are highly rigid; they are based on the standardization of work processes, as well as the implementation and enforcement of standard procedures and guidelines, which dictate the specific way in which activities need to be executed. They also define proper communication channels among participants. Project control systems focus on making sure activities are executed as planned and any deviation from the plan is immediately corrected; control is usually centralized and administered by a few key participants in a project organization. This traditional process is depicted in Figure 1.1a as, what is called in organizational behavior, “single-loop learning.” A single-loop learning focuses on correcting a situation based on a given set of operating norms and standards that have 17 been defined during the planning process (hence, those standards are based on correcting events that have been foreseen at the planning phase). It is the case, however, that single-loop learning is ill-suited for the management of complex and dynamic project environments because projects are always unique and characterized by complex interrelated patterns among many subsystems that greatly limit a project organization’s ability to foresee all contingencies (Morgan 1997). In complex organizational settings, rational actions and their long-term effects are lost in the complexity of interactions among the various components of an organization and its environment. The success of rationally-conceived plans is not therefore guaranteed, because it is impossible to predict the outcome of actions that may depend on strategies devised by other components interacting both within the organization and in its greater environment. All project activities cannot be mapped out beforehand at the planning phase because project activities evolve based on the project context, the particular problems to solve, the evolving project requirements, contingencies and so on. Thus, project organizations must be able to implement “double-loop learning” (represented in Figure 1.1b). Double-loop learning is a three-step process of: (1) scanning the environment to detect critical changes that may impact a system; (2) challenging the ability of the established structure and norms to effectively respond to the environment; and (3) allowing a new organizational pattern to emerge. Double-loop learning is a creative process because it forces an individual or organization to challenge the status-quo; to self-organize; and to adapt to discontinuities in the environment. 18 Environment Activity Performance Plan, Rules & Standards Control Response Performance Gap Figure 1.1a: Single-Loop Learning Environment Activity Performance Plan, Rules & Standards Do standard rules apply? Control Response Adaptive Response Yes Performance Gap No Self-organize and innovate Figure 1.1b: Double-Loop Learning The rigidity and linearity of project management activities impede the free flow of information and the flexible collaboration required for a project team to innovate and adapt to discontinuities in the project environment. Indeed, the usual project 19 management response in the face of contingencies is to tighten control on performance and ensure that the pre-established procedures are implemented so as to keep a project within pre-defined parameters, i.e., to maintain the equilibrium. This behavior that typically results from the fear of unexpected consequences often triggers resistance to change. Resistance to change in a complex environment consequently results in exacerbating problems, thus creating more fear and hostility among the participants (Stacey 1996). Today’s management paradigm, argues Stacey (1996), leads to a vicious cycle. In our attempt to design organizations so that they yield successful outcomes, we attempt to rationally analyze our environment and design organizational structures that fit the context. However, the increasing complexity of our environment reduces our ability to foresee and control, which makes us anxious about failure. In our pursuit to remain in control, we jump from one management recipe to another, as none really works over the long haul. Despite all this, we somehow produce order and innovation (i.e., the anarchy we expected to occur when we are not in control actually does not take place); this confuses us even more. Thus, AEC project organizations face the following paradox: on the one hand, they must monitor project performance in order to ensure that the project organizations will meet the cost, schedule and other performance criteria. On the other hand, project management tools and processes impede project organizations’ ability to flexibly adapt to environment discontinuities and limit their innovative capabilities that are critical for delivering value to clients. 20 However, this paradoxical reality does not need to be. Its existence is grounded on the old, common belief that one can, indeed, control a project’s destiny from the outset and manage complexity through better planning and tighter control. This idea, however, is an illusion. AEC projects are complex social systems whose behaviors cannot be anticipated because project subsystems interact with one another and with the environment following nonlinear patterns, which over time create outcomes that cannot be predicted. The more control we apply to keep complex systems within the parameters of our pre-defined plans, the more our actions are amplified in unexpected ways, which, in fact, increase uncertainties. Managing complex AEC project organizations requires adaptive management capabilities, i.e., a project organization must be able to learn from its environment and self-organize. The result of self-organization is the emergence of new processes and structures that are better adapted to the situation. These adaptive capabilities require AEC professionals to cooperate and trust one another; to utilize a holistic and integrative approach to project management; and to utilize a flexible organizational structure. Unfortunately, as we will show in the next section, outdated AEC business practices and beliefs often hinder AEC project organizations’ adaptive capacity and thus, limit their ability to overcome adaptive challenges. 1.3. Outdated AEC Business Practices and Beliefs The adaptive capacity of AEC project organizations is limited by outdated business practices and beliefs that restrict professionals’ ability to leverage the knowledge network within which they operate and to integrate project subsystems into value-added solutions 21 for clients. Current AEC business practices originate from the separation of design from construction, which occurred in the late 19th century in the U.S. This created what we know today as the triangular relationship among client, architect and contractor – a triangular relationship that seals the separation between capital, art and technique. Today, the schism is functional, legal, organizational and cultural (Thomsen 1999). As a project organization is now subdivided into different subsystems managed by different professionals having different legal responsibilities, conflicts of interest arise, which exacerbate the already adversarial relationships among the stakeholders on the project. Consequently, project delivery models such as design-bid-build are sequential, hierarchical, competitive (if not hostile) and characterized by a long delivery time. As an alternative to the traditional design-bid-build model, the design-build project delivery model has been developed to facilitate the integration of A/E professionals and construction professionals into one team with a single contract. Owners have found multiple advantages in design-build delivery methods. First, the design-build model provides the owner with a single source of accountability. This removes the adversarial relationships between designers and contractors. The risk is thus shifted from the owner to the design-build entity. Second, construction issues are addressed early in the design phase, thus allowing the project team to resolve problems as they arise and therefore, eliminating costly disputes between A/Es and contractors that are typical of the design-bid-build system. Risk is minimized and the project may potentially be of better quality. Third, the design-build process facilitates innovations aimed at reducing costs. In the traditional design-bid-build process, architects, engineers, consultants, procurement specialists, suppliers and so on work independently within their 22 disciplinary scope and have very few incentives to coordinate their work to develop innovative ideas so that value is created for the whole project. Conversely, in the designbuild method, the designers, contractors and other participants have an opportunity to collaborate in the value engineering process. Design alternatives are assessed together and the one that best answers the owner’s requirements is selected by consensus. Fourth, the integrated nature of the design delivery method may result in early project completion so that owners can enjoy the earlier utilization of their facilities and thus earn higher return on their investments. Finally, the design-build team may provide owners with extended services such as financing for a turnkey project. In spite of its advantages, the design-build model is, nevertheless, insufficient to support the delivery of value to clients. While design-build contracts are often associated with a reduction in project costs, reports in Engineering News-Record, however, show that actual cost performances vary from one project to another. Each design-build project is different and success depends on how well participants are able to collaborate and implement flexible design and construction processes. The traditional architectural community has argued that the design-build delivery method is appropriate only for projects that are simple, straightforward, industrial and time-sensitive (The Haskell Company 1997). The design-build method is often used to fast-track projects. The disadvantage is that the projects often become constructiondriven. When a contractor exerts pressure on designers to minimize costs, it is often at the expense of design creativity, aesthetics and overall quality. Design-build proponents argue that the design-build process can provide a balance between beauty and function. The controversy around the design-build delivery process may be due to the way the 23 model is implemented in practice: for instance, owners with design expertise often provide the design team with construction documents that are 40 to 60 percent complete. This “draw-build” process, as Ricketts (1999) calls it, defeats the purpose of the collaborative “design-build” process whose advantage resides in the integration of all required expertise early enough in the project so that value can be created for the owner. This is also the viewpoint of Paul Doherty (2001), vice-chair of the American Institute of Architects’ Design Build Professional Interest Area, who states that: The term “design-build” is being used to describe many different project procurement methods. The fear of an easy description has many architects involving themselves in what they are told is a design-build project, when in fact it is not better than build design, or in extreme cases, ethical extortion of values…When design is not the leading value of a design build project, there are inherent problems that will manifest themselves as the project progresses. If a design build project is not design-led, (not designer-led, design-led) it is doomed for failure. Just don’t do it. Finally, it should be noted that while the design-build delivery process is an attempt to better integrate design and construction activities, the management methods used throughout a project’s life cycle are still sequential: a project remains a mere assembly of sub-components. Shareholders often do not have a real understanding of how to create value for the customer, as they focus on achieving predetermined objectives without an ongoing assessment at every phase of the project life cycle of the level of value of the solutions proposed. As Howell (2001) points out: …efforts such as the Design-Build movement have tried by changing contracts and relationships to shift the optimization target from the activity back to the project. But effectiveness of this contracting approach remains limited by the same underlying assumption; a project is a set of sequentially dependent activities with such low uncertainty that it can be managed with central control. But a project isn’t so simple and maintaining the fantasy only causes problems. 24 Outside of a systemic understanding of project management, the benefits of the design-build delivery system are very limited. To compensate for this deficiency, a lean management and philosophy evolved based on an integrated, systemic and holistic approach to project management, which aims at generating value to clients and minimizing waste in all its forms.4 Lean management was first developed in Japan as a new model of production in the automobile manufacturing industry. In more recent years, the lean management philosophy has been applied to the context of the design and construction of AEC projects. It aims at providing a solution to the “simplistic conceptualization of work and a deficient definition of control” inherent in traditional project delivery systems (Howell 2001, p.1). This philosophy focuses on six essential points: control; performance; project delivery; value; coordination of action through pulling and continuous flow in information; and decentralization. Control is defined as “causing events to conform to plan and triggering learning and replanning,” while planning, one of the essential components of control, is a means of “defining criteria for success and producing strategies for achieving objectives” (Howell 1999, p.1). Thus, lean philosophy takes a proactive approach to management, in which planning and control are part of the same ongoing process. Lean management is based on concurrent engineering, that is, the simultaneous design of a facility and its production processes with the goal of improving the performance of the whole versus the independent performance of the parts. Lean management integrates the design and making of products into a single, integrated 25 framework, i.e., the way the work is to be performed on the construction field is an inherent component of the definition of objectives, which drives the design process. To create this integrated process, it is necessary that downstream project participants (e.g., contractors, suppliers, etc.) be included in the up-front planning and design of the facility. This explains why the lean delivery method focuses on the management of design-build projects. In order to minimize waste creation, lean management focuses on pull-driven processes so that only what is necessary to the downstream process is produced upstream. Lean management emphasizes the decentralization of decision-making processes in order to improve project participants’ capabilities to react to eventual contingencies. To provide team members with access to the entire production system, lean management facilitates information sharing and faster feedback loops. Progress can be better monitored and interdisciplinary conflicts better anticipated: therefore, dependencies issues can be resolved earlier in the project life cycle. While it is based on a systemic approach, the lean management philosophy mainly focuses on supply-chain management, especially the management of the resources (information or materials) that flow among components of a project. While resource management and information flow are critical to the adaptive capacity of a project organization, there are other organizational characteristics that need to be taken into consideration in adaptive work – for instance, the richness of the relationships between the participants; the level of anxiety created by uncertainties; the degree of power differentials among participants and so on – which influence the creative and adaptive capacity of a project organization. Such issues are not addressed by lean management. 26 Moreover, the lean management approach does not explicitly consider a project as a self-organizing, living system. Thus, it does not take into account the social structures of a project – i.e., the vision, values and culture of the professionals involved and of the project as a whole – as critical organizational characteristics that influence a project team’s adaptive capacity and thus, which need to be managed. Green (1999, p.24) has argued that lean management is sometimes implemented as a coercive method, instead of an enabling method. Instead of being based on “loyalty, empowerment and consensus,” it is sometimes implemented as “management-by-stress and exploitation.” This is why integrating the value and culture of the project organization is critical to ensure the integrity of the project organization. In addition, lean management theory investigates neither the self-organizing and emergent properties of complex projects nor what these properties mean for the adaptive capacity of an AEC project organization. Because the lean project delivery system focuses on supply-chain management along the project life cycle, it does not provide an approach to support transformative work required by some clients (as the case study of the San José Library will demonstrate), i.e., project work that is ill-defined, requires learning, and/or for which there is no technical solution. Transformative projects often have deep and complex implications for the environment and the ecosystem in which they are embedded. For instance, the construction of a new transportation system may have critical impact on community relocation, taxes, social equity of the solution and so on. All these issues are non-technical in nature and need to be addressed prior to the design and construction of the transportation system. 27 Finally, critics have also argued that lean management proponents apply their philosophy without proper consideration of the project context. For lean management to be effective the project delivery has to be design-build; but there are many examples of projects that have been successfully delivered under the traditional design-bid-build process. Beyond the effectiveness of project delivery models, cultural issues and beliefs have a great impact on AEC project organizations’ ability to integrate their activities and implement collaborative practices. For instance, the myth of the architect as the “individual genius,” who is the sole owner of creativity and innovation and who tops the hierarchy of the project organization is still present in the AEC industry. Based on this idea, other project participants’ roles are reduced merely to executing the ideas of the “genius” and to ensuring the conformity of the building to his/her specifications. This model is a far cry from reality. As Dana Cuff (1996, p.13) notes in the introduction of her book Architecture: The Story of Practice, the “myth of the autonomous architecturalhero,” which she had thought to be a reality during her graduate study, was not confirmed when she observed professionals in action. Design and construction activities are the outcome of a very complex social process in which the architect, whose design creativity is absolutely critical, is nevertheless only one component of a greater system. As Cuff (1996, p.13) notes, [T]he design of our built environment emerges from collective action…the artifacts of practice, buildings, are socially constructed by the hands of individual architects, their coworkers, the organizations they work within, the array of contributors from clients to consultants and their colleagues, and by larger socioeconomic forces that affect the profession. 28 The old myth of the autonomous designer has diverse negative consequences: it creates an unrealistic and unnecessary hierarchy within a project organization; it undermines the leadership behavior that may be exercised by other project participants; and it exacerbates the dichotomy between design and construction activities. Interestingly, the lack of cooperative behavior which is evident in Western society, in general, and in AEC professional culture, in particular, may have deep origins. As Miles et al. (2000, p.316) explains, People in Western societies have been taught that self reliance is a virtue and that firms should pursue the maximization of their own interests and returns. Therefore, the idea that one might be just as concerned about the returns of a collaborator as about one’s own returns does not sound ‘reasonable.’ At a deeper level, Western civilizations over the past centuries have evolved two basic philosophies: (1) the libertarian, which emphasises individual responsibility and freedom and is the underlying logic for neoclassical economics, and (2) the socialistic, which emphasises collective ownership and returns based at least partially on need. The concept of collaboration, with its demands for both individual and shared responsibility, does not fit squarely into either of these philosophies. A new or hybrid philosophy would be required to fully justify collaborative behaviour among individuals and firms. The lack of understanding of AEC project organizations as complex social systems and knowledge networks operating in complex and dynamic environments is grounded in a rationalistic and mechanistic worldview which pervades Western society. The origins of this traditional worldview are presented in the next section. 1.4. The Inheritance of an Old Worldview Over centuries, worldviews and paradigms have influenced science and common beliefs within society as a whole. As Kuhn (1962) pointed out in the first edition of his book The Structure of Scientific Revolutions, the particular paradigm we embrace dictates our scientific research methodologies and the types of questions we ask and, consequently, 29 drives science toward the development of specific solutions.5 The interactions between worldviews and science are, in fact, complex and mutually determined. On the one hand, science is embedded within the matrix of society and, thus, as explained by Kuhn, worldviews influence the direction of scientific research; on the other hand, scientific breakthroughs, once they are disclosed to the broader audience – most often using powerful metaphors to facilitate laymen’s understanding – pervade all aspects of our society: our mental models, our language, our technical developments and ultimately, our behavior. Since the 17th century, Western worldviews have been influenced by two important figures in philosophy and science: the French mathematician and philosopher René Descartes (1596-1650) and the English physicist Sir Isaac Newton (1643-1727). The combination of Cartesian and Newtonian ideas has been so powerful that it has shaped not only science but the mental models and metaphors that guide our social behaviors as well. From Descartes, Western society inherited two main and related philosophical and scientific concepts: the idea of the mind-body dichotomy; and the concept of “rational analysis.” With the well-known assertion “I think, therefore I am,” Descartes argued for a disembodied mind having no influence from or on the body. This powerful idea allowed Descartes to reject the existence of any subjective reality. From a Cartesian perspective, the essence of humanity is rationality, that is, our ability to think logically, to set goals for ourselves, to make decisions among different alternatives, and so on. Rationality is dictated by universal principles which consider “rational choice” as being literal, logical, disembodied, dispassionate, and consciously calculable. Knowledge of 30 the world, therefore, is always gained independently of experience through rational analysis. To discover the “Universal Truth,” Descartes provided us with a reliable “scientific method:” to gain knowledge about an object one must subdivide that object into parts small enough to be known by our rational mind; the knowledge about the whole object is, therefore, reduced to the sum of knowledge about the parts. Cartesian philosophy has influenced generations of scientists for 300 years and is at the basis of the greatest scientific discoveries of our time. Gleick (1988, p.14) points out that, “Implicitly, the mission of many twentieth-century scientists – biologists, neurologists, economists – had been to break their universes down into the simplest atoms that will obey scientific rules.” Newton’s discovery of the Universal Laws of Motion gave mankind precise, predictive power over nature. The universal law of gravity, for instance, explains both why an apple falls and why the moon maintains its orbit. As Gleick (1988, p.12) notes: “…the laws of motion provide a bridge of mathematical certainty. Understand the laws and you understand the universe.” Newton’s laws are deterministic: they imply that for any given system, the same initial conditions will always produce the same output. In other words, we can always determine the output of a system if we know its initial conditions and vice versa. At the end of the 18th century, the French mathematician and physicist Simon Laplace (1749-1827) built on the earlier Newtonian theory to depict the universe as a huge clockwork mechanism. The belief that universal laws can explain all the mechanisms of the universe was very appealing to seventeenth century Western society and it quickly became pervasive.6 Following the footsteps of physics and natural science, the deterministic worldview influenced thinking in physics, astronomy, 31 philosophy, economics, psychology and politics; by the early twentieth century, it also influenced the emerging social sciences. Determinism and the power of rational thinking have influenced economics and traditional management approaches. Neoclassical economics theory is grounded in the work of famous philosopher and economist Adam Smith (1723-1790). In “An Inquiry into the Nature and Causes of the Wealth of Nations,” Smith (1776) depicts the economy as a “well-oiled machine,” in which, theoretically, the overall allocation of resources is globally optimal because resource allocation is based on the rational behavior of selfinterested agents, whose objective is to maximize their own benefits. It is also on the ground of rational thinking that Frederick Taylor (1856-1915) built his “Scientific Management” theory. Taylor (1911) showed that productivity of workers could be increased through rationalization of the work, i.e., by making workprocess standardization explicit. This concept meant that the ways in which the work was performed; the tools which were used; and the time that it took to perform the work required to be analyzed and standardized to meet predefined criteria. The principles of the rationalization of work have been implemented in the vertical corporation to create a linear work environment, i.e., an environment that is predictable and controllable by the management hierarchy. Even today, the bureaucratic form of most organizations leverages the following principles: hierarchy of control; specialization of functions; centralization of information and control; behavior governed by formal rules; codification of work processes in policies and procedures; and so on. For the German sociologist Max Weber (1864-1920), the bureaucratic form of organization is a powerful tool in that it creates an environment that encourages people to 32 respect and follow the rules, thus making them more efficient. Thus, Weber’s model of bureaucracy focuses primarily on authority structures that consist of top-down command and control. The value of this bureaucratic model resides in its predictability, efficiency, volume of production and accountability (Weber 1957). In his foreword to Arie de Geus’s book, The Living Company (1997), Peter Senge points out that our Western mental models are so much ingrained in the Newtonian and Cartesian worldviews that all our thinking and actions are driven by mechanistic and reductionistic assumptions. Similarly, Margaret Wheatley (1992, p.6) observes: We manage by separating things into parts, we believe that influence occurs as a direct result of force exerted from one person to another, we engage in complex planning for a world that we keep expecting to be predictable, and we search continually for better methods of objectively perceiving the world. While the scientific management approach and the bureaucracy model have both greatly helped organizations in the industrial era to increase their productivity, these approaches have important limitations in today’s dynamic business environment. These forms of management are especially ineffective for dealing with innovation, change and environmental complexity because formal processes do not provide the flexibility required for an organization to adapt to contingencies and to coordinate activities that are based on knowledge exchanged (Burns & Stalker 1961; Thompson 1967). The management of complex systems requires a new framework that abandons the idea of a predictable world operating as a linear cause-and-effect system. Complexity theory that started to develop in the mid-20th century provides a new, holistic approach that facilitates understanding of behavior in and of complex systems. We next present 33 the origins of this new science that is dramatically transforming the way our society perceives the world and interacts with it. 1.5. From the Perspective of a New Science The spectacular technological developments we have witnessed since the 17th century have proven the success of the reductionistic and analytic procedural method of “classical” science; nevertheless, by the middle of the 20th century, the complexity of new practical and theoretical problems faced by science and society as a whole has forced the scientific community to admit the limitations of classical scientific methods in interpreting complex natural phenomena. The scientific paradigm shift that consequently occurred was driven by the convergence of factors such as the development of information technology, especially the development of powerful computers that simulate the behavior of complex systems; new technological endeavors such as ballistic missiles and spacecraft, which require collaboration among diverse disciplines; biological research related to metabolism and homeostasis; the discovery of quantum physics; the emergence of new problems such as air and water pollution, traffic congestion, city planning and so on. These developments share the same general characteristic: they are all very complex because of the many interactions between their constituent parts and the understanding of such systems requires a new theoretical framework. Complexity theory whose main concepts started to crystallize during the mid-1980s at the Santa Fe Institute – a think tank that brought together an impressive array of scientists from different fields for collaborations with one another – provides an alternative framework to understand the behavior of complex systems. 34 Complex systems are nonlinear systems, i.e., their output is not proportional to input. They have emergent properties that are not exhibited by the parts, but that arise from the interactions and relationships between the parts. Thus, one cannot predict the behavior and performance of complex systems by using traditional, rational and analytical methods of classical science. Dealing with such complexity requires a holistic approach that, in contrast to the reductionistic method, considers the interactions between the parts as critical to the understanding of the behavior of the whole system. This new approach is commonly expressed by saying that in a complex system, “the whole is more than the sum of its parts.” Complex systems are adaptive systems: they self-organize to adapt to their changing environments. Through the interactions between their parts, they create emergent structures. They do so at the edge of chaos – a paradoxical state of simultaneous order and disorder, stability and instability. Instability is required to trigger change and adaptive behavior, while stability is needed to prevent the disintegration of systems in the face of uncertainties. Complex adaptive systems do not merely respond to change in the environment, but try to take advantage of discontinuities to create ordered patterns and evolve to a higher level of organization (Waldrop 1992). The understanding of the behavior of such complex systems developed quickly, thanks to increased collaboration among scientists from diverse fields. It is important to acknowledge that until the 1940s, scientists in different domains of research worked in isolation from one another. They did not share their findings, nor did they build connections between their research and the potential application of their findings to other fields. However, by the late 1940s, this isolation started to change as biologists such as 35 Ludwig von Bertalanffy began to apply principles from one field of research to another. The development of generic models that can be applied to different domains (e.g., the concept of homeostasis, which was first applied to physiological processes, was extended by von Bertalanffy to encompass a wide range of phenomena, from single cells to whole populations) provided a driver for the development of an interdisciplinary field called “general system theory” (Russell 2000). The principal investigator of general system theory, von Bertalanffy (1969, p.37), argues that there are systemic principles and laws, which are isomorphic to diverse fields such as physics, chemistry, biology, psychology, cognitive science, social science and so on. Thus, the aim of general system theory is the “formulation of principles that are valid for ‘systems’ in general, whatever the nature of their component elements and the relations of ‘forces’ between them.” It thus promotes the integration between various sciences, both natural and social. For instance, complexity theory revolutionized research in biology and evolution because the self-organizing and adaptive properties of complex systems exemplify the properties of life. Therefore, living systems can be considered to be complex adaptive systems that thrive at the edge of chaos. The understanding of biological processes in terms of complex dynamics from which emerge characteristic patterns of order was a radical shift from the conventional notion of a world evolving randomly based on natural selection. More recently, economists and organizational theorists started to consider the economy and organizations respectively as complex adaptive systems (Arthur 1988; Stacey 1996; Wheatley 1992). Some business organizations consider themselves as living systems evolving within a larger industry ecosystem. For instance, Arie de Geus (1997) wrote The Living Company based on his professional learning, while working for 36 Royal Dutch/Shell; and Visa, the well-known credit card institution, was founded by Dee Hock on the principle of a self-organizing, adaptive and nonlinear complex system (Dee Hock 1999). Most interestingly, in the late 1980s, the U.S. Army, an organization that is famous for its rigidity and the implementation of hierarchical structures, initiated the implementation of adaptive and self-organizing strategies as an alternative to its conventional command-and-control model. The need to develop a more organic approach in the army was driven by the increasing amount of information available to troops – information that they needed to make decisions in real time on the field (e.g., information about location of the enemy; the nature and targeting of the enemy’s weapons systems and so on). Within the framework of the “Commander’s Intent,” which defines the scope of an engagement, fighting units may exploit arising opportunities and self-organize in a way that may differ from what was planned.7 This being said, it is important to remind ourselves that complexity theory applies to complex and dynamic environments, i.e., environments that cannot be predicted. The traditional mechanistic approaches are perfectly appropriate in stable environments where equilibrium prevails. We may argue that even in today’s business environment, organizations perform at least some of their activities within relatively stable environments; therefore, the knowledge and understanding of traditional management methods remain valuable. Moreover, as we discuss at length in the remaining chapters, because self-organization and creativity occur at the edge of chaos, i.e., in a state of simultaneous stability and instability, it is the synergy and tension between traditional coordination and management methods and adaptive, self-organizing activities that will allow AEC project organizations to innovate and achieve efficiency. Understanding how 37 an organization may leverage the synergy between both types of management approaches is critical to achieving quality performance in the 21st century. What can we learn from the findings in complexity and living systems theories for the management of AEC project organizations? This is the question we attempt to answer in the remainder of this dissertation. We state the hypothesis that AEC professionals’ understanding of their project organizations as complex living systems will transform the way professionals behave, both individually and in groups. New behaviors, in turn, may change the final outcomes of AEC professionals’ actions in ways that, to be consistent with our new worldview, we cannot foresee. The benefits of the new worldview and its management implications have yet to be fully demonstrated. However, there is no reason to believe that the new paradigm will cause a decline in the quality of performance anymore than the implementation of our current theories that are grounded on outdated interpretations of the world and traps us in an illusion of predictability and control. 1.6. A New Performance Model for AEC Project Organizations In the following chapters, we consider AEC project organizations through the lens of complexity theory and living systems theory. From this perspective, a project organization is a complex adaptive system that is embedded within a larger ecosystem and interacting with it. When faced with environmental uncertainties, a project organization creatively self-organizes and develops new patterns of interaction among its components. The outcome of the adaptive capacity of a project organization is the emergence of a new structure that helps the project organization to thrive in dynamic 38 environments. Self-organization and emergence occur at the edge of chaos – a paradoxical state of simultaneous stability and instability. In order to develop a new model of organizational performance, which is grounded on a framework provided by complexity theory, we need to redefine our research questions. While the mechanistic perspective focuses on equilibrium, the complexity perspective focuses on self-organization at the edge of chaos; therefore, the following questions arise: What are the forces that push an AEC project organization to the edge of chaos and keep it there long enough for the project organization to selforganize and innovate? What are the sources of stability and instability in complex AEC projects? How do the processes of self-organization and emergence operate in an AEC project organization? What are the benefits of adaptive and self-organizing AEC project organizations? These questions will first be answered theoretically through a review of complexity and living systems theories. Based on this theoretical framework, a new model of AEC project organization management will be defined. This model will be tested through the case study analyses of two projects mentioned earlier in the chapter: the Terminal and the Library projects. Both projects, while different in nature and scope, were complex and faced high levels of uncertainties; each used different management approaches to manage those uncertainties. The remaining chapters are organized as follows: Chapter 2 presents the main theoretical concepts from complexity and living systems theories and shows how these concepts can be applied to the understanding of social systems and organizations. 39 Chapter 3 builds on the concepts presented in Chapter 2 to develop a model of AEC project management at the edge of chaos. Findings in organizational theories will be tied back to complexity theory to demonstrate how a project organization may devise strategies to ensure it has the adaptive capacity required to manage complexity and uncertainties. Chapter 3 also demonstrates that managing at the edge of chaos requires a synergy between seemingly opposing forces such as cooperation and competition; positive and negative feedback and so on. Chapters 4 and 5 present the case studies and analyze them based on the model defined in Chapter 3. The analysis of the Terminal project will show the limits of the traditional project management model, especially the inability of a project team to adapt to unexpected events, when utilizing rigid coordination methods, and the consequent creation of waste for the project as a whole. In contrast, the analysis of the Library project will show how flexible, adaptive coordination and project management methods facilitated the emergence of a project culture that benefited the entire project in all its different phases. This case study also demonstrates how a project team may minimize the creation of waste along the project life cycle. Chapter 6 summarizes the findings and discusses the implications of complexity and living systems theories for project management; AEC professionals’ skills development; and the role of AEC institutions in improving the adaptive capacity of the AEC industry. Suggestions on ways to improve the adaptive capacity of a project organization are also provided. 40 CHAPTER 2 Complexity Theory and Living Systems To consider social systems as living systems requires a profound change in our worldviews. In order to explore this fundamental paradigm shift, this chapter presents the theoretical framework for developing a model of organizations, which is based on living systems theory. The theoretical framework is grounded in complexity theory – the broader theory against which living systems theory is researched. Complexity theory provides a new perspective on the process of evolution, which complements the broadly accepted, yet incomplete, explanation provided by Darwin (1809-1882) that evolution is driven by random mutations of organisms adapting to the environment. Complexity theory demonstrates that the emergence of a new order occurs spontaneously in nature, as organisms interact with their environments to develop new structures that facilitate their development and evolution. As Stuart Kauffman (1995, p.viii) points out: 41 Whether we are talking about molecules cooperating to form cells or organisms cooperating to form ecosystems or buyers and sellers cooperating to form markets and economies, we will find ground to believe that Darwinism is not enough, that natural selection cannot be the sole source of the order we see in the world. In crafting the living world, selection has always acted on systems that exhibit spontaneous order. This chapter presents a three-phase journey to understanding living systems principles. First, we review the key findings in complexity theory, which have challenged the traditional mechanistic and reductionistic paradigm and demonstrated its limitations. Borrowing from physics, chemistry, computer science and evolutionary biology, we show that, at a state called “the edge of chaos,” a complex nonlinear dynamic system is able to spontaneously self-organize and create order and structure. While the specific behavior of the system is unpredictable, the pattern of behavior is constrained within some limits. The fact that order and patterns can autonomously emerge from disorder and chaos has challenged traditional, deterministic worldviews and forced the scientific community to consider spontaneous, bottom-up behaviors as critical to the understanding of life and evolution. Second, we discuss living systems principles. Living systems are defined as complex adaptive systems and autopoietic systems (i.e., self-creative systems). Complex adaptive systems are self-organizing systems that are able to learn, i.e., change their internal structures as a way to cope with their environments. The emergent structures display a higher level of order that is fundamental to life. Autopoiesis theory provides a deeper level of understanding of living systems from a cognitive perspective. The theory defines the process of cognition as embedded. From this perspective, the process of cognition is the fundamental process of life. An autopoietic system is a system that uses 42 cognition as a means to change its internal structure and create its own parts, while maintaining its overall organization. Third, we review critical arguments on the application of complexity theory and living systems theory to social systems and argue for the legitimacy of that approach. 2.1. Paradigm of Complexity Theory This section introduces key findings in complexity theory. First, we use the example of two early discoveries of the phenomenon of chaos to present two fundamental characteristics of chaotic systems: (1) sensitivity to initial conditions; and (2) ability to produce order and patterns. Second, we introduce the concept of strange attractors. Strange attractors are forces that maintain a system in a paradoxical phase of both stability and instability: the edge of chaos. Third, through a discussion of the work of Christopher Langton on cellular automata, we illustrate the creation of order and pattern at the edge of chaos. Fourth, we present the theory of dissipative structures; this theory provides an explanation of why open systems and living systems, in particular, are able to self-organize at the edge of chaos. Fifth, we discuss the concept of “life” and consider the fundamental attributes that differentiate living systems from nonliving systems. Finally, we summarize the findings on complexity theory. 2.1.1. Discovery of Chaos During the three centuries that separated Newton’s (1643-1727) definition of the Universal Laws of Motion and the early 20th century, science focused primarily on the study of linear systems; these are predictable systems in which the output is proportional 43 to the input. In accordance with the Newtonian worldview, scientists believed that by gaining enough knowledge about the systems they were studying, they would be able to predict their behaviors. It was assumed that small variations in system input would have a negligible impact on system behavior. This view was first challenged in the late 19th century by French mathematician Henri Poincaré (1854-1912) who was working on the problem of the relative motion of three bodies under mutual gravitational attraction.1 Contrary to Newtonian theory, Poincaré discovered that the system he was studying was defying determinism: the problem’s solution could not be predicted because the system was highly sensitive to very small changes in its initial conditions. This meant that the tiniest perturbation in the system’s initial conditions could be amplified in such a way that the system’s behavior would change unpredictably. The phenomenon of extreme “sensitivity to initial conditions” has now come to be called dynamic instability or, simply, chaos. The term, “chaos,” within the context of this discussion, should not be understood in its common language usage as “a state of utter confusion and disorder” (Rosenhead 1998, p.2). In this context, a “chaotic” system refers to an entity that “display behavior which, though it has certain regularities, defies prediction” (Rosenhead 1998, p.3). An example of a chaotic system is water dripping from a faucet: it is impossible to predict when the next drop will fall. Subject to small accidents and/or uncertain data, chaotic systems defy determinism and challenge our ability to predict the future. The implications of Henri Poincaré’s discovery were far too challenging for the spirit of the late nineteenth century to generate much interest in the scientific community. 44 It was not before the 1970s that chaos theory gained some ground in scientific fields such as astronomy, chemistry, mathematics, evolutionary biology and meteorology. An early discovery of the principle of chaos was, in fact, made by accident in 1963, by meteorologist Edward Lorenz (Gleick 1987, pp.9-33). Like many of his colleagues, Lorenz was searching for a means to provide accurate weather forecasts. Using one of the primitive computers available at the time to run a very simplified model of weather behavior, Lorenz looked into the computer’s output for patterns or rules that could be used in forecasting methods. One day, as he manually entered the initial conditions of a sequence he wanted to re-examine into the computer, the computer generated output that was completely different from the output generated in the previous run. Lorenz quickly realized that the only possible difference between the two sequences was in the number of decimals he had typed into the computer: while the computer was storing numbers with six decimals, Lorenz entered numbers with three decimals, assuming this small round-off would not make any difference. His assumption could not be further from the actual reality. In what is now commonly known as “The Butterfly Effect,” Lorenz confirmed Poincaré’s finding that tiny variations in a dynamic system can produce very large effects, which are basically unpredictable. But unpredictability in a nonlinear dynamic system does not necessarily mean randomness. A nonlinear dynamic system’s behavior is unpredictable because, when plotted on a multi-dimensional map, the specific position of any of its points cannot be forecasted; only its general geometry and behavior can be predicted. In contrast, a system that exhibits random behavior does not create predicable patterns. Lorenz’s weather model did, indeed, exhibit a “fine geometrical structure” – some kind of order 45 amidst chaos. Turning his attention to simpler nonlinear systems (e.g., systems that are subject to friction such as, for instance, a waterwheel or convection in a gas or fluid), Lorenz showed that chaotic systems actually produce ordered structures and patterns. Chaotic systems are paradoxical in that they generate both order and disorder, simultaneously. How order, structure, and patterns are created in chaotic systems is a fundamental question that complexity and chaos theorists are trying to answer. To understand this phenomenon, we need to explore the behavior of strange attractors. 2.1.2. Strange Attractors Attractors are well-known behaviors of nonlinear dynamic systems. Stacey (1996, p.54) defines an attractor as “[a] pattern of behavior into which a system ultimately settles in the absence of outside disturbances.” From a mathematical stand point, there are three kinds of attractors: stable, unstable and strange. Both stable and unstable attractors are well-known behaviors of deterministic nonlinear feedback systems; but it is the strange attractor that is of concern to complexity and chaos theorists.2 Definitions and some examples of the three attractor types are provided below. First, however, we need to define some concepts relating to feedback mechanisms, which is critical to the understanding of the behaviors of attractors. Feedback mechanisms are the focus of cybernetics. Feedbacks are control processes used by a system for self-regulation. Stacey (1996, p.287) defines feedback as “[a] process in which information about the outcome of an action is fed back into the decision-making, or regulation, process to affect the next action.” Feedback is negative 46 when the information about a gap between expectation and outcome is fed back to dampen deviations from the expectation. In contrast, feedback is positive when the information about a gap between expectation and outcome is fed back to amplify deviations from the expectation. Feedback mechanisms occur in nonlinear deterministic systems that are affected by disturbances. Now consider attractors. A stable attractor uses negative feedback to dampen down small, gradual disturbances; it pushes a system back to equilibrium. A stable attractor generates a predictable pattern. Under the effects of a stable attractor, a system may be drawn to a single point (e.g., a pendulum swings until it comes to rest due to friction) or to some periodic cycle (e.g., a thermostat brings the temperature of a room back to the desired level). An unstable attractor uses positive feedback to amplify disturbances disseminated into a system at a very high rate: it pushes a system toward complete instability and randomness (e.g., the snow-ball effect created by over-fishing has the consequence of depleting a fish population to extinction; other examples are chain reaction; inflation; and so on). A strange attractor is paradoxical because it is simultaneously stable and unstable. A strange attractor forces a system’s behavior to remain within some limits (i.e., the system is constrained or bounded), but the system defies determinism because it is impossible to predict its specific position over time. For instance, Lorenz’s Butterfly Attractor – a three-dimensional graphical representation which depicts the chaotic behavior of the weather – exhibits a trajectory that always follows the same pattern, although no one can tell what the exact position of any of its points will be at a given 47 time. The general pattern of the system can be predicted, but not its specific trajectory. Within the limits of a strange attractor, a system moves freely based on tiny changes in the environment. In the realm of mathematics, it is the system’s freedom within the boundaries of a strange attractor, which generates the creativity exemplified in the famous Mandelbrot’s fractal geometries.3 A fractal image consists of the repetition of the same pattern over time at ever-smaller levels of scale. Fractal geometries can be found everywhere in nature: in the patterns of clouds, leaves, vegetables such as broccoli and cauliflower, canyons, and so on. The place where stability and instability occur simultaneously was dubbed the “edge of chaos” by Christopher Langton (1989, 1992), founder of “artificial life,” while he was working on simulations with cellular automata in the 1980s to model the basic biological mechanisms of evolution and life itself. The next section describes the characteristics of the behavior of a system at the edge of chaos. 2.1.3. The Edge of Chaos In the field of artificial intelligence, computers are used to model thought processes. Similarly, computer-based cellular automata are used in the field of artificial life to model very simplified laws of the universe. Cellular automata were first created by John von Neumann in the late 1940s to study self-reproduction.4 A cellular automaton consists of a grid of cells (like a chess board), where each cell can take different values and is influenced by its neighboring cells. For instance, a one-dimensional cellular automaton consists of a set of cells organized in a single row and a set of rules. Each cell can be in one of several states. In a two-state automaton, 48 each cell can be either one of two colors (e.g., either black or white. One may also say “dead” or “alive.”); in a three-state automaton, each cell can be any one of three colors (e.g., blue, red, or yellow); and so on. Cells change from one state to another over time by following rules. The rules are based on the state of each cell and its neighbors at a given time. For instance, in a onedimensional-two-state automaton, a cell can be either black or white. Each cell (including itself) has three neighbors. According to the rules for that particular cellular automaton, if all three of these cells are white, then the new state of the cell will be white; if all three of the cells are black, then the new state of the cell will also be white; in any other case, the new state of the cell will be black. Through his studies, Langton confirmed von Neumann’s theoretical findings that self-reproduction is possible in cellular automata. In addition, Langton showed that selfreproduction specifically occurs at the edge of chaos, i.e., a transition phase where patterns emerge spontaneously. Prior to this phase, cellular automata create a predictable picture (i.e., all the cells on the computer screen are either “dead” or “alive”) as if they are under the effect of a stable single point attractor; past this phase, cellular automata create completely random patterns on the screen as if they are under the effect of an unstable attractor; but right in-between is a phase where geometry and patterns are created, where order and disorder coexist, where creativity takes place and life emerges as if the cells are under the effect of a strange attractor.5 In more general terms, it has been demonstrated that, before reaching the edge of chaos, a deterministic nonlinear feedback system may alternate between predictable and completely random behaviors. When in a state of equilibrium, the system does not have 49 any opportunity for creativity (i.e., for developing new internal patterns) because negative feedback forces the system to come back to a predictable state. When in a state of disequilibrium influenced by positive feedback, the system exemplifies completely random behavior. At the edge of chaos, however, when information flows into the system and the system components are able to interact with one another, the interplay between positive and negative feedback creates new patterns. The edge of chaos is a state where the components of a system self-organize spontaneously, without any blueprint, to create order and structure. This paradox – that order emerges out of chaos under specific conditions – is the focus of the theory of dissipative structures, which we turn to next. 2.1.4. Theory of Dissipative Structures Ilya Prigogine (1917-2003) won the 1977 Nobel Prize in chemistry for his work on dissipative structures, which demonstrates that order can arise from disorder. Prigogine showed that, left on its own, a mixture of specific substances in a dish results in a chemical reaction that creates concentric patterns spiraling for several hours at the surface of the dish. The chemical solution loses energy to the environment and, at the same time, creates internal structures. This phenomenon seems to contradict a fundamental law in physics: the second law of thermodynamics. The second law of thermodynamics states that a system isolated from the environment can only lose its free energy (i.e., energy necessary to do useful work) and, thus, can only decrease its internal organization and ability to do further work. Such a system is drawn to complete randomness or “heat death.” For instance, the energy 50 released by a burning piece of wood can be used to boil water (i.e., useful work); but once burned to ashes, that piece of wood cannot be re-used to do more useful work. In addition, the process is irreversible: once burnt, the wood cannot be reconstituted from its ashes, smoke and heat (Russell 2000). The entropy in such a system, i.e., the amount of energy no longer available for useful work, is increasing. All processes within the system come to a stop at a state of equilibrium. In Prigogine’s chemical substance, the dissipation of the system’s energy did not push the system to a random state; on the contrary, the system increased its level of organization. In fact, the ordered patterns represent a decrease in entropy (i.e., increase in order) within the dish – entropy that is dissipated to the environment. The explanation of this phenomenon resides in the fact that the second law of thermodynamics only applies to closed systems, i.e., systems that are isolated from their environment. In Prigogine’s experiment, the system is open, i.e., energy and matter flow between the system and its environment. In addition, order and patterns emerge when the system is far from thermodynamic equilibrium, i.e., at the edge of chaos. The edge of chaos is created by positive feedback that amplifies small fluctuations in the environment, thus preventing the system from counterbalancing and returning to equilibrium. Far from equilibrium and under great internal disturbance, the system’s initial form disintegrates, but without causing its death. Instead, the system is able to reconfigure its internal structure at a higher level of complexity, which is better adapted to the environment. For the system to be able to create new order, its components must be self-reinforcing (or self-reproducing), i.e., some components catalyze the production of new elements of the same kind. This means that the molecules in the chemical solution are in fact 51 communicating (communication between the components of a system is a prerequisite for self-organization to occur). While communication is an accepted property of living systems, it is a phenomenon quite unexpected in a chemical solution. While Prigogine’s dissipative structures lacked some fundamental characteristics of traditionally-recognized “living systems,” the evidence of autonomous self-organization in such chemical solutions forced the scientific community to review the definition of “life” and extend it to what was previously considered non-living. Indeed, as we argue in the next section, the theory of dissipative structures has profound implications for our understanding of evolution and the origin of life. 2.1.5. What is Life? What differentiates living systems from nonliving systems is their high level of internal organization in diverse interrelated structures. A molecule is not a randomly aggregated set of atoms. Instead, a molecule is composed of atoms organized in a very specific way. It is the specific organization of the atoms, which defines the molecule’s properties; if the organization of the atoms changes, so do the characteristics and properties of the molecule. Similarly, the human body is a highly structured system of cells and organs, organized in a very specific way that differentiates it from a mere random aggregation of cells. The trend toward increasing complexity and organization in living systems seems to be at odds with the second law of thermodynamics – a law that applies to all physical systems. Furthermore, this law implies that the universe as a whole is moving steadily toward increasingly disordered and random states – the entropy of the whole universe is increasing. Consequently, we are left with critical questions: if the universe is evolving 52 toward increasing chaos, how can order and one of its most important representations – life – be arising from disorder? Why do living systems display increasing order? The answer resides in the fact that living systems are open systems that continuously exchange energy and matter with their environments to maintain their internal order. Living systems behave like dissipative structures, i.e., they use the flow of energy and matter from the environment to build and maintain a high degree of order in their internal structures – an order that is critical to life. In doing so, they dissipate entropy in the environment; consequently, the environment is moving toward more randomness and chaos. Self-organization and the emergence of new structures can be found everywhere in nature, ranging from the level of a single-cell organism to bacteria and even to the level of social systems such as ant colonies and bee swarms. In contrast to living systems, mechanical systems do not self-organize: without proper maintenance from their operators, they fall into decay. The theory of dissipative structures provides new insights for understanding the evolution of the universe as a whole. It has influenced evolutionary biologists and other scientists to challenge the traditional Western scientific view that the essential building blocks of life have come about by accident from chemical reactions in the “primeval soup.” Indeed, for neo-Darwinian biologists, life has “put itself together bit by bit” (Harman & Sahtouris 1998). The “bits” are assumed to be nonliving matter and Earth is a nonliving geological ball In contrast to this view, the physicist, James Lovelock (1979), developed the Gaia hypothesis. It proposes that Earth is a self-organizing living system interacting with its environment. From this viewpoint, Earth is a complex adaptive system of interconnected 53 components (e.g., biosphere, atmosphere, oceans, soil) that interact with one another, using feedback mechanisms (i.e., Earth is a cybernetic system) in order to maintain itself in a homeostatic state. As with all living things including humans, Earth uses feedback mechanisms as a means to maintain itself in relatively constant internal conditions. Earth captures the energy of the sun for photosynthetic processes and rejects it into space as long-wave radiation, while matter is recycled within the Earth. To envision Earth as a living system, an analogy may be useful. Consider the California redwood trees, Sequoia gigantea. These trees that may be as old as 3,000 years may stand as high as 300 feet and weigh as much as 2,000 tons. In some way, they are similar to Gaia because 97 percent of their tissues are dead (e.g., the wood of the trunk and the bark of the tree is dead). The living matter consists only of a small rim of cells along the periphery of the trunk: The trunk of the tree is similar to the Earth’s lithosphere with a thin layer of living organisms spread across its surface. The bark, like the atmosphere, protects the living tissues, and allows for the exchange of biologically important gases, such as carbon dioxide and oxygen.6 Few, if any, people would maintain that a redwood tree is not a living entity. Similarly, many of the elements making up Earth are dead; yet even nonliving things contribute to the Earth’s living processes, and we may, therefore, consider Earth to be alive. The Gaia hypothesis has been very much criticized within the scientific community because it has been perceived as being teleological. That is, scientists believed that the Gaia hypothesis required that Earth’s natural processes be guided by a purpose. (Lovelock has always maintained that it is not the case.) In addition, from a reductionistic scientific worldview, the idea that life emerged out of nonlife implies that 54 consciousness developed as a by-product of brain activity. However, the definition of life as a property of all self-organizing systems forces us to take a dramatically different stand and hypothesize that life, and ultimately consciousness, preceded matter. Russell (2000, p.36) pointed out: [W]e could take the view that since consciousness evolves out of life, consciousness is already inherent within life in some potential latent form. Likewise, since life evolves from apparently inanimate matter, life is already inherent within matter in a latent form. Perhaps the potential for each new order is always present, awaiting the particular conditions that would allow it to become manifest. What are these conditions? At least in part, the answer would seem to be a progressive increase in complexity. From a post-Darwinian worldview, the Gaia hypothesis does not stand. To understand the Gaia hypothesis requires a new theory of living systems and evolution, i.e., in the words of Kuhn (1962), it requires a paradigm shift. Such a paradigm shift has occurred, thanks to the work of two Chilean biologists, Humberto R. Maturana and Francisco J. Verela, who, in the late 1960s, created a new theory of living systems called autopoiesis. Autopoiesis theory states that nature and life are conscious and intelligent and that this is the case without the need for an overall design or purpose. Autopoiesis theory is the only theory that provides a definition of life. In contrast, other theories have merely provided descriptions of the properties of living systems. Autopoiesis theory is presented in detail in the next section, where we focus on defining principles of living systems. 2.1.6. Summary Let us now summarize the findings from complexity theory. Complexity theory provided the scientific community with the mathematical tools and methodologies to model 55 nonlinear complex systems. Consequently, it opened scientists’ eyes to new phenomena that could not have been seen by using mechanistic, reductionistic methods. Complexity theory has brought to light the limits of the post-Newtonian worldview that all system behaviors are linear and predictable. It has shown that very simple equations and rules, iterated many times, can create spontaneous, awesome complexity. The new theory has pushed scientists to focus their attention on patterns of organization, as opposed to a quantitative analysis of systems and, consequently, to interpret reality in a completely different way. Complexity theory has also provided new directions toward gaining a deeper understanding of the principles of life and evolution. It has shown that equilibrium can only drive a living system to its death. Life, creativity and the emergence of new structures and patterns of organization seem to occur at the edge of chaos – a paradoxical place of simultaneous stability and instability. At the edge of chaos, dissipative structures use positive and negative feedback to freely self-organize within the limits of strange attractors. Living systems are open systems that behave like dissipative structures: they are able to decrease their entropy by exchanging information, energy and matter with the environment. In doing so, they create a higher level of internal order. 2.2. Living Systems Principles With complexity theory as our new theoretical framework, we now consider the principles of living systems. Living systems are first defined as complex adaptive systems, i.e., systems whose agents interact with one another to coordinate the behavior of the whole system. Second, we review some examples of self-organization in living 56 systems. Third, we discuss the concept of fitness landscapes, which provides an explanation of why self-organization occurs only at the edge of chaos. Finally, we review the autopoiesis theory that defines living systems as cognitive systems, capable of creating their own parts. 2.2.1. Complex Adaptive Systems Stacey (1996, p.10) defines complex adaptive systems as consisting of “a number of components, or agents, that interact with each other according to sets of rules that require them to examine and respond to each other’s behavior in order to improve their behavior and thus the behavior of the system they comprise.” Based on the above definition, complex adaptive systems do not need a centralized control mechanism to coordinate their behavior; instead, independent agents interact with one another based on local rules that dictate what behavior to follow in a particular context. For instance, an ant colony and a bee hive are complex adaptive systems in which the agents (the individual ants and bees) coordinate their actions by following simple rules inherent to their individual species. Similarly, the human brain is a complex network of neurons (the agents in the system) that fire when stimulated by some external or internal stimulus. The resulting interactions between the neurons create a pattern of electricity, which becomes associated with the stimuli. Repeated many times, the operations in the brain define diverse rules (or symbol systems) that are associated with different experiences. Humans are complex adaptive systems as well, acting as independent agents within a larger system, namely, society. As independent agents, humans follow rules that are coded in symbols such as words and images. 57 A crucial characteristic of complex adaptive systems is that they have emergent properties as a whole, which are not exhibited by the parts, i.e., the whole is greater than the sum of the parts. The properties arise from the interactions and the relationships between the parts (Capra 1996). In living systems, the emergence of complexity at the level of the system can be commonly witnessed in the decentralized coordination of foraging in an ant colony or in the flocking of birds. The components of a complex adaptive system may be very different, each having a distinct function in the system. For instance, a bacterium – one of the simplest organisms on Earth – may contain a handful of DNA molecules, hundreds of thousands of RNA molecules, a million molecules of protein and some hundred million smaller organic molecules (Russell 2000). Moreover, any complex adaptive system presents a highly organized internal structure: in contrast to crystals that can be uniquely described by the combination of their chemistry and atomic arrangement, a simple bacterium would require a lengthy description to depict the complexity of its internal structure. The components in a complex adaptive system are nested: each individual component is itself a complex system of interconnected parts. An individual agent in a system is sometimes called a holon – a term first introduced in 1967 by Arthur Koestler (1905-1983) in his book The Ghost in the Machine. Taken together, holons form a holarchy.7 An example of a holarchy may be: cells-tissues-organs-body-familycommunity-ecosystem-bioregion-planet-star system-galaxy-and so on (Harman & Sahtouris 1998). Hence, when looking at a living system from a holistic perspective (as opposed to considering the components of the system independently of the whole), biologists consider the components of the system, the system itself and its ecosystem as 58 living systems. At each level in the holarchy, holons exemplify the same properties; thus, one would say that complex adaptive systems have a fractal or holarchic nature. At each level of the holarchy, “[t]he parts interact continuously to re-create the wholes and the wholes in turn affect the interaction of the parts…” (Stacey 1996, p.37). The components of a complex adaptive system interact with one another to exchange resources and information necessary to maintain the system’s internal balance (i.e., homeostasis). To self-regulate, a complex adaptive system uses control processes such as feedback loops that move the system toward its internal goals (Capra 1996). In other words, a complex adaptive system adjusts its behavior to better achieve its purpose. Complex systems are adaptive because they have some purpose, i.e., they aim to perform some tasks necessary to their existence (Stacey 1996). It is interesting to note that the concept of purpose in living systems is foreign to conventional biology because the reductionistic method did not provide biologists with the tools to understand mutual interactions between components of living systems; but, as we now know from cybernetics theory, all self-organizing systems are purposeful systems because they operate based on feedback mechanisms that are always directed toward a goal (Capra 1996; Stacey 1996). Therefore, while purpose has been mainly considered an attribute of human behavior, it can be now extended to all living systems. As Harman and Sahtouris (1998, p.29) note: [T]he insistent tendency among living things for bodily development to reach and maintain as a ‘goal,’ an organized living system of a definite kind, and the equally persistent directiveness or goal-seeking that is an essential feature of human behavior and mental activity, are fundamentally the same thing. Feedback mechanisms require that the components of adaptive systems learn and use their learning as the basis for action. “Learning,” in this context, should not be 59 understood in the sense of acquiring new information. Learning consists of a system’s ability to scan its environment to detect variations and apply a set of rules in order to adjust its behavior appropriately. There are two kinds of learning in complex adaptive systems: (1) single-loop learning occurs when a system applies its program (i.e., the set of rules that defines the system’s behavior) without change; in contrast, (2) double-loop learning occurs when a system changes its program in order to adapt to the environment. In each case, a system adjusts its behavior to the stimuli in its environment so that its behavior is better adapted to the situation; but double-loop learning is much more involved, since it requires the system to make changes in its internal program, i.e., it requires the system to be creative. Creativity in complex adaptive systems occurs at the edge of chaos – a “space of novelty” – that Stacey (1996) defines with five characteristics: 1) A phase transition. This phase is characterized by tension such as simultaneous order and disorder; stability and instability; along with constraint and freedom. 2) A state of paradox. Paradox is created from the bounded instability of a system. Examples of paradox may be the presence of both conformity and diversity; agents that are richly, but not too richly, connected; behavior that is both predictable and unpredictable; competition and cooperation; and so on.8 3) Actualization of archetypes. Archetypes are rules that drive the behavior of agents in a system (examples of archetypes may be flocking behavior or traillaying behavior). Stacey (1996) notes that, “The actualization of the archetype is sensitively dependent upon the precise interactive experience of the system” – 60 that is, the system behavior is contextual in the sense that it always displays the same general pattern, but is never quite exactly the same. 4) Creative destruction. At the phase transition, creativity (i.e., double-loop learning) requires destruction of some of the system’s existing rules and the creation of new ones. 5) A critical point for control parameters. Three control parameters drive a complex adaptive system’s behavior and dictate whether the system will move to and remain in the space of novelty or if it will, conversely, fall into complete chaos and disintegration: (1) the rate of information flow through the system; (2) the richness of connectivity between agents in the system; and (3) the level of diversity within and between the schemas (i.e., rules) of the agents. The outcome of the process of learning in a system is self-organization and its corollary: emergence. All living systems self-organize and create emergent structures that are better adapted to change in their environments. Both properties are fundamental to life. We look at the phenomenon of self-organization in more detail in the next section. 2.2.2. Self-Organization in Living Systems Capra (1996, p.85) defines self-organization as “[t]he spontaneous emergence of new structures and new forms of behavior in open systems far from equilibrium, characterized by internal feedback loops and described mathematically by nonlinear equations.” This definition includes the three main characteristics of living systems: (1) the ability of open 61 systems to (2) create new structures and modify their behavior, thanks to (3) the high degree of interconnectedness between the system’s components. The concepts of self-organization and emergence were foreign to the neoDarwinist worldview. From the neo-Darwinist perspective, evolution occurred through random mutations of organisms adapting to the environment. The concept of the “survival of the fittest” means that those organisms that adapt better and faster than their competitors survive or, in the words of Elisabet Sahtouris, that the “dying out of failures adequately explains the successes” (Harman & Sahtouris 1998, p.54); however, there are puzzles in nature that cannot be explained by this worldview. In particular, the idea that evolution occurred through gradual variations or the accumulation of small changes is in conflict with evidence of the sudden appearance of new species, as the two following examples demonstrate. The biologist Lynn Margulis has shown in her study of bacteria – the oldest organisms on Earth – that interconnectedness, networking and merging between organisms are at the basis of evolutionary change on Earth – in addition to, and beyond, deadly competition (Margulis & Sagan 1997). Bacteria are able to exchange genes with bacteria of a different type so that the receiving bacterium can use the foreign genes to perform functions that it could not have previously performed with only its own genes. Bacteria are also capable of merging with one another. The results of one of the early mergers were protoctists, our microbial ancestors, which are at the origin of our digestion, movement, as well as tactile and visual systems. Acknowledging that humanity has evolved from these complex microscopic beings “challenges our ideas of individuality, uniqueness, and independence” (Cole 1993, p.3). 62 Another example of self-organization in relatively simple cells can be found in slime mold (Johnson 2001, pp.11-17). Within normal weather conditions (for slime mold, this means a cool, damp environment), slime mold operates as thousands of distinct, single-cell units moving separately. As external conditions change to become warmer and dryer, the cells of the slime mold coalesce into a single, larger, organism. For many years, biologists thought that – consistent with the traditional model of centralized command and control – such an incredible example of coordinated behavior was the result of “pacemaker” cells that used a substance called cyclic AMP to order the other cells to begin aggregating. Each single cell was believed to relay the signal to its neighbors as if playing a giant game of “telephone.” There was, however, a problem with this theory: all the cells in the slime mold were identical, and nothing indicated the presence of a cell hierarchy. Twenty years elapsed before Evelyn Fox Keller, a Harvard Ph.D. in physics, and Lee Segel, an applied mathematician, demonstrated that slime mold could behave without a centralized coordination system. They did so by adapting a mathematical model developed by Alan Turing (1912-1954) – the famous English mathematician who broke the code during World War II – to study the mechanism of “morphogenesis.” Turing’s mathematical model demonstrated how a complex organism could assemble itself, without relying on a master plan. In a paper published in 1969, Keller and Segel demonstrated that slime mold cells can behave independently of a leader by releasing cyclic AMP and then following trails of the pheromones they encounter. If the slime cells pump enough cyclic AMP, clusters of cells will start to form. Cells will begin following trails created by other cells, creating 63 a positive feedback loop that encourages more cells to join the cluster. It took another ten years for a series of experiments to demonstrate Keller’s and Segel’s hypothesis to be correct. Self-organization and emergence are common phenomena around us: They explain how dental plaque grows so strong as to require a dental hygienist’s tool to remove it; how organisms develop resistance to antibiotics, herbicides and pesticides at great speed; how bioengineered crops can merge with non-bioengineered ones so that they can resist pesticides; and so on. We now know that self-organization occurs at the edge of chaos: as we previously discussed, tension and paradox are critical for a system’s learning and adaptive capacities. However, the question that has yet to be addressed is: Why do living systems need such an environment in which to self-organize? The concept of fitness landscapes, discussed in the next section, provides an explanation of why this is the case.9 2.2.3. Fitness Landscapes The term fitness has been used by biologists to describe the effectiveness of an organism’s long-term survival strategy. It has been demonstrated that the fitness of organism A depends not only on A’s strategy to improve its own fitness but, also, on the fitness of the organisms with which A interacts in its ecosystem. This can be illustrated by the following example: a given antibiotic only remains effective as a human survival strategy until the bacterium it is supposed to fight mutates and develops its own survival strategy. By improving its own fitness, the bacterium impacts the fitness of the antibiotic and, ultimately, of a human being. Fortunately, beyond an “either-or” survival strategy, it 64 is also possible for two or more systems to follow a strategy that improves their fitness as a group – for instance, by cooperating with one another (Stacey 1996). A fitness landscape consists of a set of peaks and valleys representing the different successful and unsuccessful survival strategies of a system respectively (Kauffman 1993). For example, an organism, A, attempting to maximize its chance of survival, will try to find the highest peak, using diverse strategies. If A does so by incrementally improving on a given strategy, assuming all other systems in the ecosystem will not change their own strategies, A is likely to get stuck on the first peak it comes to, because all subsequent incremental steps will only go downhill. Because it is unlikely that A will find the highest peak in its first attempt to improve its fitness, getting stuck on the first peak means that A will never discover a higher peak unless it accepts a decline of fitness before the next improvement. In contrast, if A chooses a disorderly walk in which it sometimes falls downhill, its chances of finding a higher peak are greater.10 As Stacey (1996, pp.83-84) points out, “This is the same as saying the system should not seek adaptation or equilibrium but should follow strategies that hold it away from them, at the edge of chaos.” As discussed, the edge of chaos requires both instability and stability. Instability is created by competition that requires a system to learn and change its strategy. Stability is created by the system’s internal rules that may create “lock-in” of a particular form of behavior (Kauffman 1991, 1993). Kauffman (1991, 1993) has shown that the state at the edge of chaos occurs when landscapes are neither too smooth nor too rugged. When landscapes are very smooth (i.e., with no peak) or extremely rugged (i.e., with many very low peaks) it is easy for all the agents in a system to reach their peak and thus impact the 65 fitness of the other agents. Therefore, the agents’ landscapes will be constantly and rapidly changing. Conversely, if the landscapes of the agents are neither too smooth nor too rugged (i.e., the landscape consists of many high peaks), the system will not change so violently and self-organization will be possible. In the next chapter, we will review the concept of fitness landscape and apply it to social systems and organizations. At this point, we will consider the autopoiesis theory that has deep implications for our understanding of self-organization, learning and the definition of life. 2.2.4. Autopoiesis Theory The fundamental principles at the basis of self-organization and adaptation are the focus of autopoiesis theory. The term “autopoiesis” comes from the Greek words auto (self-) and poiesis (creation, production): autopoiesis means self-creation. The theory states that a living system continually creates its own parts and its own structure, while maintaining its own overall identity or pattern of organization. Autopoiesis theory has been critical in advancing our knowledge of living systems. The autopoiesis theory was developed by two Chilean biologists – Humberto R. Maturana and his former student Francisco J. Varela – in the late 1960s as they were studying cognition and perception. (Interestingly, the theory equates the process of living to a process of cognition.) For Maturana and Varela (1980, p.7), cognition and perception are of interest to biologists because “[c]ognition is a biological phenomenon and can only be understood as such.” All living systems able to perceive change in the 66 environment are cognitive systems: “Living systems are cognitive systems, and living as a process is a process of cognition” (Maturana & Varela 1980, p.13) Maturana and Varela’s (1980) definition of cognition sharply contrasts with the conventional notion of the process of cognition defined as a process of representation of an external, independent world. In the conventional view, cognitive beings internally process information or signals that exist “out there” in the external world. In contrast, for Maturana and Varela (1980), cognition is contingent on embodiment because it is the internal structure of a system, which specifies the object of cognition (as opposed to representing it), while the cognitive domain in which the system may interact is dictated by the system’s organization. In other words, a living system’s cognitive process is equivalent to the system’s ability to modify its internal structure in response to environmental change, while maintaining its overall organization. Therefore, from this perspective, cognition is the process of living and autopoiesis theory is a theory of life. Let’s now understand how this can be. First, we need to clarify the difference between the organization of a system and its structure, because the difference explains how an autopoietic system can change and evolve, while maintaining its “self.” The organization of a system (living or nonliving) refers to the configuration of the relationships between the components of the system and, as such, determines the system’s identity – that is, the system’s membership in a particular class – for instance, a dog, a car, a chair and so on. Organization determines the system’s properties as a whole (it does not specify the properties of the system’s components). 67 The structure of a system refers to the actual components of the system and the way these components relate to one another in order to constitute a composite unity. The structure does not define the system’s identity; it is only the “physical embodiment of its organization” (Capra 1996, p. 98). As Maturana and Varela (1980, p.77) explain: The organization of a machine (or system) does not specify the properties of the components which realize the machine as a concrete system, it only specifies the relations which these must generate to constitute the machine or system as a unity. Therefore, the organization of a machine is independent of the properties of its components which can be any, and a given machine can be realized in many different manners by many different kinds of components. In other words, although a given machine can be realized by many different structures, for it to constitute a concrete entity in a given space its actual components must be defined in that space, and have the properties which allow them to generate the relations which define it. The following example may clarify the fundamental difference between organization and structure. A car is a system that comprises a certain number of components such as a body, an engine, a steering mechanism, wheels, and so on. In order for these components to form the machine known as a “car,” there must be some specific functional relationships between them. The complete configuration of these functional relationships constitutes a car’s pattern of organization. The structure of the car is the physical embodiment of its pattern of organization, i.e., the particular shape, materials, color and so on, of the components. The same pattern – “car” – may have many different structures that will differentiate, for instance, a sedan from a sports car. In Baldwin and Clark’s (2000) terms, the structure of the artifact “car” consists of the car’s design parameters (e.g., dimensions, materials, color, and so on). Changing the values of the design parameters (e.g., the color may take the values “blue” or “red”), while it affects the car’s structure, does not change the car’s pattern of organization. 68 The key characteristic of a living system is its ability to change its structure, while maintaining its organization. This is equivalent to saying that the system is able to create its own parts, while preserving its identity. If, because of a structural change, there is loss of identity, the system disintegrates as such. As Capra (1996, p.81) points out: The study of pattern is crucial to the understanding of living systems because system properties, as we have seen, arise from a configuration or ordered relationships. Systemic properties are properties of pattern. What is destroyed when a living organism is dissected is its pattern. The components are still there, but the configuration of relationships among them – the pattern – is destroyed, and thus the organism dies. The pattern of organization of all living systems is a network pattern: it is characterized by nonlinear relationships between the system’s parts, i.e., feedback loops that are used by the parts to maintain the system’s organization (Capra 1996). As Maturana (1970) states, “Living systems…[are] organized in a closed causal circular process that allows for evolutionary change in the way the circularity is maintained, but not for the loss of the circularity itself” (cited in Capra 1996, p.96). Structural changes and adaptations in an autopoietic system are triggered, but not determined, by perturbations in the environment. It is the structure of a living system that selectively determines which perturbations in the environment the system will notice or ignore, and which new interactions will be created between its components. One says that the behavior of a living system is structure-determined. “According to Maturana, the behavior of a living organism is determined, but rather than being determined by outside forces, it is determined by the organism’s own structure – a structure formed by a succession of autonomous structural changes” (Capra 2002, p.36). The concept of structure-determined systems allows us to reconcile the notions of freedom and determinism. An autopoietic system is both determined and free: it is determined by its 69 own structure, but it maintains the freedom to define which perturbations from the environment will trigger its internal change and adaptation. Structural coupling is the term used to depict a living system that engages with another system or with the environment. Consequently, structural coupling relates to the processes of coordination (i.e., how two or more systems coordinate their activities) and co-evolution (i.e., how the systems change and adapt over time) (Whitaker 1996). When a system couples (i.e., interacts) with its environment and is influenced to change its structure, the system goes through a process of adaptation to the environment without disintegration. When a system couples with another system, the systems mutually influence each others’ changes, i.e., structural changes in one system influence the other system to also change its internal structure. Each change is bounded to what each system’s structure can afford at a given time; therefore, it is not possible to talk about changes in one of the systems without referring to its interactions with the other system, i.e., the domain of possible changes in the systems is not reducible to descriptions framed only in terms of either one. Over time, the history of the structural changes in a system defines the system’s development path (i.e., evolution). At any given point in the evolutionary path of a living system, the structure of the system is “a record of previous structural changes and thus previous interactions” (Capra 2002, p.36). This means that all living systems have a history that consists of the set of all structural changes. It is the history of structural changes underwent by a system, which differentiates its behavior from that of another system of a similar type. This is why a system’s history is important to the understanding of its development and evolution. 70 The history of the structural changes in a living system is its history of learning (i.e., double-loop learning). Living systems learn through their ability to structurallycouple with their environment or with other systems in order to change the relationships between their components (i.e., change their internal structure) and thus adapt to their environment, while maintaining their organization and identity. Double-loop learning occurs when the components of a system use positive feedback to amplify a disturbance in the environment, which triggers structural changes and thus, adaptation. In contrast, single-loop learning occurs when the components of a system use negative feedback to dampen disturbances in the environment and apply a standard set of rules to follow to remove or minimize the effect of the disturbances. Single-loop learning does not trigger structural change in a system but, instead, brings the system back toward equilibrium. Since an autopoietic system is able to determine on its own what triggers its adaptation, the system is said to be operationally closed. This does not mean, however, that the system is isolated from the environment and not influenced by it: the system is energetically open (as Prigogine’s dissipative structures prove). Closure in autopoietic systems means that “their change of state in response to changes in their medium are realized and propagated solely within the network of processes constituting them (as they are defined)” (Whitaker1996, p.5). An operationally closed system is able to re-define its boundary by choosing the components with whom it will interact. There is a dialectic between the environment and living systems in that they influence and shape one another. As Sahtouris (1996) highlights, “…[E]nvironments are not ready-made places that force inhabitants to adapt to them, but ecosystems created of, by, and for living beings.” (cited in Harman & 71 Sahtouris 1998, p.61). Consequently, an observer may define what is inside or outside a system; for the system, however, “inside” and “outside” have no meaning, since the system is operationally closed and able to determine, on its own, the components of its internal structure, at any given time. An autopoietic system is self-referring, i.e., its structural changes are always consistent with itself. Self-reference in a living system implies that the system must go beyond itself in order to refer to or represent itself, i.e., it needs to transcend itself (Harman & Sahtouris 1998). This means that a living system is capable of self-reflective consciousness, i.e., it is able not only to be conscious, but to be conscious of being conscious (Russell 2000). (The word “consciousness” here is used to mean “the field within which all experience takes place” (Russell 2000). In this sense, consciousness is a prerequisite for all experience.) As Russell (2000, p.32) points out, “An important characteristic attributed to conscious beings is the ability to form internal models of the world they experience; the greater the consciousness, the more complex the models.” Humans are able to develop complex internal models of reality that include the self – the “modeler” – in the model. Mental models in humans are at the basis of our ability to formulate values, beliefs, goals and strategies (Capra 2002). However, it is important to point out that all autopoietic systems are intelligent and conscious, no matter how developed. Humans are not a superior species that is separate from other systems; rather, they are interconnected with the entire universe (Harman & Sahtouris 1998). Once again, this view is in sharp contrast with the reductionist approach that argues that consciousness can be explained in terms of neural events in the brain, and life, in terms of 72 organic chemistry. The alternative view is to consider consciousness as being inherent in nature (Russell 2000). The discussion of consciousness is relevant here because self-reflection is necessary for double-loop learning: this type of learning requires the system to question and modify its internal set of rules in order to develop and evolve. To increase its complexity, therefore, a living system must know its internal capabilities and choose which ones to use in different contexts. Self-reflection in a living system means that the parts understand the whole. This has an interesting consequence when we consider a living system as being embedded within a holarchy. Indeed, a living system can selfrefer to any level of the holarchy to which it belongs and thus, extends its boundaries to whatever level the system identifies within the holarchy. Consequently, autopoietic theory is consistent with and supports the Gaia hypothesis: if we consider Earth or, in fact, the universe as holarchies made of holons all interacting with one another, then both Earth and the universe are alive. To complete this presentation of autopoiesis theory, we now return to the process of cognition. As discussed, an autopoietic system is a cognitive system. In autopoiesis theory, the mind is not a thing, but a process that does not require a nervous system or brain. The definition of cognition in autopoiesis theory goes beyond the thinking process to include perception, emotion and action, as well as language and conceptual thinking in humans (Capra 1996). From this perspective, “cognition, then, is not a representation of an independently existing world, but rather a continual bringing forth of a world through the process of living” (Capra 2002, p.36). This view is to be contrasted with the Cartesian 73 dichotomy between mind/matter, that is, the complete separation between subjective phenomena and objective reality of the world. The new understanding of cognition has some implications for both research methodology and information theory. Conventional scientific methodology consists of providing an objective description of reality; but, from the standpoint of autopoietic theory, it is not possible to separate the observer from the phenomena it observes, since the observer is a structuredetermined living system. For Maturana and Varela (1980, p.136), “cognitive reality” is circumscribed to the system’s “cognitive domain,” that is, “all the interactions in which an autopoietic system can enter without loss of identity.” If the system observes a phenomenon, the system’s cognitive domain is limited to “the domain of all descriptions which it can possibly make” (Maturana & Varela 1980, p.136). Findings in quantum physics have shown that separation between an observer and an observed phenomenon does not exist when dealing with atomic entities (Heisenberg 1927). This phenomenon is now extended beyond quantum systems: it seems to be a characteristic of all living systems. From this perspective, all cognitive activity is embodied and context-specific; therefore, all scientific endeavors are inherently subjective and context-specific. This does not mean that we should give up scientific rigor. As Capra (2002, p.43) points out: When we speak of an ‘objective description’ in science, we mean first and foremost a body of knowledge that is shaped, constrained, and regulated by collective scientific enterprise, rather than merely a collection of individual accounts. The autopoietic definition of cognitive activity provides a new way for understanding communication between living systems (and humans in particular). As we have seen, new information is necessary for a living system to evolve and create new order: in isolation from the environment, a system’s entropy increases and the system 74 falls into decay. Information is the “creative energy” of living systems (Wheatley 1994). While information has often been considered by information theorists to be a “thing” to be processed and managed, Wheatley (1994, p.102) argues that information is “both the underlying structure and the dynamic process that ensures life.” As she explains, while the cells in our bodies are always regenerating and changing, the information in the DNA remains constant. Physician Deepak Chopra (1989) points out that a cell is “a memory that has built some matter around itself, forming a specific pattern. Your body is just a place your memory calls home” (cited in Wheatley 1994, pp.103-104). This view that information is more constant than matter is consistent with the idea that consciousness preceded the known material universe, i.e., that consciousness transformed itself into matter, in contrast to the traditional view that considers consciousness and intelligence as emergent phenomena in evolution (Harman & Sahtouris 1998). Understanding information as a dynamic process has important consequences for the way we deal with it in our social lives, especially in our business organizations. In contrast to information theory that considers information and knowledge to be contextand value-free, autopoietic theory holds that information does not exist independently of a system; it is not something “out there” that the system picks up from the environment. Instead, the system selects which disturbances to notice and consequently, creates information and assigns meaning to it through structural coupling.11 Thus, communication is not a process by which organisms exchange information, but instead, it is a process used for coordinating behavior. Autopoiesis theory is, in this sense, similar to action theory (Winograd & Flores 1987) that states that if participants in a communication process are operating from 75 different views of topics, intentions, and outcomes, true communication will not occur. Since every representation of the external world is an interpretation – and that there may be as many interpretation as there are interpreters – focusing on a particular message (i.e., on the thing) is misleading because there is no guarantee that it will be assimilated by the participants involved in the communication. The message is important only in its relationship to the participants. The focus should be on the mutual orientation of the conversants, mutual adjustment and integration of knowledge, especially tacit knowledge. In human systems, tacit knowledge is deeply embedded within people’s past experience and people’s ability to apply tacit knowledge is mainly contextual (Nonaka & Konno 1995, 1998). The meaning we assign to information is contextual, grounded in individuals’ experiences and based on our cultural conventions. As Capra (2002, p.84) points out, [W]e interpret something by putting it into a particular context of concepts, values, beliefs, or circumstances. To understand the meaning of anything we need to relate it to other things in its environment, in its past, or in its future. Nothing is meaningful in itself. From the perspective of an autopoiesis theorist, meaning is not required for two systems to communicate effectively. Maturana and Varela (1980) consider communication as the coordination of behavior between living organisms through mutual structural coupling. This type of communication occurs with or without nervous systems. This definition of communication does not require language, as it is understood in the sense of linguistic communication. All organisms communicate when they interact with one another; for instance, honeybees dance to depict sophisticated patterns, thus indicating the location of specific flowers. Even the molecules in Prigogine’s chemical solution communicate in the sense that each molecule “knows” what the others are doing. 76 For Maturana, “Language arises when a level of abstraction is reached at which there is communication about communication. In other words, there is a coordination of coordination of behavior” (cited in Capra 2002, p.53). While all animals are able to communicate via mutual interactions to coordinate their behaviors, only human beings are able to refer to a previous communication through the use of language, i.e., a system of symbolic communication in which symbols such as words, gestures and other signs serve as coordinating action. These symbols, when they become associated with our mental images of objects, become the basis of further coordination and linguistic communication (Capra 2002). Therefore, as Capra points out, human communication has a dual nature: on the one hand, since it is based on the use of symbolic language, it generates ideas and meaning; on the other hand, as with communication in all living organisms, it involves coordination of behavior that, in humans, means rules of behavior, or social structures. (We come back to the concept of social structures in the next section as we extend the autopoiesis theory to social systems.) 2.2.5. Summary Let’s now summarize this section on the living systems principles. A living system is a complex adaptive system whose internal agents use feedback mechanisms to regulate the system’s behavior and create emergent order. To cope with environmental discontinuities, the agents in a complex adaptive system self-organize at the edge of chaos – a phase transition characterized by paradox. At the edge of chaos, when a lot of information flows between the diverse agents of a system and when the agents are richly connected, the system uses cognitive processes to perceive discontinuities in the 77 environment and uses that information to trigger structural changes that are consistent with the system’s identity. A living system is a self-referring system in that each part “knows,” or is aware of, the organization of the whole. Consequently, a living system’s adaptive learning is always consistent with the history of its past structural changes. If, for some reason, the structural changes are inconsistent with the identity of the system, the system disintegrates. The history of a living system’s structural change is its history of learning. The learning process consists of the destruction of some of the system’s internal rules and the creative definition of new rules that are better adapted to the environment. A living system embedded within a holarchy brings forth the reality of its environment and autonomously self-refers to any level of the holarchy. In doing so, the system is able to redefine its boundaries and its environment. Communication between structurally-coupled living systems aims at coordinating the individual behaviors of the systems and may occur without language. Language adds a higher dimension to the communication process by adding meaning to a particular context. The principles that characterize self-organizing living systems are indicated in Figure 1 below. The different terms at each corner of the triangle define essentially the same concept and have been used by different authors to describe living systems. Matura and Verela (1980) define an autopoietic system as a system in which the pattern of organization is embodied in the structure of the system; the embodiment of the structure occurs through a cognitive process that is a process of life. Wheatley (1992, p.102) considers information as the “dynamic process that ensures life.” It is self-generated 78 through the relationships between system components and participates in defining the system’s identity. Finally, Capra (2002) considers the form of a system as its pattern of organization; matter as the system’s structure; and process as a process of cognition. As he extends the living systems principles to social systems, Capra adds meaning as the fourth element to the model (not shown in the figure below). Matter Relationships Structure Process Information Cognition Living Systems’ SelfOrganization Principles Form Identity Pattern of organization Figure 2.1: Principles of Self-Organization in Living Systems 2.3. Complexity Theory Applied to Social Systems The question of whether complexity theory and living systems theory can be applied to social systems has been much debated (Rosenhead 1998). Indeed, complexity theory has originated from scientific fields such as astronomy, chemistry, mathematics, evolutionary biology, geology and meteorology. As Rosenhead (1998, p.2) points out, “[T]here is no unified field of complexity theory, but rather a number of different fields with intriguing points of resemblance, overlap or complementarity.” Over the last few years, however, much has been written to demonstrate that social systems are living systems, and many theorists in sociology, organizational management and economics have used complexity 79 theory to explain the behavior of social systems, organizations and economies (Arthur 1988; Capra 1996, 2002; Luhmann 1990; Stacey 1996; Wheatley 1992; Winograd & Flores 1987). As Capra (2002, p.82) argues, “Since social systems not only involve living human beings, but also language, consciousness, and culture, they are evidently cognitive systems – it seems rather strange to consider them as not being alive.” As we shall see, however, the application of autopoiesis theory to social systems, while possible, is a little more problematic. In this section, we shall review two main arguments that are critical to understanding organizations as living systems. The first argument is from Ralph Stacey (1996) who has probably been the most influential writer in demonstrating the pattern of self-similarity in the behavior of an individual mind, a group and an organization. Using a very detailed and rigorous method, Stacey (1996) shows that individuals, groups and organizations are complex adaptive systems and that the creative process (i.e., doubleloop learning) in an individual, a group and an organization are essentially the same. The second argument concerns the extension of autopoiesis to the social domain. While the work of Maturana and Varela (1980) restricts the application of autopoiesis to biological systems, it is German sociologist Niklas Luhmann (1990) who has made the most extensive use of the theory by developing a “social autopoiesis” that is centered on the idea that social networks are self-generating networks of communication. Capra (2002) extends this idea by arguing that human systems are networks of communication that create social structures in which meaning is embedded (e.g., value, belief, judgment, and so on). In other words, the production of meaning is to social systems what the production of physical structure is to living systems in general. 80 2.3.1. Self-Similar Nature of the Creative Process in the Mind, Groups and Organizations Building on psychoanalytic literature, Stacey (1996) demonstrates patterns of selfsimilarity in the creative process from the human mind to the group and then to the organization. Stacey (1996) shows that the human mind, groups and organizations are complex adaptive systems that self-organize at the edge of chaos. In each system in question, the process of creativity requires a state of paradox including both sources of instability and sources of stability. Creativity emerges from the dialectic between these two opposing sources. While the specific, long-term evolution of the systems is unpredictable, the processes will exemplify archetypal patterns that are consistent with the systems themselves. Stacey (1996) shows that an individual mind consists of two subsystems: a dominant symbol system and a recessive symbol system. The dominant symbol system consists of rules expressed as routines, habits, procedures, theories in-use, checking and control behaviors, customs and rituals and so on. These rules are the source of stability in the mind (single-loop learning) and drive current engagement with external reality. The recessive symbol system consists of rules that govern play (i.e., fantasy): these are expressed as speculations, images, dreams, metaphors, analogies, fantasies, espoused theories, myths and so on. These rules are available to an individual, but are not currently being used to engage reality. They govern the inner life of an individual and are sources of instability in the mind because they usually conflict with the dominant symbol system. These rules facilitate the process of creativity (i.e., double-loop learning). Creativity and innovation require an individual to change its internal structure 81 – a destructive process; therefore, creativity creates anxiety that needs to be contained, if one does not want defensive mechanisms to kick in, which will impede creativity. The containment of anxiety is achieved by calling on the dominant symbol system: it requires an individual to have a strong ego and a good holding environment (friends, family, colleagues, etc.). Creativity only occurs if it will improve fitness that in turn depends on the responses of the immediate entourage with which the creative agent interacts. Consequently, individual creativity and innovation lie in interactions within a group and are dependent on the cooperative or the competitive behaviors of the interacting agents. At the level of the group, the space of creativity consists of the tension between group members’ cooperative and competitive behaviors. On the one hand, cooperation translates to searching for win-win solutions; discussing and testing multiple alternatives; and holding off all use of power and hierarchical position to avoid obstructing the process. On the other hand, competition is exercised by members who attempt to win the process, retain control and restrict participation in the decision-making process. This can also be expressed by the tension between the legitimate system that attempts to maintain the status quo and the shadow system that drives change. When the legitimate system is the stronger of the two systems, the group remains in the equilibrium zone and is not, therefore, capable of creativity. When the shadow system is the stronger of the two, the group tends toward anarchy. Thus, group creativity only occurs when the group is able to maintain a balance between the forces created by the legitimate and the shadow systems. Moreover, if the group is creative, its level of fitness will depend on the fitness of other groups in the fitness landscape; therefore, a group’s creativity also depends on whether it 82 can engage the creativity of other groups with which it interacts or of the organization to which it belongs. Similarly, creativity at the organizational level occurs when the organization engages with other organizations in its ecosystem through both cooperative and competitive processes. Here again, we may see a tension between the organization’s legitimate system of norms, procedures, protocols and so on, which, in addition to society as a whole, provide stability and the shadow system used by groups that gain enough power to challenge the status quo. Stacey (1996, pp.189-190) concludes his demonstration by stating: From this perspective creativity is not an attribute of an individual but a property of a hierarchy of interlocking systems ranging from an individual mind nested in a group nested in an organization, which is in turn nested in an economy and a society, which are nested in an international community, which is nested in the earth’s ecology. The interaction between these nested systems is always circular, however, with one level affecting and being affected by the levels above and below. 2.3.2. Social Autopoiesis Luhmann (1990) considers a social system (e.g., a group, a firm, a community) as a unit system to which autopoiesis is ascribed, that is, the social system is distinct from its participants. As we now know, autopoietic systems create their own structures and their own parts. If one considers individuals as the parts of a social system, the idea of a social system creating its own parts is absurd. A social system, in order to be autopoietic, needs to create social structures. The question, then, is to define what should be considered the constituent components of an organization. Luhmann (1990) resolves this dilemma by defining social systems as being realized in a domain of communications. For Luhmann (1990, cited in Capra 2002, p.83), the elements of a social system are communications: 83 Social systems use communication as their particular mode of autopoietic reproduction. Their elements are communications that are recursively produced and reproduced by a network of communications and that cannot exist outside of such a network. A group, for instance, uses networks of conversations that are inherently circular: a conversation triggers further conversations such that the network of conversations takes the form of multiple feedback loops. The group is a closed system in the sense that the network of conversations generates shared beliefs, explanations, values and so on, which are continually used and reused within conversations. The network of conversations that is created constitutes the boundary of the system. In contrast to a biological system that has a topological boundary, a social system has a social boundary that is characterized by expectations, confidentiality, loyalty, and so on (Capra 2002). This boundary is continually maintained and renegotiated within the network of communication and creates the identity of the group. As we discussed previously, human communication generates both meaning and social structures. The concept of “social structure” is central to social science, ranging from the works of the early founders of modern sociology such as Emile Durkheim (1858-1917) and Max Weber (1864-1920), to the work of Talcott Parsons (1902-1979) during the 1940s and 1950s, and finally to the works of contemporaries such as Anthony Giddens and Jürgen Habermas. Social structures consist of beliefs, values, norms, or practices, which we associate with a given culture. Capra (2002, p.87) explains: [C]ulture arises from a complex, highly nonlinear dynamic. It is created by a social network involving multiple feedback loops through which values, beliefs, and rules of conduct are continually communicated, modified, and sustained. It emerges from a network of communications among individuals and, as it emerges, it produces constraints on their actions. In other words, the social structures, or rules of behavior, that constrain the actions of individuals are produced and continually reinforced by their own network of 84 communications…People’s behavior is informed and restricted by their cultural identities, which in turn reinforces their sense of belonging. It is from the perspective of a self-generating network of communications that we will apply autopoiesis to social systems. The organization of a social system is its network of communications. Its structures are social structures that are continually selfgenerated by the system through communications and are always consistent with the system’s identity, i.e., its culture. The system’s cultural identity gives shape to the system’s boundary of meaning and expectations that constrain the system’s behavior. 2.4. Organizations as Living Systems The definition of organizations as living systems provides a fresh opportunity for the interpretation of communication and coordination processes; cooperation and competition; power relationships; the role of institutions and organizational rules; the role of information and tacit knowledge; along with the role of technology in supporting organizational performance. As emphasized in Chapter 1, organizational performance in today’s business context requires an organization to have both the capacity to innovate and to deliver products and services efficiently. Innovation and efficiency seem very much at odds, since one requires the organization to remain at the edge of chaos and the other requires the organization to consistently apply its business rules and protocols to remain in a state of equilibrium. We take a dual approach in the next chapter and contrast conventional organizational theories to the living organization theory in order to determine the components that contribute to organizational performance. 85 CHAPTER 3 Managing AEC Projects at the Edge of Chaos Complexity and living systems theories provide a new framework for understanding how AEC project organizations may efficiently deliver high value to clients in turbulent environments. Based on this theoretical framework, an AEC project organization is a living and complex adaptive system operating in an uncertain and always-changing environment. Faced with environmental discontinuities, a project organization selforganizes at the edge of chaos, i.e., far from equilibrium. In doing so, it may create new communication structures that emerge out of the interaction between project participants. Thus, a project organization is capable of double-loop learning – that is, it is able to perceive changes in its environment; challenge its existing structures and norms; and allow new communication patterns to emerge, which are better adapted to the situation. As we know, the edge of chaos is a paradoxical state that is characterized by a tension between stability and instability. Thus, two questions arise: (1) What is required for an AEC project organization to be able to exit the equilibrium state and migrate to the 86 edge of chaos? (2) What are the forces that push an AEC project organization to the edge of chaos and maintain it in this state long enough for it to self-organize and innovate? Or, in other words, what are the sources of project stability and instability at the edge of chaos? In this chapter, we combine findings from complexity theory and organizational theory to answer the above questions. First, we present a model of AEC project organizations as complex adaptive systems. Second, we briefly review the processes of self-organization and adaptation in a project. Third, we discuss the control parameters needed for a project organization to exit its stable state and migrate to the edge of chaos. Fourth, we discuss the concept of strange attractors in the context of a project organization and show how they may maintain a project organization at the edge of chaos. 3.1. AEC Project Organizations as Complex Adaptive Systems An AEC project organization may be considered as an ecosystem, i.e., a system of nested systems such as professional firms and diverse industry regulatory agencies (ecosystems in themselves), which interact with one another in a nonlinear way and exchange resources necessary to achieve project goals. A project organization is purposeful in that it aims to complete the design and construction of a facility. An AEC project is, in turn, a sub-component of the larger ecosystem that is the AEC industry, which is, itself, a component of our whole society. Project team members are linked by various types of contractual relationships and professional rules that determine the structure of the project organization; the project 87 delivery system; and ultimately, the way disciplines coordinate work with one another. The contractual agreements remain in place until each AEC professional firm has completed its tasks and the building is delivered to the client. At the project’s completion, the project organization dissolves. Thus, an AEC project organization is a temporary organization whose members’ interaction is limited in time (Mohsini 1992). The peculiar temporary characteristic of an AEC project is important because it influences the behavior of its participants along the project life cycle. Because of the heterogeneity of an AEC project organization and the strong interdependencies between activities of project participants, the performance of any one firm in the project ecosystem (local performance) directly impacts the performance of other firms involved in the project, as well as the performance of the project ecosystem as a whole (global performance). For instance, a firm’s poor performance may influence the quality of a project in which it is involved and consequently, it may also influence the performance of its project partners. The project may be delivered late and exceeds its budget, which may impact the client’s ability to operate the facility efficiently. In the case of a commercial building, for instance, the client may not be able to start operating the facility on time, which, down the line, will negatively affect the client’s customers and ultimately, the entire value chain in the project ecosystem. In a decentralized system such as an AEC project, local optimal performance does not guarantee global performance. Therefore, the focus should be on maximizing global performance. When multiple agents are involved in a situation, the optimal solutions are the ones that satisfy all sides (Guinier et al. 1997). Acknowledging that the success or failure of each 88 component profoundly affects the health of the whole should drive new professional behaviors based on cooperation, collaboration and trust. A project organization is a decentralized environment in terms of the physical location of its resources and components. However, even more importantly, project management and decision-making are also decentralized. In contrast to the traditional Master Builder of Antiquity who integrated all the design and construction activities in one single human brain, integration in today’s AEC projects occurs by connecting the “distributed collective intelligence” of all the individuals involved in a project. An AEC project organization is, therefore, a cognitive network (or knowledge network), and it may be compared to a virtual global brain in which information and ideas flow freely between AEC professionals (the neurons of the brain).1 The complexity inherent in a project organization resides in the many connections and nonlinear interactions that occur between its members. It is, therefore, an emergent property of the project. Decentralized decision-making and control means that it is impossible for a single professional in a team to dictate what the final design decision should be. Thus, in order for the process to move forward, it is critical for all the participants to be fully engaged so that they may provide input about their disciplinary constraints and use each other’s input to progressively adjust their own discipline-specific solutions. In this paradigm, the act of “designing” becomes a collaborative co-creation process among team members. In fact, the notion of the architect as a “Grand Designer” has always been something of an illusion, even in a traditional design-bid-build project delivery model, since an architect’s decisions are always constrained by the other disciplines’ requirements. Nevertheless, the structure of typical AEC contractual mechanisms often imposes linear and sequential 89 delivery methods that impede a tight integration between team members. In such a context, large and complex AEC project organizations struggle to deliver projects fast with high-level performance. In contrast, in integrated project teams, disciplinary solutions (e.g., program, design, construction and so on) unfold simultaneously for all the disciplines involved, through negotiations and mutual adjustments. Conflicts arising in an integrated project organization are resolved collaboratively, as each member knows that self-interest and group interests are not in contradiction. Rather, self- and group interests must be balanced because the project cannot be successful without the concomitant success of all of its members. Thus, there should be contractual incentives that promote positive, reinforcing reciprocity among project participants who must help one another in order to succeed. Such incentives, coupled with the need to integrate disciplinary knowledge, should encourage project team members to develop behaviors, based on cooperation and trust. Cooperation indicates a shared agreement between professionals to jointly support others in achieving common objectives. With cooperation, there is a real sense of belonging to a team and a willingness to make tradeoffs for the sake of the project. Trust is the lubricant that facilitates the decision-making process and the resolution of conflict. In addition, transparency is also critical for supporting an environment based on cooperation and trust because it facilitates the integration of distributed knowledge. Transparency, in this context, refers not only to the clear definition of handoffs, but also to the clear definition of goals and intentions of each of the participants. Aligning each individual discipline’s goals with the overall goals of the project should be an important activity in a project organization. 90 Considering a project organization as a living system challenges our traditional definition of control. The conventional definition of control implies central authority and compliance with a norm (i.e., the act of controlling is one of bringing the system back to its desired, initially planned state after its divergence). Conversely, in living systems, control is distributed among the system’s components: it reflects the components’ ability to self-regulate and self-control. This means that each professional in a project organization is actively involved in making numerous local informed decisions, using multiple feedback loops that take the input of all the team members into account. Thus, each project participant shares in the control of the project. When decisions have to be negotiated, judgment and values may be used to influence the project leader to make a decision in one way or another. 3.2. Self-Organization and Adaptation in AEC Project Organizations In routine times during the course of a project, AEC professionals collaborate to mutually adjust their disciplinary solutions and to anticipate potential design and construction problems; however, due to the complex interdependencies between project components and uncertainties in the environment, many issues inevitably arise along a project life cycle, which cannot be foreseen. Even in well-integrated project teams, such situations are unavoidable and greatly impact project performance. Faced with discontinuities in the environment, an AEC project organization must self-organize. As we know, self-organizing systems are open systems interacting with their environment and dynamically adapting to the forces outside of their own boundaries. In the context of an AEC project, self-organization may occur all along the project life cycle 91 – for instance, when a project organization’s existing structures and standard work processes are not effective in the face of an adaptive challenge, or when an unforeseen event interrupts the flow of activities; challenges the plan; and/or disrupts the process. When contingencies arise, AEC professionals should be able to reflect upon their project environment and flexibly re-organize their project team so that it can better respond to the crisis. In other words, the project team must be capable of double-loop learning. Monge and Eisenberg (1987, p.331) describe the process of “reorganizing” as a “process by which linkages are continually added and deleted as organizational members respond to external and internal factors.” In doing so, a project team creates new project activity patterns: new participants may be involved; the communication flow and types of interactions between participants may be redefined; and new work processes may be implemented. Therefore, in an adaptive project organization, the pattern of activity is never static or predetermined; instead, it varies in accordance with the particular needs of the client; the project team; the project scope; the environment; and the specific issues that need to be addressed. The process of reorganizing may not be driven from the top (i.e., from a central authority such as the architect or project manager), but may occur as team members interact with one another based on the needs of the moment. In doing so, the team members create an informal communication network that is superimposed on existing formal project structures. The informal communication network, therefore, emerges as new communication structures are built between participants based on their information needs. Emergent communication networks are not planned by a project team and thus, cannot be anticipated; they are different from one project to another, from one project 92 phase to another, or from one project activity to another. Thus, one may say that the project organization is a self-generating communication network that creates new connections and structures as needed. In order for an informal communication network to emerge, the boundaries of a project organization need to be very permeable, letting information flow easily among project participants, the project and its (external) environment. Without such input, professionals cannot gain awareness about the state of a project. Consequently, the project as a whole lacks the required intelligence for learning and adapting. When an issue arises, if the team’s internally available resources are not sufficient for solving the problem, the team should have the organizational flexibility to look for expertise beyond existing organizational boundaries. This requires professionals to tap into their social networks to find the expertise necessary for solving an issue. Once the expertise is located, the team must be able to integrate new professionals into the existing project team framework. We may recall that an autopoietic system can only behave in a way that is consistent with itself and its history of development and evolution. As noted by Wheatley (1992, p.94), “Self-reference is what facilitates orderly change in turbulent environments” because self-reference helps a self-organizing system to coordinate the actions of all its individual parts in a consistent way. Thus, when a project organization is required to change, it may do so in a way that is consistent with its identity. Essentially, the project’s established values, culture, traditions, aspirations and competencies may serve as reference points as the project participants self-organize in a way that is coherent and that facilitates adaptation. Due to its heterogeneity, an AEC 93 project team faces many challenges when attempting to build a shared culture among its members so as to facilitate its adaptive capacity. As we have learned from complexity theory, creativity and double-loop learning cannot occur within a system’s state of equilibrium: a complex adaptive system selforganizes at the edge of chaos. Therefore, in order for an AEC project organization to adapt to discontinuities in its environment, it must be able to exit its stable state and migrate to the edge of chaos. In the next section, we analyze the conditions necessary for a project organization to be able to migrate to and then maintain itself at the edge of chaos. These conditions are driven by control parameters whose levels indicate whether a complex adaptive system will remain in the stable zone, migrate to the edge of chaos, or disintegrate into complete chaos. In Section 4, we discuss the effects of strange attractors in the context of a project organization and how they operate as sources of both stability and instability. 3.3. Influencing the Control Parameters Three control parameters influence and indicate a complex adaptive system’s ability to migrate to the edge of chaos: (1) the rate of information flow; (2) the degree of diversity among the components of the system; and (3) the richness of connectivity between components. In the realm of human systems, Stacey (1996) argues that there are two additional control parameters that need to be included in the list: (4) the level of contained anxiety; and (5) the degree of power differentials. As Stacey (1996) points out, there are critical points in parameter values, which an organization needs to reach, in order to exit the equilibrium space and move to the edge of chaos. Beyond these critical 94 points, however, an organization may lack sufficient structure for containment and disintegrate into complete chaos. Therefore, an adaptive organization must figure out: (1) what the critical points for each control parameter are; and (2) how to reach and remain at these critical points until an adaptive challenge is overcome. Unfortunately, finding the critical points is not a hard science: they may change based on the internal context of the organization and its external environment. Often, the best that can be done is to identify the variables that influence control parameters levels in one direction or another and to recognize that any critical point’s specific location may remain context-specific. Variables that any organization can influence in order to affect control parameters are: the organizational structure; the communication and coordination processes; the technologies used to support communication and team interaction; organizational culture; and the culture and behavior of the members of the organization. By taking into consideration these variables in the modification of its strategy, an organization may also influence its own environment that may in turn affect the control parameters. This dynamic is shown in Figure 3.1. For each control parameter, we first consider its impact on a project organization’s ability to move to the edge of chaos. Second, we use findings in organizational theory to assess how a project’s organizational variables influence the control parameters. 95 influences ENVIRONMENT Control Parameters influences Organizational Variables STRUCTURE PROCESSES CULTURE & BEHAVIOR TECHNOLOGY influences RATE OF INFORMATION FLOW AGENTS’ SCHEMAS DIVERSITY RICHNESS OF CONNECTIONS LEVEL OF CONTAINED ANXIETY DEGREE OF POWER DIFFERENTIAL Figure 3.1: Organizational Variables’ Influence on the Control Parameters 3.3.1. Rate of Information Flow Whether a project organization is able to move to the edge of chaos or not depends on its ability to operate as a dynamic communication network that thrives on information and knowledge sharing (Lesser & Prusak 2004). However, there are limits to the capacity of an organization and/or the abilities of individuals to absorb information. When crossdisciplinary information is not available, AEC professionals have no opportunity to perceive discontinuities within their own disciplinary environments and thus, they may not be able to mutually adjust their activities in a timely fashion. Conversely, when too much information flows between project participants, too much ‘noise’ may be created, which may be overwhelming for project participants because every aspect of the project 96 may be constantly changing for every discipline. Lacking the requisite stability, the project organization may become completely chaotic. The rate of information flow is dictated by connectivity levels and interactivity in the team. Connectivity relates to ongoing accessibility to – and communication of – distributed knowledge resources essential to project activities (i.e., accessibility to explicit knowledge). Interactivity relates to collaboration, i.e., people’s ability to work with one another and make decisions, both virtually and in real time (i.e., integration of tacit knowledge). Connectivity is a prerequisite to interactivity: people exchange information and knowledge as they work with one another to achieve a common goal (Nonaka & Konno 1998).2 The more rigid and hierarchical the project organizational structures, delivery models and formal communication processes are, the less connectivity and interactivity may occur between team members and therefore, the slower the rate of information flow may be. For instance, the linearity of design and construction activities in the traditional design-bid-build delivery method hinders the transfer of information and knowledge about construction activities to the design process. Because the contractor’s feedback occurs after design decisions are made, a project team is often placed in a reactive mode – forced to fix problems after the fact instead of being able to anticipate them. Thus, one may expect the rate of information flow to be higher in a design-build model in which the design and construction activities run in parallel with one another. Professionals may supplement, if not bypass, the formal organizational structures by creating ad hoc, informal networks of communication, through which information may flow more easily. Thus, the flow of information is influenced by a project team’s internal ability to self97 organize and its capacity to develop linkages externally with the professionals possessing the expertise necessary to the project activities. It is often the case, both in informal knowledge networks (Lesser & Prusak 2004) and in highly centralized structures (Jin & Levitt 1996), that one member causes a bottleneck because s/he has the knowledge that is required by all the other network members. A bottleneck will greatly reduce the flow of information. Thus, it needs to be removed by having that member delegate responsibilities to others. The development of information technology – the advance of the Internet, in particular – has greatly increased connectivity levels between AEC professionals. Information technology facilitates both connections to disciplinary information and explicit knowledge that resides in each professional firm and the exchange of information in real time, which is essential to shortening the feedback loops between project members. In addition, Web-based collaboration tools facilitate synchronous and asynchronous collaboration between professionals. To improve connectivity, a project team must have the ability to effectively manage the project knowledge and information (i.e., explicit knowledge). The management of explicit knowledge consists of the processes of codification, distribution, and retrieval of knowledge that is captured in a digital form such as specifications, standard procedures, guidelines and even formal work processes. The management of knowledge is based on the development and maintenance of data and information repositories such as databases, libraries, content management systems, expert systems and so on; and also on the ability to retrieve information located in these repositories through search engines, along with visualization and representation tools. When 98 effectively managed, explicit knowledge can therefore be easily transferred from one individual to another or from one firm to another. Real-time interactivity between project participants facilitates the integration of inter-disciplinary tacit knowledge (i.e., knowledge based on people’s accumulated experience and their capacity to apply knowledge in their work). Effective integration consists of “cross-learning,” i.e., developing a mode of interaction between experts so that knowledge possessed by, for instance, a structural engineer is integrated with an architect’s knowledge about space design and a general contractor’s knowledge about construction processes so that the design may integrate spatial, structural and constructibility issues into a coherent whole. Integration of knowledge occurs as design issues are solved within a specific context. It requires professionals to see the connections between their own disciplinary constraints; the constraints of their peers; and the context of the project. An organization’s ability to develop its culture and behavior around the sharing of knowledge is called the organization’s “information ecology” (Davenport, cited in Schrage 1997). A project’s information ecology based on knowledge sharing, a collaborative culture and behaviors are required for an AEC project to become an adaptive system. In order to manage its information ecology, an adaptive project requires a particular style of management and leadership, which is characterized by “tolerance for paradox;” a letting go of control; and the fostering of interactions and relationships between project members.3 Instead of imposing a path, a “true” leader allows new processes to emerge freely within certain guidelines (Lewin & Regine 2001). A leader 99 facilitates mutual adjustment precisely because it spurs creativity and the creation of new knowledge. It is true, however, that too much transparency in disciplinary information may drive a project team into a chaotic state. As Cross et al. (2004) point out, “The quantity of communication in any situation has no necessary bearing on its quality” (cited in Lesser & Prusak 2004, p.49). The “right” information needs to flow to the “right” people at the “right” time. Defining what information should be shared, with whom, when and how often is the focus of a lean production system and effective management (Ballard 2000). It should be an important activity at the planning phase, as well as along a project life cycle. Moreover, Cross et al. (2004) add that connecting everyone to every one else is not always desirable because, “[R]ather than leading to improved collaboration and problem solving, an indiscriminate increase in connectedness can be a drag on productivity, as people get bogged down maintaining all their relationships” (cited in Lesser & Prusak 2004, p.51). In contrast, the authors advise that professionals should be connected when there is a strategic benefit in doing so – that is, when AEC professionals face an adaptive challenge. 3.3.2. Diversity between AEC Professionals’ Schemas The schemas of an organization’s agents are dictated both by the agents’ mental models (the conscious and unconscious rules and patterns of behavior that individuals typically follow, which have been acquired over time from experience and education) and the organizational institutions that constitute the framework within which the agents may behave and which create the culture of an organization (Stacey 1996).4 A low level of 100 diversity among agents’ schemas is usually characterized by conforming members who operate in the legitimate system and do not challenge the organization’s established rules. An organization with a low level of diversity among agents’ schemas will remain in the equilibrium state. A high level of diversity among agents’ schemas is characterized by members who are able to challenge the pre-established institutional framework imposed by an organization. An organization that presents too much diversity among its agents’ schemas will tend to move toward anarchy. Ideally, an organization should have a level of diversity, which is high enough to allow the members of an organization to question the existing state, but not so high as to inhibit the maintenance of some level of cohesion. This is the particular state required for an organization to move toward and maintain itself at the edge of chaos. One may expect that the level of diversity among agents’ schemas may be lower within the context of a single firm than within the context of a cross-organizational network such as an AEC project organization. Indeed, it has been demonstrated that one of the benefits a firm provides is to create the context in which a sense of community may be developed around the firm’s identity and culture (Kogut & Zander 1996). Perhaps a firm designs its institutional framework to minimize uncertainties by structuring human interactions. If the identity and culture of a firm are built around the enforcement of rules and procedures (as is the case in bureaucracies), the adaptive capacity of the firm will be low, especially if all firm’s members identify with the firm’s identity and abide by its rules. Indeed, a high level of homogeneity will influence every member of an organization to perceive environmental discontinuities in exactly the same way, thus potentially missing out on things that could impact the organization. 101 While a firm may face the challenge of increasing its level of diversity in order to maintain its adaptive capability, a cross-organizational network may face the need to homogenize. Obviously, one of the benefits of a network resides in its heterogeneity: activities in a network build on differential advantage such as specialization, know-how, expertise, culture and so on (Thorelli 1986). However, excessive diversity creates instability and an environment prone to confrontation. When one considers the context of the AEC industry, it may seem obvious that a very high level of heterogeneity inherent in AEC professionals’ specialization and the diversity of projects’ social structures (i.e., diversity of professionals’ beliefs, values, norms, practices, culture and so on) are often at the root of conflicts within projects. Only through an understanding of the foundations of AEC professionals’ belief systems can the various components that drive common behaviors in current AEC industry practice be understood and potentially changed to facilitate the emergence of a new AEC adaptive business model. AEC professionals’ mental models are the product of over a hundred years of practice and tradition and they may be explained through an analysis of the process of professionalization. Professionalization defines the framework of practice; dictates the structure of power between professionals; defines the codes of ethics; defines a set of rules for the participation of many interested parties in a project; and shapes the professionals’ values and beliefs. In particular, the socialization of architects, which occurred within a broader community that includes other disciplines such as contractors and engineers, has shaped the behavior of these disciplines respective to the architects. The process of professionalization entails three key activities: (1) the development of formal education and training, which provide professional technical expertise; bestow 102 esoteric knowledge in the subject area; and enforce authority over the lay person and in some cases, over other professionals; (2) the development of professional associations, whose role is to guarantee first, the technical efficiency of its members by testing their ability to practice before being admitted (through the process of licensing, for instance); second, an imposed code of ethics; and third, the protection of the profession from the invasion of those who are unqualified; (3) the development of a monopoly on the market served by the profession in order to differentiate a profession from mere business activities (Hugues 1963; Larson 1977; Marschall 1939; Parsons 1939). The professionalization of architecture, however, has not been fully successful, due to the inherent nature of the profession, which is doubly positioned between ‘profession’ and ‘art’ on the one hand, and ‘profession’ and ‘business’ on the other (Dostoglu 1986). First, the architect’s authority over other project participants has always been problematic because architectural competency based on scientific rationality has been difficult to claim, due to the fact that there is no ultimate objective, scientific knowledge in architecture (as opposed to engineering that is considered largely a science, not an art form). Second, the American Institute of Architects (AIA) that was founded in 1836 has made much effort to reinforce the identity of the architectural profession through a process of exclusion of other professionals related to the building industry – builders, contractors, developers, surveyors, draftsmen and amateur architects. While the separation of architect and builder (and other professions) has provided the architect with a distinct professional identity and scope of practice (and thus, increased autonomy over the building professions), the separation has also created greater dependence by the 103 architect on other professionals who are essential to the architect’s successful completion of his/her task. Since control has to be distributed among the project participants, distribution of power becomes a problem in time of crisis, as each party may fight for his individual solution on the ground that each possesses the requisite knowledge and technical expertise. As Larson (1993) points out, the contradiction between the autonomous pursuit of architecture and its inherent heteronomous conditions results in an essential complexity in architectural professional practice. The conflict rests in old beliefs implicit in professional practice and the antagonistic attitudes and behaviors that follow, thus furthering conflicts and limiting the potential for cooperation. Third, professional monopoly over the market has been difficult to achieve for architects because of the episodic relationships they hold with their clients; the fact that clients may directly address builders; and that architects’ practice of design and supervision of building construction must be shared. Therefore, it is evident that the traditional definition of professionalism, as a specific form of domination of one set of experts over another, is very much at odds with the current, information-and knowledge-based business environment. The integration of knowledge across professional and organizational boundaries requires a new definition of the professional – one in which the professional finds his legitimacy within the context of collaboration with other parties. Social and institutional change, however, is a slow, if not moribund, process. First, institutional change is path-dependent, i.e., it is dependent on the history of previous organizational changes (North 1992). Therefore, whatever new rules may be conceived will remain more or less consistent with the existing industry and firm 104 framework, thus impeding radical institutional innovation. Second, the process of institutional change is very slow and incremental. Values that people have assimilated over time through a process of deep enculturation and passed on to others though generations is not always conscious; thus, it is difficult to change the entrenched values of the people (North 1992). Individuals’ understanding of what rules to follow and when – or, what behavior to demonstrate in a given context – is implicit and embedded within their practice (Tsoukas 1996). When professionals are unaware of their mental models, they cannot evaluate their origins and thus, they cannot change them in response to changes in the environment. The resultant inertia can impede learning and institutional changes. Nevertheless, to function effectively as a network, a project organization faces the needs to create a project identity and culture, which are different from, yet consistent with, the ones that exist in its constituting firms. Goal alignment is critical to team integration. The development of a shared vision, as well as explicit and implicit rules of conduct, is necessary for smooth network operation (Jarvenpaa & Ives 1994). In order to develop such a vision, partners in the network will need to learn from each others’ behavior and, as noted by Powel (1998), develop routines and norms that can mitigate the risks of opportunism. To facilitate integration, Mitchell (1979, p.96) argues for a process of “centralizing the concepts.” As he puts it: To optimize results, the centralization of the concepts has to be aligned with selforganizing systems. The development of self-organizing systems, combined with centralized concepts, is the key to successful strategic control. Centralizing the concepts means defining the “rules of the game,” which each of the team members must follow to facilitate coordination and integration. What may be 105 centralized are critical concepts such as vision and goals; communication and behavioral protocols between organizations; inter-organizational communication standards; performance measures and critical success factors; conflict resolution processes; and other strategic issues. What should not be centralized are those rules that would prevent the team members from being flexible and undermine their ability to adapt to contingencies. In particular, rules that dictate how to do an action, instead of what needs to be done, should not be centralized. 3.3.3. Richness of Connections While connectivity is a critical variable that influences a network’s ability to selforganize, the quality of the connection between agents and the embeddedness of relationships (i.e., social relations) are even more important variables that can affect the efficiency of economics activities (Granovetter 1973; Uzzi 1997). Rich connections are characterized by repeated interactions (often face-to-face), trust, reciprocity, collaboration, cooperation, integration, low level of self-interest, altruism, personal ties and so on. Rich connections promote knowledge transfer and learning, while lowering transaction costs (Uzzi 1997). While neo-classical economics theory refers to market transactions as the most efficient mechanism for optimizing the allocation of resources based on the rational behavior of self-interested agents, it has been demonstrated that interfirm networks provide for embedded exchanges that cannot be found in marketbased transactions that are derived from arm’s-length ties (Nooteboom et al. 2000; Powell 1990; Uzzi 1997). 106 In his study of interfirm networks, Uzzi (1997, p.42) found that embeddedness characterized by high levels of trust in relationships between network participants facilitates “fine-grained information transfer.” Communication in such networks was ongoing and based on the sharing of strategic information and tacit knowledge critical to forecasting, thus, helping network participants to make better decisions and become more responsive to change. Because of their strong relationships, the network participants were motivated to “do more than the letter of a contract” and they were better able to develop problem-solving mechanisms that facilitated coordination and resolution of problems “on the fly,” thus minimizing rework. Therefore, in an AEC project organization, the tighter the relationship between participants along the project’s life cycle, the higher the degree of embeddedness and trust among participants and the richer the connections will be. Overembeddedness, however, may have a negative impact on the performance of a network. Here again, there is a critical threshold that defines whether the networked organization will be able to migrate towards the edge of chaos or will slip over the edge into chaos itself. In a well-known paper called “The Strength of Weak Ties,” Mark Granovetter (1973) demonstrated that knowledge and information, which circulates between members of a group via strong ties, will be locked within the group, thus impeding their dissemination across groups. Indeed, strong ties are based on a closed network structure: participants in such networks may not have the opportunity to gain from new insights and knowledge acquired through the inclusion of new expertise in the group. Conversely, the benefit of weak ties stems from its facilitation of relationships between groups. Weak ties are loose relationships that individuals build as they 107 encounter each other – for instance, when professionals build work relationships over the duration of a project. When the project organization dissolves and professionals move on to work on other projects, the ties that have been created remain (in particular if the working experience has been a rewarding one) and may be leveraged, later on, when needed. However, if ties between a project’s participants are too strong, feelings of obligation and friendship may be hindrances at times when a professional needs to expand his relationships to include new partners who may have the resources necessary to solve a particular problem. Developing a relationship with a new partner outside of the project network could be perceived as some kind of defection and may therefore spoil or detract from the relationships in the project. Indeed, rich connections are extremely vulnerable to any perceived defection. Axelrod (1984) has shown that even in a simple game such as tit-for-tat, any misunderstanding or perception that one of the members of a network is cheating or defecting will likely spoil the relationship, sometimes permanently, and it will provoke retaliations from the members who feel that they have been cheated. Cheating may even have long-term repercussions that occur beyond the life cycle of a project, as the reputation of a professional who has cheated may haunt that professional in other business contexts. Transparency of intent, clear talk and straightforward actions will prevent misunderstanding. 108 3.3.4. Level of Contained Anxiety Many people would argue that anxiety is a feeling that is counter-productive when dealing with change and creativity. Stacey (1996), however, affirms the opposite. As he points out, “Creativity requires that we are able to hold, rather than avoid, high levels of anxiety, because only then can we perform complex learning” (Stacey 1996, p.222). To demonstrate his point, Stacey (1996) borrows from psychoanalytic studies of individual creativity. He explains that at an early stage of development, infants do not distinguish between what is inside and outside their mind, that is, between internal fantasies and external reality. Since the earliest fantasies are those of omnipotence and omniscience, infants believe that they are the center of everything. Satisfaction of an infant’s every need confirms this fantasy, whereas frustration of needs is experienced as persecution and generates fear, anger, violence, hate and ultimately, anxiety. Humans’ maturation process consists of the separation between the mind and the outside world, which requires the gradual abandonment of the earliest fantasies of omnipotence. This only occurs if the infant has experienced some frustrations; if the infant is always satisfied, s/he will not abandon the fantasies. Thus, the maturation process generates anxiety because of the tensions between opposite feelings: love and hate; dominance and submission; separation (or differentiation) and ultimate belonging or conformity. If these tensions lead to too much anxiety, the infant will not be able to cope with the maturation experience; with too little anxiety, the infant will not embark on the maturation process. Thus, the process of maturation requires strategies that allow an infant to hold high levels of anxiety. Similarly, the creative process (i.e., the process of change) in an individual requires strategies to hold an individual’s high levels of anxiety. These strategies must create 109 some level of frustration, but not too much. That is, an individual needs to be able to hold ambiguities and paradoxes without being overwhelmed by anxiety. What does this means for professionals in an organizational setting? The level of anxiety in individuals and organizations may be raised by environmental discontinuities and their consequent organizational changes, since they force individuals and organizations to disrupt their routines and standard activities and move to an uncomfortable zone characterized by uncertainties and chaos. It has also been demonstrated that very high levels of integration between diverse professionals and experts working with one another can generate anxiety. Allen (1994) showed that the implementation of information systems designed to increase activity integration among different experts by providing real time information increases the level of “mutual control.” Such systems force professionals to look for problems outside their own domains and to coordinate efforts in response to others’ unexpected changes. However, working in such a dynamic environment often increases psychological pressures on the individuals concerned and provokes anxiety in the individuals who may not have the skills and/or abilities to cope with such an integrated environment. Conversely, the implementation of routine and standard procedures is always a means to contain individuals’ anxiety levels since it contributes to a stable environment. Conformity to organizational institutions or to an industry’s implicit business rules – for instance, conformity to professional norms as defined by a professional association such as the AIA in architecture – may reduce professionals’ levels of anxiety by influencing them to make business decisions that are consistent with the accepted norms. At an individual level, people devise strategies such as using defensive mechanisms or 110 developing comforting fantasies connected with childhood in order to avoid anxiety (Stacey 1996). Unfortunately, these strategies increase resistance to change and their counterparts at the organizational levels hold an organization outside of the edge of chaos. While maintaining the level of anxiety is critical if one wants to avoid disintegration, the strategies used to do so must not keep an organization in the equilibrium state. Stacey (1996) posits three means by which anxiety levels can be controlled, but not avoided, within an organization at the edge of chaos. These means are: (1) improving the quality of the relationships between the individuals in an organization or group; (2) using self-reflection; and (3) improving the quality of leadership and the way power is exercised. A trusting environment allows the members of an organization to contain their levels of anxiety because they feel mutually supported by their peers. Feeling supported increases the likelihood that they will be willing to take risks by raising issues and concerns that may destabilize the organization and drive it to the edge of chaos. Selfreflection allows members of an organization to reflect upon the system of which they are a part and engage in transformative work (double-loop learning) that both provokes and contains anxiety when performed in a supportive and trusting environment. Leadership, as opposed to authority, is another means to hold anxiety in a group. By showing tolerance for paradox, fostering interactions and influencing the group members to freely contribute, a leader may help a team to hold its level of anxiety, while the team selforganizes in response to an adaptive challenge. In some cases, as pointed out by Stacey (1996), the leader may hold the anxiety for the team; in other cases, all members of the team may hold anxiety. While a leader may increase the level of anxiety in a team by 111 challenging the team and forcing its members to descend from their individual fitness peak into the edge of chaos, a good leader may also encourage a team to define meaning and a sense of purpose in an incoherent and chaotic environment, thus lowering the level of anxiety of the team members. Meaning and sense of purpose help an organization transcend the messiness of its organizational context and provide a point of reference on which individuals can base their actions (Wheatley 1992). 3.3.5. Level of Power Differentials The type of power, which is exercised in an organization, greatly influences the organization’s ability to move to the edge of chaos. Power, here, is defined as the ability to influence the decisions or actions of others (Thorelli 1986). A power differential is created by a leader-follower differentiation. Extreme power differential (i.e., when an elite owns most of the power and exercises it by oppressive authority) may create fear among members of an organization who, consequently, will not perform within the shadow system and will not challenge the authority imposed on them: in such a case, the organization will operate in the stable zone. Conversely, if power is evenly distributed among all the members of an organization and no one really exercises that power, the organization may disintegrate into a chaotic state. Between these two extremes, there is a situation where anxiety is contained by a clear hierarchical structure that enables leadership. Yet, at the same time, organization members are not afraid to express their opinions and to challenge the status quo (Stacey 1996). Interdependence in a network influences the distribution of power between members of the network; thus, understanding the flow of power and information is 112 critical. In a network such as an AEC project organization, each professional firm holds a “position” of relative power over, and influence on, the other members of the network. Each firm’s position is influenced by: (1) the particular role played by the firm in the network (i.e., division of labor); (2) the power a particular firm may exercise over the other members of the network; and (3) each firm’s respective power in the larger ecosystem (e.g., AEC industry) (Thorelli 1986). Thus, each firm’s position in a network is relative to the position of the others and firms’ relative positions may change over time. For instance, AEC professionals involved in a project hold positions based on the role their discipline plays in the project: the architect who oversees the entire project development may have a stronger position than an HVAC consultant, whose scope of work and influence may be limited. In a design-build project, the contractor may have a position stronger than, if not equal to, the position of the designers, since the structure of the project may give him the overall project responsibility. A firm that benefits from a very high reputation in its field, or in the overall AEC industry, may be able to use its reputation to increase its influence and power over other project participants. Thorelli (1986) specifies five different, but interdependent, sources of power in a network that influence power differentials. These sources of power are: economic base, technology, expertise, trust and legitimacy. In an AEC project, economic power may be manifested through the client’s decisions over budget allocation and the contractor’s ability to remain within budget. Technological superiority may be expressed through a discipline’s capacity to innovate, or it might be evidenced in the processes used, the amount and type of knowledge owned and so on. Expertise is reflected in professionals’ capability to apply their knowledge and expertise to resolve project issues. Trust may 113 also be a source of power when it is reciprocal and it forces participants to cooperate to avoid retaliation from other members of the team. Formal legitimacy as a source of power may originate from the type of contractual mechanism used between project participants (e.g., long-term relationships, joint venture arrangements and so on) and ownership of the project. As Thorelli (1986, p.41) points out, “[T]he daily activities in the network may be viewed as either building or drawing on differential advantage.” This is done though the multiplicity of connections between the network members. Along a project life cycle, power differentials may change, depending on the particular involvement of different professionals at a given time; the context of the involvement; and the specific issues to solve. When power is exercised exclusively by one member of a project team, one may assume this will increase the level of anxiety of the other participants and that the project team will not be able to self-organize as needed. Similarly, if all project participants hold the same level of power at all times and no one exercises that power, anxiety will also arise because of the lack of leadership, which may send the project into an anarchic state. 3.3.6. Summary Through a review of findings in organizational theory, we have identified specific organizational variables that influence the level of the control parameters and thus dictate whether a project organization will be able to migrate to the edge of chaos. Figures 3.2a through 3.2e summarize the organizational variables that we have discussed in this chapter and that influence each control parameter. The list of identified variables is not exclusive and further research may complete it. 114 Control Point _ + Rate of Information Flow Organizational variables - Level of connectivity and interactivity - Rigid, hierarchical structure versus flat, network structure - Project delivery method - Informal communication network - Information bottleneck - Information quality - Information technologies - Information management system - Asynchronous/synchronous communication - Feedback delay - Formalization, i.e., type of communication/process procedures (rigid versus flexible) - Mutual adjustment - Other, as identified by further research Figure 3.2a: Variables Influencing the Rate of Information Flow 115 Control Point _ + Agents’ Schemas Diversity Organizational variables - Diversity of culture, beliefs, norms, and values - Diversity of language - Diversity of practice - Type of organizations and institutions involved - Professionals’ personality - Project culture - Professionals’ differences in term of codes of ethic - Experience in working in a heterogeneous environment* - Degree of individualism - Goal alignment - Shared vision - Level of agreement on project rules, protocols, and standards - Explicit and implicit rules of conduct - Others, as identified by further research * Variables that apply to the individual firms and professionals involved in a project and not to the Figure 3.2b: Variables Influencing Agents’ Schema Diversity 116 Control Point _ + Richness of Connections Organizational variables - Nature of personal and organizational ties (i.e., personal, arm’s-length ties, weak, strong) - Cooperation - Trust - Reciprocity - Integration - Self-interest* - Altruism* - Sharing of strategic information - Rivalry - Defection - Transparency of action, clear talk and straightforward actions - Tolerance for paradox - Others, as identified by further research * Variables that apply to the individual firms and professionals involved in a project and not to the overall project organization Figure 3.2c: Variables Influencing the Richness of the Connections 117 Control Point _ + Level of Contained Anxiety Organizational variables - Internal tensions - Level of frustration* - Ability to hold ambiguities and paradoxes* - High level of integration, mutual control - Routines, standard procedures - Level of conformity to professionals norms - Defensive mechanisms, resistance to changes* - Quality of professional relationships - Individual and team self-reflection - Leadership - Methods used to exercise power - Others, as identified by further research * Variables that apply to the individual firms and professionals involved in a project and not to the Figure 3.2d: Variables Influencing the Level of Contained Anxiety 118 Control Point _ + Level of Power Differentials Organizational variables - Distribution of power - Level of interdependence - Ability to influence others* - Exercise of power - Position in the organization* - Roles and division of labor - Reputation in the industry* - Economic base* - Technological influence* - Expertise* - Trust - Legitimacy* - Others, as identified by further research * Variables that apply to the individual firms and professionals involved in a project and not to the Figure 3.2e: Variables Influencing the Level of Contained Anxiety 119 3.4. The Power of Strange Attractors in AEC Project Organizations Developing effective strategies to maintain the five control parameters we just discussed at the control points is critical in order for an adaptive organization to have the capacity to move to the edge of chaos when it is challenged by environmental discontinuities. In this section, we discuss the forces that push an organization to the edge of chaos, maintain it there and allow the organization to self-organize without disintegrating. These forces are strange attractors whose concept was introduced in the previous chapter. As we know, the complex interplay between positive and negative feedback processes in a system creates a strange attractor – a basin of attraction into which a system is drawn. A strange attractor creates simultaneous stability and instability in a system and lures the system to the edge of chaos. A system’s behavior is free within the limits of the attractor’s boundaries. The system is able to learn, self-organize and adapt to discontinuities in its environment. One should not think of a strange attractor as a single thing that causes a system to behave in a particular way. A strange attractor consists of a set of factors that interact with one another and catalyze a system’s emergent behavior (Pascale et al. 2000). A strange attractor is a force – or, more precisely, a tension between opposite forces – that shapes the independent behaviors of diverse individuals to move in the same direction without centralized control. Stacey (1996) identifies three sources of instability in a complex adaptive system: (1) sensitivity to initial conditions (i.e., positive feedback); (2) competition; and (3) exposure to creative tension set up by play in recessive schemas (i.e., learning capability). The sources of stability in a system are the opposite of the conditions that create 120 instability: (1) constraints (negative feedback); (2) cooperation; and (3) protection from creative tension by dominant schemas (i.e., inability to learn). In this section, we discuss how strange attractors operate in the context of an AEC project organization. Since a creative tension between the sources of stability and instability needs to be present at the edge of chaos, we review the sources listed above in pairs. 3.4.1. Balancing Positive and Negative Feedback Processes As complexity theory indicates, positive feedback is necessary for change and adaptation. Without it, an organization cannot exit the equilibrium state and migrate to the edge of chaos. While positive feedback processes push an organization outside of its comfort zone and force it to change, built-in negative feedback processes create resistance to change and bring it back toward equilibrium. When negative feedback processes prevail over positive feedback processes, an organization will maintain itself at equilibrium. Conversely, positive feedback with no balancing processes may create a chaotic state. In contrast, the synergy between positive and negative feedback processes creates a strange attractor that maintains a system at the edge of chaos and allows it to self-organize and adapt to its environment. (It is the interplay between positive and negative feedback processes, which allows the human body to achieve homeostasis, i.e., internal balance.) In the same way that tiny errors in the initial input in Lorenz’s weather model were amplified to create a completely unexpected weather pattern, positive feedback processes in an organization can amplify the effects of seemingly unimportant events or decisions in ways that can dramatically change the future of the organization. Popular 121 terms such as self-fulfilling prophecies, vicious or virtuous cycles, bandwagon effects, chain reactions, momentum and so on are all examples of reinforcing feedback loops. For instance, in an AEC project organization, a client’s decision to build or not to build a facility, combined with economic changes, competition and other external factors, may have long-term consequences that the client cannot foresee. Decisions made at the planning phase or at the design phase – for instance, the choice of a particular material or structural component may lock the project team into a particular path that may have longterm positive or negative consequences. For example, material prices may dramatically increase, thus forcing the client to increase its budget or the designers to revise their design, which will consequently impact the project schedule. The example illustrates the importance of path-dependent processes, i.e., that historical accidents plus increasing returns can play a critical role in locking-in a particular outcome over another, which may have been rationally preferable.5 Positive feedback may also create good outcomes: in the example above, the client may have benefited from a fall in material prices. Positive feedback can also lock-in very good products through “word of mouth” and trend factors as is the case with the economic success of the iPod made by Apple. But, as Peter Senge (1990) explains, positive feedback cannot continue unchecked in nature. Eventually a balancing effect (i.e., negative feedback) will occur that will limit and reverse the acceleration of change. Negative feedback processes are at the basis of all goal-oriented and control mechanisms in a project organization. Any control process within a project requires that the team has predetermined its objectives and standards that have to be met and that constitute the measures of organizational performance. Professionals define objectives and standards through planning processes 122 and monitor the achievement of these objectives and standards through negative feedback processes. Therefore, the planning process usually focuses on the achievement of longterm goals. While the use of negative feedback is predominant in traditional management theory, the impact of positive feedback has often been ignored or minimized because it is commonly believed that there will always be a negative feedback process to counteract the effect of positive feedback and to bring an organization back to its equilibrium state. Indeed, within the rationalist tradition, it is argued that one of the main activities of social systems, and organizations in particular, is to develop tools and methods that facilitate rational decision-making – usually through planning activities and a centralized model of control. However, as the examples above have shown, the effects of positive feedback processes are indeterminate and thus, very difficult to predict. As thorough as it may be, planning cannot anticipate the outcome of positive feedback processes. It is often the case in organizations that negative feedback processes take the shape of resistance to change. Senge (1990) points out that the balancing processes used by organizations are both explicit and implicit. More often than not, resistance is an implicit process that will impede change within an organization, when overlooked or ignored. Resistance usually arises from threats to traditional norms, mental models and institutions, which are embedded within an organization’s structure, power relationships, control mechanisms and distribution of authority. Consequently, an organization that faces an adaptive challenge will need to identify the sources of resistance and devise means to overcome them. 123 Finally, most feedback mechanisms operate with a delay. For instance, customers’ satisfaction (or lack of it) may take some time to get fed back into an organization’s production system. If the delay is significant, the organization may not have the chance to react to the market’s needs in time and may miss the mark in improving its product’s functionality. In a project organization, the lack of checks and balances, which may not be immediately recognized by a project team, may increase the discrepancies between different disciplines’ solutions over time: by the time the project team members recognize the problem, it may be too late to avoid redoing much of the work. Lack of communication and effective coordination between participants may lead to the magnification of small issues and create conflicts down the road. Long delays are sources of great instability in a system. When using balancing processes, it is often the case that organizations over-react when they finally perceive the effects of a delay, thus resulting in the creation of more environmental instability, rather than balance as intended. This is the case, for instance, when an organization overproduces an item in response to high market demand, while demand may have shifted by the time the product actually hits the market. Another example is when a project team implements rigid methods of control after project management has been identified as being too lax. Understanding how delays operate is very important in managing longterm performance. 3.4.2. Synergy between Competition and Cooperation The adaptive capacity of a system resides in its agents’ ability to simultaneously compete and cooperate. When competitive behaviors are not supplemented by cooperation, a 124 system will remain locked into win-lose or lose-lose situations. Some agents may do well for a while, until they get challenged to descend from their fitness peaks by other competing agents; or none of the agents will achieve their goals and thus, will remain at lower fitness peaks. Such situations will ruin the relationships among the agents of the system and will drive the system as a whole to a lower level of performance. In such an environment, the system will not be able to be creative and innovative. Similarly, a system in which agents only cooperate, continuously make concessions and accept onesided losses will only achieve sloppy performance, as all potential creative ideas will be immediately washed out because of the risk of creating conflicts among the agents. In contrast, competitive behaviors in a system that also exemplifies cooperative behaviors between its agents will challenge the status quo and will drive the system to the edge of chaos without disintegration. Competition, in an AEC project organization, is due to specialization and the heterogeneity of the team; it is inherent in a project context. Competition arises because of the interdependencies between different disciplinary activities and it is driven by professionals’ needs to achieve their disciplinary goals: architects are concerned with buildings’ functionality, articulation of spaces and aesthetics; engineers focus on the design of structurally sound and energy-efficient buildings; contractors must build to the architects’ specifications and within their clients’ budgets and schedules. As professionals try to maximize their own disciplinary performance within the constraints imposed by their peers, they may fight each others’ ideas and work. Consequently, due to the competition of ideas and of disciplinary goals, frictions and conflicts may arise, 125 which, if not properly managed, may ruin teamwork, disrupt processes and hamper progress. Conventional thinking is built on the idea that frictions and conflicts are always a nuisance. Therefore, AEC professionals have devised strategies to reduce, if not eliminate, conflicts. Partnering, for instance, is a procedure that has been developed as a dispute-avoidance and resolution method. Early in a project, a project team may participate in an informal, off-site retreat of one- to three-days’ duration, which is directed by a facilitator. The objectives of the retreat are to allow the participants to get to know each other; to create a convivial atmosphere between team members; to broaden participants’ provincial views; and finally, to help participants agree on commonly defined goals and procedures. Working relationships are carefully and deliberately built based on mutual respect, trust and integrity. Ideally, the team meets often during the project to continue the partnering process, particularly when issues need to be resolved. Competition, however, is not necessarily harmful or a hindrance to performance. In fact, it may be quite the opposite. In some specific conditions, competition may be very healthy and vital to a project team’s ability to innovate. In his analysis of the automobile design process at Nissan Design International (NDI), Jerry Hirshberg, founder and president of NDI, showed that “creative abrasion,” i.e., the competition between different design ideas provides the “creative energy” necessary to original thinking and the achievement of high quality design. Hirshberg (1998, p.33) recalls how he developed the concept of “creative abrasion” in the following terms: It [the concept of creative abrasion] evolved from an understanding of innovative thinking best described by the scientist/educator/writer Arthur Koestler as ‘the sudden interlocking of two previously unrelated skills or matrices of thought.’ And it grew out of an urgent need to deal with an atmosphere of colliding cultures 126 as well as the natural friction from an interaction of broadly varied personalities and viewpoints. Just as the creative fusion of ideas can occur by holding seemingly antithetical thoughts in the mind simultaneously, so creative collaboration between people can occur by an effort to retain conflicting cultural and disciplinary viewpoints in the mind without discarding or allowing either to dominate. It is therefore critical for professionals to hold on, to some extent, to their own disciplinary constraints, standards and interests in order to create some level of abrasion. The juxtaposition of divergent ideas stimulates individual and team creativity and helps a team to transcend the obvious solutions. Interestingly, competition between ideas encourages collaboration because it forces team members to work with one another to work out the issues. Creative abrasion also requires professionals to know when to hold onto their individual interests and when to give in: doing so requires much selfmanagement and leadership. Consequently, productive competition and creative abrasion must always occur within a cooperative environment. Cooperation is driven by mutual dependence and the need for professionals to share information and expertise. Cooperation creates a supportive teamwork culture that is based on mutual respect, trust and integrity. In an environment such as the AEC industry that has been based historically on adversarial relationships, trust will not exist per se. A project team may have to spend time building, and then managing trust relationships. Indeed, adaptive challenges always test the level of trustworthiness between partners. Conflicts, successfully managed, will create a reinforcing process: project team members’ levels of trustfulness will rise, in turn generating more trustworthy behaviors. Conversely, the defection of one participant may dramatically erode immediate and long-term relationships. 127 We should note that cooperative behaviors and the practice of reciprocity may not always originate from trust, but from the pursuit of self-interest instead, without the influence of external motivation mechanisms. Axelrod (1984) demonstrates, in a computerized version of the “Prisoner’s Dilemma” game, that the “shadow of the future,” i.e., whether agents in a system will have an opportunity to work with one another again in the future, influences their choice between cooperation and defection. The motivation, here, is the reputation that agents gain in cooperating or not, as there is a risk, after defection, of being ‘punished,’ not only by the agents in the system, but also by other agents outside the system, with whom they may have an opportunity to work in the future.6 Because an AEC project spans many years, professionals should have the time and incentive to work out their relationships, as well as develop trust and reciprocal behaviors. Moreover, an opportunity to work together on another project may create an incentive for participants to cooperate with one another. In local markets, professional reputation built on “word of mouth” may also influence behavior, since the opportunity for firms and professionals to encounter each other in another project may be higher than in a global market in which a firm may always work with different partners. The process of “coopetition,” i.e., the simultaneous competition and cooperation between professional firms in an AEC project organization, may be modeled using a fitness landscape in which professionals correspond to the agents in an ecosystem. As the earlier discussion about fitness landscapes has demonstrated, the performance of an agent in an ecosystem depends on the strategy and performance of the other agents in the ecosystem. Similarly, the individual performance of one professional discipline in a 128 project is influenced by the professional goals and design solutions of the other members of the project team. As a professional tries to maximize the performance of his/her own disciplinary solution, s/he creates constraints on the other team members, which force them to lower their level of performance (i.e., they may be forced to descend to a lower fitness peak). If professionals were only competing with one another to maximize their own design solutions, the entire project would become completely instable and chaotic as each professional would attempt to reach a higher fitness peak without considering the impact on the other disciplines’ performance and the performance of the overall project. Therefore, in order for each professional to improve his/her own disciplinary performance, s/he must collaborate with others and negotiate individual requirements and priorities against the requirements of the overall project. As pointed out by Harman and Sahtouris (1998), in a holarchy (or ecosystem), holons enjoy only relative autonomy because their autonomy is constrained by the next larger holons. As each holon pursues its own interest, it has to negotiate its self-interest with the overall interest of the entire holarchy: “How this endless negotiation of selfinterest functions at every level, or in each nested holon, can be seen as the search for dynamic balance or mutual consistency through negotiation between parts and wholes, or rather between holons and holarchies” (Harman & Sahtouris 1998, p.23). For instance, a system such as a couple is constituted by three sub-systems: the two spouses and the couple itself – an abstract, yet very real entity with its own personality. Each entity has its own identity, autonomy and integrity. As Harman and Sahtouris (1998, p.23) notes, “The integrity of each individual has to be endlessly negotiated with the integrity of the couple, seeking its own integrity or autonomy.” Similarly, in an AEC project 129 organization, each professional’s autonomy is constrained by the overall project organization. Professionals need to negotiate the integrity of their own disciplinary solutions, while maintaining a mutual consistency between the different disciplinary solutions. Creativity and innovation occur through the synergy between competition (creative abrasion) that ensures the integrity of each disciplinary solution and cooperative behavior that ensures the performance of the whole project. 3.4.3. Creative Tension between Dominant and Recessive Schemas An organization may be drawn to the edge of chaos by the tension between dominant schemas – that is, the rules that drive the performance of primary tasks and govern the current engagement of external reality – and the recessive schemas – that is, rules that govern play and challenge the dominant schemas (Stacey 1996). The dominant schemas support an organization’s legitimate system and define its formal culture. In a project organization, the dominant schemas consist of all the project management structures and activities that aim at defining goals; monitoring progress; facilitating work coordination; supporting checks and balances; maintaining professionals’ behaviors within the implicitly and explicitly accepted norms; and so on. Thus, the dominant schemas drive “serious” work and play a critical role in containing professionals’ anxiety. The dominant schemas maintain a project organization in the equilibrium zone by using negative feedback to manage any discrepancies between the performance of an activity at a given time and the pre-established measure of performance. However, the dominant schemas may also act as a driver of changes when an organization perceives that the changes are necessary to maintain the level of required performance. 130 The recessive schemas consist of “the mental contents of an organization’s members that are used not to engage current reality, but to engage in playful activities in the shadow system” (Stacey 1996, p.221). Recessive schemas are at the origin of professional jokes – “out-of-the-box” thinking; testing of unconventional ideas; and so on. They challenge an organization to move outside of its comfort zone and face uncertainties; however, recessive schemas may also be used to maintain the status quo in an organization, especially when they are applied to resist any change in the legitimate system. One of the main reasons for the recessive schemas to sabotage a process of change in the legitimate system is that change increases individuals’ levels of anxiety and this factor elicits defensive behaviors that attempt to maintain the status quo. When the shadow system supports the legitimate system, an organization will obviously occupy a stable zone. When the shadow system becomes so pervasive that it cannot be maintained by the legitimate system, an organization will disintegrate into a chaotic state. If, however, the shadow system comes to challenge the legitimate system and the legitimate system provides enough structure to hold the anxiety of an organization’s members, then the organization will move to the edge of chaos and will be able to perform double-loop learning. Thus, creativity takes place when the organization is able to break free from the rules, mental models and norms, which dictate how work ought to be done. It requires the destruction of some part of the dominant schemas, whether it is an idea, a value, a rule and so on. It is interesting to note the differences between what drives (serious) work activities and what drives play activities: • Work is about discipline. Play is about exploration. 131 • • Work is about achieving goals. Play is about the journey. Work is about performing within constraints. Play is about removing the constraints. • • Work is about following a plan. Play is about creating a vision. Work is about maintaining control. Play is about looking forward to being surprised by the unknown. • Work subdivides activities into tasks for increased local efficiency. Play integrates divergent ideas to spur innovation and global performance. • “Work tends to be a convergent activity, focusing in on the task at hand. Play is a divergent activity. It opens out and is not easy to contain” (Hirshberg 1998, p.116). • Work focuses on enforcing procedures. Play challenges the status quo. Ultimately, the dialectic between work and play activities is what allows an organization to innovate and be efficient at the same time. For instance, a project organization performing within this dialectic may be able to innovate and at the same time deliver to the client on budget and on schedule. Arie de Geus (1997, pp.63-66) argues that “to play is to learn” and that the playing that occurs in organizations is not that different from the way children play.7 Play is most often conducted with a “transitional object,” i.e., an object a player can manipulate and with which s/he can experiment. Similar to the way children use toys, professionals in a project organization also use transitional objects such as sketches, CAD drawings or computer-based models of the building they design, spreadsheets, scheduling tools, financial models and any other means to simulate reality. By playing with models, 132 professionals experiment with ideas, test out hypotheses and envision the future before a final solution is implemented in reality. Professionals also use models as “boundary objects,” i.e., objects that facilitate the autonomy of different disciplinary worlds, while simultaneously providing a bridge between different worlds (Leigh Star & Griesemer 1989).8 For instance, a 3D CAD model may be used by architects to visualize space and volume, while it may be used by engineers to test out the structural components of a building. When used by both architects and engineering simultaneously, the model becomes a shared space around which conversations take place and ideas and alternatives may be debated. Thus, the model is both an exploration tool, as well as a means to visualize constraints and manage expectations. Around the model, play enhances the acquisition of knowledge among individuals and groups (Tavistock Institute) and thus facilitates informed decision-making. Thus, the model is simultaneously a source of uncertainties (through the play activity) and a source of certainties (by facilitating crossdiscipline integration). The model operates, therefore, as a strange attractor. Organizational mental models, values and institutions influence the way an organization plays and goes about serious work. Organizational culture influences whether it is more important for the members of an organization to produce than to explore; more important to achieve the pre-established project goals than to spend time redefining what value actually means for a client; more important to deliver on schedule than to ensure that the product developed is not flawed and truly adds value; and so on. In his book, Serious Play, Michael Schrage (2000) analyzes the way companies use models and prototypes to innovate. As he notes, the culture of an organization shapes the prototypes and models an organization is able to create. In some organizations, the idea 133 that play can be an effective process for developing innovative products is foreign to conventional managers who have been educated to believe that predictability and control are essential to performance management. On the other hand, Schrage (2000) recognizes the peril of play when it is done in silo and ignores the constraints of project management. In architecture, this pitfall can be illustrated by the development of a beautiful design that cannot be built either because of its high cost; failure to acknowledge users’ requirements; or constructability issues. Once again, outside of the creative tension between the formal and informal culture and processes, along with the associated trade-offs between them, an organization cannot innovate. 3.4.4. Attractors’ Meaning As we have just discussed, the power of strange attractors resides in the tension between opposite forces, which maintain an organization at the edge of chaos. When one of the forces takes over, a project organization will either remain in a stable zone or will disintegrate into chaos. Project organizations that wish to deliver high-value facilities, while effectively responding to the budget and schedule constraints imposed by their clients, should be cognizant of this phenomenon and devise strategies to facilitate synergy between those opposing forces. Wheatley (1992) argues that, to be effective, a strange attractor must resonate within the individuals in an organization, i.e., it must be meaningful enough to the agents in an organization to attract them and make them receptive to change. Meaning is cocreated: it requires high interconnectivity between agents, as well as rich communication between them. Meaning provides a sense of purpose and a point of reference to which 134 individuals and organizations are willing to align their actions. Similarly, a project organization at the edge of chaos will be able to self-organize if its members are able to create meaning out of the tension created by strange attractors that challenge the stability of the environment. Only then can a project organization develop new structures, processes and modes of interaction, which enable its members to overcome the adaptive challenges they may face. 3.5. Testing out the Model This chapter has provided a theoretical framework within which to analyze the adaptive capacity of a project organization. In the next two chapters, this model will be applied and tested on two project case studies: the International Terminal at the Airport in San Francisco and the Martin Luther King, Jr. Library in San José. Chapters 4 and 5 present the two case studies respectively and analyze the adaptive capacity of each project. In particular, we consider whether the project organizations were able to influence the control parameters by modifying their organizational variables and thus, whether they were able to manage complexity at the edge of chaos. We also investigate what forces or strange attractors, if any, were present in each project to push and maintain the project organizations at the edge of chaos. 135 CHAPTER 4 International Terminal at the SFO Airport The new International Terminal at the San Francisco Airport (henceforth referred to as “Terminal”) was completed in December 1999 and opened its doors to the public, almost a year later, during the fall of 2000, after the airport staff and equipment had moved from the old terminal to the new facility. It is the centerpiece of the airport’s multi-billion dollar expansion program that started in 1989, more than 16 years ago. Some components of the program are still under construction today. The client is the San Francisco Airport Commission and the San Francisco International Airport Bureau of Design and Construction. The Terminal is a design by San Francisco-based architecture firm, Skidmore, Owings & Merrill L.L.P. (SOM), in a joint venture with architectural firms Del Campo & Maru and Michael Willis & Architects. The design was led by SOM' Craig Hartman, FAIA, with Navin Amin s (formerly of SOM) as chief structural engineer. The design was developed in a competition held by the San Francisco Airport Commission in 1993. Tutor-Saliba Corp., 136 in a joint venture with Perini Corp. and Buckely & Co. Inc, is the general contractor firm that won the contract through competitive bidding in 1996. The Terminal is 1.65 million square feet: 866 feet long by 280 feet wide, with a central clear span of 380 feet. The building is the largest terminal in North America based on square footage (it could host 35 football fields) and the largest project SOM had ever designed at the time. Its final cost was $495 million. The Terminal is considered the “Gateway to the West.” Its dramatic steel roof structure that resembles big wings to symbolize the metaphor of flight can be seen from approaching roadways and the air. The Terminal’s façade is a huge glass-and-metal curtain wall that screens the direct western sun and lets in natural light. The building is five stories tall and is built on a sizeconstrained constructible land. As such, the Terminal presents a model for other urban airports.1 The building houses 24 gates that can accommodate the largest aircrafts. Baggage handling capabilities and U.S. Customs facilities have been expanded from their original capacity in the old terminal to expedite passenger traffic. Ticket counter space is large to shorten check-in times for departing international travelers. (The Terminal can process 5,000 passengers every hour – five times more than the old terminal could handle.) In addition, the Terminal features an aviation museum, restaurants, an interfaith chapel and six works of art. This chapter analyzes the Terminal project from a living system’s perspective and assesses the adaptive capacity of the Terminal project team throughout the project life cycle. Section 1 introduces the Terminal project case study. Section 2 provides an overview of the sheer complexity of the project. Section 3 illustrates the influence of path dependence on the project’s adaptive capacity. Section 4 analyses the project team’s 137 ability to cope with changes and unforeseen events. Section 5 shows that the lack of adaptive capacity of the project organization, especially during the construction phase, was due to the inability of the project participants to create an integrated communication network. Section 6 argues that there should be a better way to manage complex and uncertain projects such as the Terminal. 4.1. Terminal Project Case Study Introduction This section first introduces the case study methodology. Next, we present the project timeline and the project organizational structure. Finally, a table of project specifics that describes people interviewed, their company and their role on the project is included. The reader may easily refer to this table throughout the reading of the case study. 4.1.1. Case Study Methodology The interviews for this case study took place over a few weeks during the fall of 1998. The individuals I interviewed belong to organizations that formed the core project team, i.e., the client, the architects, the construction manager and the general contractor. Participants were identified based on the critical role their firm played in the project; they were themselves directly involved in the project. Interviews were conducted at a place the participants suggested, usually their office. The interviews typically lasted between one hour and 90 minutes. I only met once with all the participants, except the executive architect at SOM, whom I interviewed on two occasions. At the time of the interviews, I had not yet developed the performance model based on living systems theory, as presented in Chapter 3. My interest, at the time, was 138 to understand the impact of organizational variables on the ability of a project team to collaborate. Consequently, I asked open-ended questions that were related to the aforementioned variables. Apart from recording the interviews were recorded, I also took some notes. I then transcribed the project participants’ narratives about the project context; organizational structure; communication and coordination processes; technologies; as well as the professionals’ culture and behavior. The transcripts of these interviews served as the document source of my analysis. As I considered whether to use this case study to analyze the adaptive capacity of a project, it soon seemed obvious that the project participants’ stories could be analyzed to understand the impact of the organizational variables on the control parameters defined in the previous chapter: the rate of information flow; agent’s schema diversity; richness of the connections; level of contained anxiety; and degree of power differential. Consequently, the interviews’ transcriptions were used to assess the adaptive capacity of the Terminal project team. 4.1.2. Project Timeline and Organizational Structure In order to help the reader follow the story of the design and construction of the Terminal project, this section provides a high-level project timeline (Figure 4.1). To project participants, this timeline might seem somewhat sketchy, since only the main phases and activities that took place during the project life cycle (that lasted for over 11 years) will be mentioned here. However, such a simplification is necessary to help the reader with no previous knowledge about this project to understand the main events that took place. I use the description of the project phases as a means to introduce the individuals I interviewed, their firms and their respective roles within the project’s organizational 139 structure. Of course, this project involved thousands of people including many design sub-consultants and sub-contractors. They cannot all be mentioned here, even though they may have played critical roles and/or may have had a great influence on the project outcome. The work on the program for the Terminal started in January 1989 and lasted over three years. The San Francisco Airport Commission and the San Francisco International Airport Bureau of Design and Construction worked collaboratively to define aircraft planning, passenger flow, peek loading of the facility and functional requirements. The program was mainly developed internally with minimum help from consultants. Peter Bittenbender, architect at the Bureau of Design and Construction, participated in the early planning phase and remained active for the entire duration of the project. As the planning phase got closer to completion, the airport reached out to the design community to determine the designers for the building. Based on the results of a very laborious process designed to screen out potential design firms (this process is presented in section 4.3.1), the client team developed a short list of potential design firms for the project. Among others, the short list included architectural firms SOM, Del Campo & Maru, and Michael Willis & Architects. The principals of these three firms agreed to form a design joint venture group in order to improve their chances of being awarded the project. Of the members of the design joint venture group, I had the opportunity to interview Keith Boswell, Project Manager at SOM; Martin Del Campo, Principal at Del Campo & Maru; and Michael Willis, Principal at Michael Willis & Architects. 140 In 1993, the client organized a design competition to determine the final winner between the design firms that had made the short list. The SOM/Del Campo/Michael Willis joint venture team, whose design was voted the best not only by the jury, but also by the other competitors as well, won the competition and was awarded the project in 1995. The design team could proceed with the design of the Terminal. While the design was still in progress, the client requested that the designers prepare the bid package for the piling so that construction could get started on a fast track. The piling contract was awarded in September 1995. The general contract was awarded a few months later, in September 1996, to Tutor-Saliba Corp. in a joint venture with Perini Corp. and Buckley & Co. Inc. Michael Kaechner, from Tutor-Saliba Corp., was assigned as the project manager for the Terminal.2 The construction management services for the Terminal were assigned to SFO Associates – a joint venture of Parsons, AGS, Inc., EPC Consultants, Inc., and Business Development, Inc. By the time of my interviews, three different construction managers who had worked on the project had been replaced (either because they did not fit the job or because they gained increased responsibilities). John Draguesku, from Parsons, got involved in the project in late 1997 and remained on the job until the completion of the project in December 1999. The Terminal opened its doors to the public during the fall of 2000. 141 1989 1990 SF Airport Commission and SF International Airport Bureau of Design and Construction start Terminal planning. P. Bittenbender gets involved. 1991 Client screens potential design firms’ candidates. 1992 Design joint venture is set up between SOM, Del Campo & Maru and Michael Willis & Architects. Short list of design firms is made public. 1993 Client organizes a design competition for the Terminal among 5 design firms. The client awards the job to the SOM/DelCampo & Maru/Michael Willis & Architects joint venture. Keith Boswell from SOM is the design project manager. 1994 Terminal design starts. 1995 1996 Piling contract is awarded and construction work starts. Tutor-Saliba is awarded the general contract. M. Kaechner is the GC project manager. 1997 John Draguesku from Parsons gets involved as CM (he is the third CM on the project). 1998 Global settlement of major changes. 1999 Construction is completed. 2000 The Terminal opens to the public. Figure 4.1: International Terminal Project Timeline 142 4.1.3. Project Specifics Participant Name Peter Bittenbender Keith Boswell Martin Del Campo Michael Willis Michael Kaechner John Draguesku Company Bureau of Design and Construction Skidmore, Owings & Merrill L.L.P Del Campo & Maru Michael Willis & Architects Tutor-Saliba Corp. SFO Associates/Parsons Role Project Manager – Client Representative Design Joint Venture Project Manager Principal Principal General Contractor Project Manager Construction Manager Table 4.1: Terminal Project Participants 4.2. Challenging Environment Architectural projects of a size and scope similar to the Terminal are a challenge, not only because of their technical complexity, but also, more significantly, because their design and construction life cycle spans many years (if not decades: the planning, design and construction life cycle of the Terminal was more than 11 years). The duration of a project, from planning to opening, is a critical factor that influences project performance. Project stakeholders’ representatives change over time, thus creating discontinuities. It is impossible for project participants to anticipate all the changes that may occur over the many years of project development. The Terminal was typical of large projects in the sense that the project participants had to cope with leadership and personnel changes, economic changes and program changes, which none of the individuals involved in the planning phase could have anticipated. For instance, because the project started in a time of economic 143 recession, the city’s Human Rights Program for Affirmative Action requested the airport to provide work to as many AEC firms as possible. To do so, the client subdivided the airport expansion program into many sub-contracts that were then awarded to different firms. While this decision certainly helped San Francisco / Bay Area firms to survive the economic recession, it also forced them, down the road, to coordinate with one another, since many sub-contracts were interrelated. Of course, all economic recessions eventually come to an end. The economic boom during the two years prior to the opening of the Terminal created a lot of challenges for the general contractor who needed to find skilled labor to work on the field. This was especially difficult because sub-contractors needed to work overtime to complete the project on schedule, while the construction field was very crowded because of the overlapping contracts and delayed work due to program changes. It was also impossible for the design team to anticipate changes in the price of steel. Because the structural concept for the Terminal was based on a huge amount of steel, the sharp rise in the steel prices required the design team to formulate a new and more economical solution that could be afforded by the client. Similarly, it was impossible to anticipate changes of leadership at key positions in project management positions and their consequent changes in business strategy and direction. When a new airport director came on board right in the middle of the project and construction was already well under way, he added million of dollars of changes to the program scope, which greatly impacted project performance. The Terminal project is a great example of the complexity of typical AEC projects today and the amount of uncertainties with which they have to deal. The 144 performance of such projects greatly depends on the teams’ ability to maintain themselves at the edge of chaos. Therefore, throughout the case study’s analysis, we assess the project team’s adaptive capabilities and leadership skills in dealing with complex situations and issues, which were beyond their direct circle of influence. As we shall discover, the Terminal project team had significant difficulties in influencing the control parameters defined in the previous chapter and thus, lacked the adaptive capacity to self-organize. 4.3. Path Dependence: Coping with Project History The Terminal project provides a dramatic example of the influence of frozen accidents, indeterminism and unforeseen events in directing the history of a complex project along a specific pathway. Many decisions made at different times by different parties based on their individual needs, objectives and project constraints added-up, over the long-term, to a nonlinear series of causes-and-effects that, combined with unforeseen events, had repercussions far beyond what the project team members could have ever anticipated and which contributed a great deal to the burden on the project team. These events locked the project in a particular evolution path, thus limiting the probability for alternative histories, i.e., they limited the number of options available to the project team to steer the project toward a high-level performance. This phenomenon is what is known in complexity theory as “frozen accidents.” As Pascale et al. (2000, p.163) explain, frozen accidents “are the coincidences that become locked in. Indeterminism deals the cards. Frozen accidents are what happen at the gaming table after the cards are dealt.”3 These accidents disrupted the stability of the project environment and required the team to adapt 145 and self-organize in order to minimize the impact of these accidents on the project performance. However, as we shall see, the project team was not always successful in coping with contingencies. 4.3.1. Uncertainty of a Convoluted Design Team Selection Process To recruit the design team for the Terminal, the client imagined a very unique but somewhat convoluted process whose analysis shows: (1) the complex interplay between diverse events and decisions in an ecosystem; (2) the limit of a system’s rational decision-making process that was shaped by changes in the environment and uncertainties; and (3) the need for the components in an ecosystem to continuously adapt to their environment to survive. 4.3.1.1. Preliminary Design Firms Selection Process To select the design firms for the diverse projects at the airport, the airport client imagined a lengthy and laborious process that required a staff of five persons working part-time over a period of eighteen months. First, the airport asked architectural offices from all over the world to submit their qualifications. The design firms were specifically asked not to send any marketing brochures; instead, they were instructed to provide the names of different jobs and clients for whom they had worked in the past. The airport client team intended to interview the clients of the design firms, instead of the design firms themselves, to assess the quality of services provided by the design firms in addition to their design capabilities. As Bittenbender pointed out: 146 We were trying to find a general consensus on whether a firm was good, or mediocre, or poor, in the way they [the firms] approached the client budget, the client program in a public building. The airport received more than 600 architectural firms’ replies. Based on the information provided, the client started a fantastic research effort to find the best architects. The Terminal had to be a very durable building with very high finishes and therefore, the client team searched through the design firms’ project portfolios for those projects that had similar requirements. It did not matter whether the design firms had past experience with airport design because they were not required to get involved in the development of the Terminal program, which would be given to them to follow. After a list of high-quality public projects was developed, the client team contacted by phone individuals such as the chiefs of maintenance of the buildings to interview them about their experiences. The interviews were based on a list of twenty questions such as: Did the designers follow the program that had been given to them? How did the architects treated the client, i.e., were they arrogant, or conversely, did they provide courteous and responsive service? Were the bids within budgets and, if not, how much were they over or under? How many change orders were initiated during the construction process and what was the dollar value of these change orders compared to the original bid price? Was it difficult to get the architects and consultants to come back to look after issues that may have developed after the building had opened? And so on and so forth. The answers received by the client team were truly amazing. While some people, for instance, were kind of neutral and sparing with information about their experiences, others, conversely, just “pull[ed] out their hearts,” especially if they had had a bad 147 experience. As Bittenbender recalled, “Finally, someone was asking how things were going in their buildings and they were getting very upset.” The feedback from the interviews revealed, not too surprisingly, that a large number of facilities, while wonderful architectural statements in themselves, had actually been delivered over budget. Indeed, one may expect those facilities to have been delivered with traditional design, construction and project management practices, which do not necessarily guarantee performance. Moreover, the clients interviewed may not have used target costing (Ballard 2000) during the project definition phase, either because they did not know how to implement such a process or because the designs of the facilities they were building were driven by prestige or style and their funding required them to be more elastic (if this were the case, however, it would not really be fair for the client teams to complain about the cost of their facilities).4 During the selection process, architectural firms whose facilities were criticized by the managers ended up at the bottom of the list, while those that had satisfied clients were placed at the top. The airport team used input from the survey to create a matrix that listed, for each job available at the airport, the list of architectural firms that had been selected to compete on the job. Most of the listed firms were from the Bay Area. Three large firms and three small offices ended up in the short list for the Terminal. The architectural firm selection process that took place at the airport was very unusual; the common practice is for clients to interview the architectural firms and/or organize a design competition. The airport client’s questionnaire was an attempt to assess the design firms based on their ability to align the design of the facilities to the programs; their behavior toward the clients; and their overall effectiveness in getting the 148 job done. However, one may question the usefulness of the feedback received. From a design-program alignment’s perspective, there might be many reasons why a facility does not satisfy its clients over time. For instance, the program may have changed since the delivery of the facility or the original program may not have been correctly defined in the first place. From a budget and schedule perspectives, one cannot solely place the blame on the designers. Indeed, many other variables could have also impacted the outcome of the project: the inefficiencies of construction processes; the inability of the general contractor to effectively coordinate construction work on the fields; the capabilities of the sub-contractors; scope changes requested from the client; or, finally, unforeseen events such as economic changes or bad weather. Considering the complexity of the SFO airport construction program, one of the most critical parameters influencing project performance is the adaptive capacity of a project team. Yet, few AEC professionals understand the importance of this parameter. In Chapter 6, we will discuss how a client, an AEC firm, or a project team can assess its adaptive capacity. Such an assessment, if it had been made at the Terminal project, might have helped the project team implement strategies that could have potentially improved its performance. 4.3.1.2. Different Perceptions of the Environment As we discussed in Chapter 2, structural changes in an autopoietic (social) system is triggered, but not determined by what is going on in the system’s environment: it is the system that determines what to notice or ignore in the environment. Once the system has defined the particular characteristics of its ecosystem and assessed how those 149 characteristics interact with the system’s internal organizational structure, it can then use that knowledge as a driver for its strategic decision-making processes and action. The design firms that were included in the Terminal client’s short list differed greatly in the way they perceived their business ecosystem; this difference of perception influenced their respective destinies. Del Campo mentioned he was very surprised to find out that his medium-size firm had been selected by the airport client; he did not believe, however, that his firm could compete against the larger firms with comparatively more substantial resources. Therefore, he immediately concluded that he needed to join forces with a larger firm and decided to discuss a potential collaboration strategy with the three large design firms that had also been selected for the design of the Terminal. In contrast, the other small design firms made the bold decision to compete on their own. The way the larger design offices such as SOM perceived their environment influenced their choice and business strategy with respect to their strategy for being awarded the job. In fact, SOM was the only large firm that responded favorably to Del Campo’s offer. Both Del Campo and Bittenbender acknowledged that SOM would have probably preferred to compete on its, own instead of forming a design joint venture. But the SOM management might have realized that the firm could increase its chances of winning the job by partnering with a minority firm. Indeed, when comparing design teams with one another, the Human Rights Commission automatically assigned five additional points to minority design teams – five points that made a big difference at evaluation time. The problem, however, was that in order to be considered a minority design team, the minority firm had to own the majority of the participation. As Del 150 Campo pointed out, considering SOM’s strong design ideas and demanding design standards, the firm must have been terrified at the idea of being controlled by another firm. While SOM recognized the political advantage of partnering with another firm, such an advantage was outweighed by the risk of losing its power differential. To eliminate such a risk, SOM decided to include another minority firm within the team and chose to partner with Michael Willis & Architects; by doing so, SOM subdivided the responsibility for the majority of the work between two minority firms: SOM remained the majority firm among the three, owning 49 percent of the work, while Del Campo & Maru and Michael Willis & Architects got 36 and 15 percent respectively. In order to select the final design firm or design joint venture candidate for the project, the client decided to interview the design teams that had been selected in the short list. However, the information gathered from the interviews might not have been sufficient for the client to differentiate the teams and make its choice. Although the design candidates were convinced they would be chosen based on portfolio merit and reputation, the airport unexpectedly announced that it was organizing a design competition for the Terminal. The five design teams that had made the short list were invited to compete. The Terminal competition was a very effective means to differentiate the design firms. Just before the jury panel announced the winner of the competition, each design team was asked to grade the others’ work and determine which, of all the designs, they considered the second best, after their own. Del Campo remembered that, as soon as he saw the other designs, he immediately knew that his team’s design was the best. Every other design team agreed and, in concurrence with the jury panel, the 151 SOM/Del Campo&Maru/Michael Willis & Architects joint venture was designated the winner of the competition. One may argue that, had SOM known the client would call for a design competition, the design firm might not have accepted the offer to create a join venture with two other firms and would have taken its chances on its own, trusting that its design skills and excellence would distinguish itself from the other potential candidates. However, SOM could not have foreseen that event and thus, the firm’s best rational business strategy at the given time was to build a joint venture. The effectiveness of this rational decision was, nevertheless, undermined by the complexity of the selection process. From a living systems’ perspective, the “survival” strategy of the Terminal project client (i.e., awarding the design to the best possible design team) competed, over time, with SOM’s strategy (i.e., being awarded the job): the client’s strategy forced SOM to migrate to a lower performance peak in the fitness landscape of the firm’s ecosystem. Consequently, SOM and the two minority firms had to make the best of the situation and, as we shall see in the next section, organize the joint venture for it to become a successful design team. 4.3.1.3. Integrating Three Agendas in One Single Design Team Once the design joint venture was formed, the three design firms had to make it work. Bittenbender acknowledged that SOM was the most control-oriented firm with whom he had ever worked: “They look at the building down to the last, small detail and they insist that their collaboration partners do the same.” Indeed, SOM’s high-level design standard 152 has earned the firm a reputation for design excellence in the architectural world. While SOM remained the majority firm in the joint venture, the firm, nevertheless, worried about the success of the collaboration. Due to the complexity of the Terminal project, the reconciliation of the different disciplines’ and firms’ goals was a challenge for the team. The design joint venture team had to agree on a common design strategy for the project and had to define the three firms’ respective responsibilities and collaboration processes. In addition, the scope of the project required the joint venture team to coordinate with hundreds of consultants, one of whom, the structural engineer, had a critical role in defining strategic design decisions. There have been rumors of fairly heated discussions between the architectural firms about the design philosophy in the early stage of the joint venture. For each partner, the critical question was how much of each of the respective firms’ design was going to be acknowledged. The outcome of these discussions was obvious: while the Terminal was the result of the efforts of the joint venture, the design itself was SOM’s. While SOM’s design philosophy was imposed on the other firms, there did not seem to be much disagreement about it, as both Del Campo & Maru and Michael Willis & Architects recognized SOM’s designers’ skills and capabilities and acknowledged the design quality of the building. One may also argue that, for both minority firms, participating in this project was a unique opportunity that would not have occurred without the existence of the joint venture team. The design team put a lot of effort into developing an organizational structure that was acceptable to each one of the parties and that created an environment conducive to 153 successful collaboration. The team agreed that while the three firms would participate in all the phases of the project, each would focus on different areas. SOM, because of its design skills and past experience in building large structures, would lead the design team and would be responsible for the overall design philosophy of the building. The firm would also be responsible for construction administration. Michael Willis & Architects, a much smaller firm with a very good design reputation nonetheless, would assist SOM in the design of some specific areas of the building. Del Campo & Maru, a larger and more experienced firm, could provide the technical expertise for creating the construction drawing packages and would share with SOM the responsibility for construction administration. Each firm, therefore, complemented the others. Thus, it appeared that the design joint venture was able to successfully balance power differentials between the three firms, thanks to the clear division of responsibilities, which leveraged each firm’s individual strength and facilitated decisionmaking processes with regard to the design of the facility. SOM successfully preserved its position of relative power and influence on the two other firms by building on the design firm’s excellent reputation among the architectural professions; its strong design skills; and its expertise in managing large and complex projects. These sources of power were fully recognized and accepted by the two other firms. As duties and responsibilities changed over time, some power and leadership might have shifted to the other firms, thus diminishing the power differential within the joint venture team. To simplify the amount of coordination required between three different offices, the joint venture team decided to rent an office space and co-locate the personnel involved in the design of the Terminal. Each firm provided personnel for the joint 154 venture in numbers proportional to the firms’ involvement. Del Campo remembered that, at some point during the design process, there were over 100 designers working on the project. We may assume that isolating the designers involved in the Terminal project from their own architectural offices and co-locating them into the same space helped to minimize the amount of heterogeneity and the cultural divergences within the design team, thus facilitating integration and collaboration. Since the Terminal design was, in large part, a structural engineering architecture, aligning engineering goals with architectural goals was critical to the success of the design. Because the Terminal needed to cross over a central roadway that provides the main access to the airport, the roof structure needed to span 380 feet: a real “tour-deforce,” as Del Campo acknowledged. SOM suggested using their own structural engineer, whom they considered excellent professionals. At first, Del Campo & Maru saw this as a potential conflict of interest. But after checking out the market and comparing fees between different structural engineering firms, they realized that SOM’s engineers’ fees were reasonable. From an architectural perspective, the driving design concept was that the building should be expressive of a certain structural idea that required the structural engineers to stand by the architects during the competition and throughout the design development. In the designers’ minds, there was no separation between the architectural idea and the structural engineering idea. As Michael Willis pointed out: …[I]t wasn’t the case about the architects drawing a picture and turning to the engineer and say: I want this to look very light; make it look good and I’ll edit it. The idea was: understanding that the engineering should be part of the architecture, what can we do to express the design of this building. 155 Even the first sketches had the engineering aspect expressed in them. There was a seamless dialogue between architecture and engineering. The engineers helped the architects to develop a structure that could be built (even at the competition phase) and the collaboration made the building better. There was complete alignment between the structural engineers’ work and the design intent for the building. As we shall see later, the integration between the engineering and architectural designs became very useful for minimizing the impact of rework as important structural changes had to be incorporated at a later phase in the project life cycle. 4.3.2. Limits of Coordination without Self-Organization The Terminal project was, but one – albeit the centerpiece – of numerous projects that constituted the airport extension program. Because of the economic depression at the early phases of the airport program planning process, the Human Rights Commission’s goal was to give out as much work as possible to different design firms. Consequently, the Airport Commission chose to subdivide the design of the terminal building into three major contracts: one for the Terminal (i.e., central building) and one for each of the adjacent wings hosting the boarding areas. The client, however, required that these three pieces had a uniform exterior aesthetic so that public perception at the entrance of the airport would be one of a single front door. To achieve this uniformity, collaboration was necessary between the Terminal joint venture and the two other architectural offices responsible for the adjacent wings (Hellmuth, Obata & Kassabaum, Inc. for Terminal G and Gerson Overstreet for Terminal A). While the design teams were free to express their architectural individuality inside 156 the buildings, they had to collaborate to define a single skin for the buildings. Once again, building on its power differential and the fact that the Terminal was the most important building at the airport, SOM took the lead in defining a grid for the buildings’ skin and in selecting the materials and coloration. By following SOM design guidelines, the design firms successfully created a single look for the buildings for which they were responsible. However, the construction details related to the way the skins were attached to the structure of the buildings were left to each one of the design firms. Ultimately, these construction details ended up being completely different from one another. As the general contractor Michael Kaechner noticed, boarding area A was the most economical, i.e., the quickest to install and the least expensive; the Terminal was in itself a unique system; and the G side was totally different. This, of course, made the construction process inefficient as Tutor-Saliba who was responsible for the construction of Terminal G, in addition to the International Terminal, had to manage different construction methods for an identical architectural result. (The contractor for Terminal A was Hensel Phelps Construction Co.) Moreover, because the Terminal could not really operate without the boarding areas and vice versa, the three buildings had to be delivered on the same schedule. This created a very tight functional interdependency between the three buildings and heightened the need to carefully coordinate work and schedules during construction. Onsite coordination, however, was particularly difficult because the bidding process for each building had been done on a different schedule and thus, each building was in a different phase of construction at any given time. For instance, when the structural steel was being erected for the boarding areas, the Terminal was only at the earthwork and footing phase; 157 or, when roofing was completed on one building, some of the footing and seismic isolators of another building had not even yet been installed. The Airport Commission’s decision to distribute the work to multiple firms – a decision, which, in and of itself, was an altruistic one – ultimately created dependencies and increased coordination needs between the firms involved. It seems that the impact of such decisions and the level of complexity and constraints created were not wellunderstood by the parties responsible for making the decisions at the time. Because of this reality, no planning was done to investigate different means through which coordination between firms could have been simplified. The design firms were left to figure out by themselves how to work with one another as necessity required; but those firms had been contracted for a very specific scope and they did not really try to reorganize their teams in ways that would minimize the overall cost of the construction work taking place at the airport. Construction costs might have been lower if the firms had worked with one another to define a single construction method for the building skins (for instance, by giving that responsibility to one single firm). But the client considered each project independently and awarded completely separate contracts; this strategy actually created duplication of work and waste.5 The project participants seemed to have lacked the adaptive capacity and creativity to self-organize, while dealing with such issues. First, the airport client did not fully assess the impact of subdividing the work among three design teams. Second, no one in the design teams took the initiative to discuss with the client the means to reduce the costs incurred by having three design firms designing three different skin details for 158 the same final design outcome and the potential benefits of re-designing an organizational and contractual structure that was better aligned with the project goals. The participants involved in the process lacked a conceptual framework to consider their own project as a living system embedded within a larger system: the airport’s overall program. While the client’s requirements forced them to consider the many interdependencies between individual projects, they, nevertheless, continued to think from a reductionist perspective, figuring out how to simply answer the requirements (i.e., a single aesthetic for the skins of the buildings) without taking a holistic approach to the problem (i.e., considering how the design and construction processes for the skins of the facilities needed to be structured to minimize waste and lower costs). It is clear that, because of the complexity of such a process, an attempt to reorganize work would have pushed the design teams and the client to the edge of chaos, at least for a while. But this would have also created an opportunity to add value to the client and to the overall project. However, the design teams did not try to self-organize. Instead, they remained within their own equilibrium states and continued to improve their own performance (i.e., local optimum) at the expense of the overall project performance (global optimum). The reason for such a lack of adaptive capacity may be explained by the influence of the project’s “systemic structures.” Systemic structures are concerned with the interrelationships between key project variables such as scope of work, contractual agreements, participants’ know-how and technical expertise and so on. As Senge (1990, p.45) points out, systemic structures “generate particular patterns of behavior.” Indeed, the design teams were bound to their contracts that limited the ease by which 159 organizational innovations could be undertaken. They may not have perceived any incentives for doing more than their contracted obligations or doing things differently. The coordination costs they may have incurred by redefining the terms of their contract might have been perceived as higher than the benefits generated from the change. Moreover, redefining the scope of the contracts with the different parties involved (i.e., giving more scope to some firms, while reducing the scope of work for others) would have directly impacted the firms’ profits – positively for some firms and negatively for others. Unless particular arrangements could have been made with the firm(s) whose scope of work would be reduced (e.g., a promise of future work by the client), such scope changes might not have been well-received by AEC professionals. Ideally, since the design of a uniform aesthetic exterior for the facilities was a known requirement, the client should have taken a holistic approach to structuring the work early, at the planning phase, when the airport was figuring out how to subdivide work among multiple firms. This would have eliminated the need to re-negotiate contractual agreements late in the process and facilitated the coordination between the design firms involved. Led by SOM, the Terminal design joint venture retained control over the other design teams by influencing the design of the buildings’ skin. In doing so, the design teams achieved the client’s goal, i.e., a uniform exterior aesthetic. However, this came at a high cost for the client and the process did not generate as much added value as it could have had for the costs incurred. 4.3.3. Discontinuity Due to Change of Leadership Starting in January 1989, the Airport Commission worked over three years on the development of the program for the Terminal, collaborating with federal agencies and 17 160 airlines, all of which had to sign off on the conceptual plan that laid out the footprint of the building. When the Airport Commission hired the design joint venture, the architects received a binder including the program and drawings for the building. The designers were not allowed to alter the program or change the relationship between spaces without client approval; they, nevertheless, had free rein to do whatever they wanted architecturally, as long as they remained within budget. However, the client added new requirements to the program over time, which, as we will see later on, had very important consequences for the project. The main reason for these additions was a change in the airport leadership. While the former airport director had focused mainly on issues related to airport operations, the new director focused his interest on business and airport revenue issues. Since the airport pays for everything in revenue bonds and does not use tax dollars, the new director took a lastminute look at the concessions and decided to nearly double the program assigned to them. The airport decided not to defer the incorporation of these new requirements and added them to the project scope. In addition, the airlines that are airport partners requested additional changes. As Bittenbender noted during the interview: Nobody is going to be in the driver’s seat long enough to control it [the program] until the building is done. These were all functional changes for the airlines; those related to concessions were all good things to do. Just, the timing was terrible…We paid twice what it should have cost under other circumstances. But we make so much money off the concessions that it’s…okay…You cannot even realistically budget contingencies to cover those events, because they are going to come but you cannot know how big they are going to be. We did not have a clue we would have to deal with $60 million dollars’ worth of changes for airlines and concessions at that point. There is no way you can plan for that. And nobody would approve a budget with a contingency to cover something like that [anyway]. 161 In a project lasting more than 11 years, changes that cannot be foreseen at the outset of the project will unavoidably occur and disturb the process. Different project teams may take different approaches when faced with contingencies. Some teams may submit to project inertia and attempt to continue the work as usual, while doing their best to incorporate the changes. Others may have the courage and strong adaptive leadership to shake the project inertia and re-evaluate the situation, spending some time assessing the effects of the changes on the overall project; re-defining overall project goals; and devising a new project management strategy that provides new solutions to achieve new goals.6 It seems that inertia was stronger than leadership at the Terminal project: the large number of changes requested by the client was incorporated into the design, while the project goals and management strategy were not re-evaluated and re-aligned with the client’s added scope. Consequently, many conflicts arose between stakeholders as discrepancies between the client’s expectations and the product that was being delivered increased. 4.3.4. Unsuccessful Fast Track Process Due to Late Design Changes For different operational and political reasons, the airport chose to take the risk to fasttrack the project. Indeed, the urgency of getting the building up and running as soon as possible was extreme, because the existing terminal was overloaded to the point that passengers had to sit on airplanes for as long as two hours before they could go into the customs area – a situation that was only deteriorating. Consequently, the director of the airport asked the Airport Commission to prepare the bid package for the piling contract, while the design of the building was not yet 162 complete. The urgency to move the construction process forward was such that the Airport Commission set a record of 22 days to bid, award and get the contract certified by the controller (a process that usually takes between 90 and 120 days). Thus, pile driving started, while the design team kept working on the remaining packages: foundations, general construction, baggage and a variety of other support packages – a total of seven different contracts. The client awarded the piling contract in September 1995, while the general contract was awarded in September 1996. Thanks to the decision to fast-track the building process, the project schedule could have gained a year, if it had not been for late program changes requested by the client – changes that jeopardized the team’s ability to maintain the schedule. Bittenbender acknowledged that if the Airport Commission had had the choice, it would have much preferred not to fast-track the building. A significant amount of money could have been saved if the design joint venture had had more time to complete the design and to look at each part of the building in a more coordinated way before the bid; but, as Bittenbender noted, “If you choose to fast track it [the building], you can’t fault them [the designers] for a good percentage of the problems that come out of that decision.” While fast-tracking is effective in the context of a simple building, a cookie-cutter design, or in a high-rise where each floor is the same as the next one, it is much more difficult to manage in a project as complex, unique and extraordinary as the Terminal. For the Airport Commission, it was all a matter of choice and trade-offs; but fasttracking, combined with the client’s inability to avoid continually making changes to the program, created complex problems among the client, the general contractor, the 163 construction manager and the architects – in particular, problems related to the respective responsibilities due to conflicting goals and decisions. Because the completion date of the Terminal could not be missed, fast-tracking imposed a great burden on the general contractor and on the sub-contractors who had, for example, to work overtime. This issue was exacerbated by the difficulty of finding the necessary manpower. The construction market was booming during the Terminal construction and thus, there was a limited pool of workers, field supervisors and equipment upon which to draw. As Kaechner explained, in the best situation, fasttracking would require the general contractor to hire 400 electricians working for about a month and then, to drop that number to 250 electricians working an additional two months. In a booming market, this plan is completely unrealistic. Instead, 250 electricians were asked to work overtime over five months – a plan that is very expensive. The cost incurred due to overtime work was even higher when one considered the lowered productivity in the field due to the stacking of trades on a very congested construction site. 4.3.5. Reactive Budget Management Strategy Money is always a subject of contention in a large and complex project as firms struggle to deliver a building on budget, while trying to make a profit. The Terminal project team seems to have managed budget issues reactively and struggled with that issue all along the project life cycle. The team faced at least two main budget-related issues during the design phases and numerous issues during construction. During the design phase, the first issue arose after the joint venture team won the competition and realized that the 164 building they had designed would cost twice what the client had budgeted for it; the second issue arose later on as the designed team argued with the client about their fees. Prior to the competition phase, the client had approved a budget of $240 million for the Terminal. During the competition, cost was not the main concern for the design joint venture team whose focus was on designing a state-of-the-art Terminal and on winning the job. Moreover, the cost estimates based on conceptual drawings are only approximate. One may even hypothesize that, even though the architects knew that the facility they were designing was over the budget, they may have simply ignored the fact during the early project phase and hoped for the client to raise more funds. Indeed, since the designers’ fee is a percentage of a project’s total installed cost, they have no incentive to reduce the scope of the project at the conceptual phase, especially when the design is driven by prestige and the client’s desire to build a landmark facility. Consequently, once awarded the job and working on schematic design, the architects had to face reality and acknowledge that the cost of the facility they had designed was closer to over $400 million. Unfortunately, the city did not have that kind of budget available. While the design team convinced the client to approve more funding, the design joint venture team, nevertheless, had to figure out how to reduce the cost of the building, without jeopardizing the design intent. The designers used value engineering to compare various design options; they reduced the size of the roof by making the overhang less dramatic on the back of the building; and they negotiated a reduction of the program with the client, who agreed to give up some office space so that one floor of the building could be eliminated. 165 This reduction of the program may sound ironic when one considers the huge program changes requested by the client, later on. However, the fact that the joint venture team’s design was 67 percent above the budget seems to demonstrate the group’s failure to successfully integrate the design aesthetic, program and budget within the final solution. As noted above, such integration was not really the goal of the design team during the competition phase. Consequently, many tradeoffs between design, program and budget had to be made during the design drawing development phase, as the designers became increasingly aware of the real cost of the facility. The second point of contention came later, at the construction-drawing phase, and revolved around design fees. At the time of my interviews (Fall 1998), Del Campo told me that the project team was going through a “good natured fight” with the client. While the design joint venture team had already earned close to $15 million, the city told the architects that they should stop charging for the remaining hours they had to spend completing the drawings. The joint venture team agreed to reduce its fees and do “house keeping” type of work on a break-even basis. The architects refused, however, to deliver value-added work on such a basis. As Del Campo mentioned, in addition to their reputation for doing quality design work, SOM also has the reputation for not being underpaid for the work they do. SOM used its reputation to successfully influence the negotiation of the fees with the client. The architects were also fair to their sub-consultants. When some design consultants told the joint venture team that their firms had not quoted enough for their jobs and they needed to receive more money in order to complete their work, the design team agreed to accept their requests, providing more money, instead of finishing the jobs 166 themselves. In doing so, the design joint venture ensured that good relationships and trust would be maintained between the firms, thus facilitating collaboration between them. In multi-million dollar construction projects, the budget is always difficult to manage and the divergence of financial goals between clients and contractors is not new. The Terminal project was very typical in this regard, as discussions around costs occurred on an almost daily basis during construction, mainly because of the impact of the fasttrack process and late program changes requested by the client. Kaechner, the general contractor, was not shy about clearly stating what Tutor-Saliba’s goal was in getting involved in the project. As he pointed out: “We [Tutor-Saliba] are here to make a profit and…this is the reason [why] we bid the job.” While building a facility of such a scope and high visibility was certainly a satisfaction for the firm, such a reward was, nevertheless, secondary to making a profit. The client was not duped in that regard. Bittenbender acknowledged that, “The contractor is out there to make a profit and does not necessarily care about our budget. That’s a fact! So trying to control that on a daily basis is difficult.” The contractor attitude toward the budget certainly did not support the creation of a sense of team and the cohesion between the participants. Such a clash often occurs when clients choose the general contractor firm based on the lower bid price with no consideration of the firm culture, value and behavior toward the client. (As we shall see in the next case study, not all contractors have such an aggressive attitude toward financial issues; a softer approach may, in fact, provide higher value to all the participants, including the contractor.) 167 The issues around the budget were exacerbated because of the client’s request for many design changes after the bid – changes that added roughly 40 percent to the scope of the project. The outcomes were quite significant and placed a great burden on the project team. The changes, as we shall discuss in the next sections, required a huge amount of site coordination between the different parties (contractors, construction managers, architects and client) and cost the client a significant sum of money. 4.3.6. Unforeseen Consequences of Decisions Made The Terminal project clearly demonstrates how decisions made by different parties based on their individual needs and objectives added up, over the long-term, to a nonlinear series of causes and effects, which created consequences that contributed a great deal to the burden placed on the project team. It is not easy, in retrospect, to figure out how the client could have anticipated the effect of the early decisions made and thus, planned for them. Perhaps the program could have been better defined if more time had been spent working with the airlines and concessions. Perhaps the client should have authorized more project contingencies. Perhaps more up-front coordination with each entity (not only between the firms participating in the Terminal project, but also between all the other constituencies working on different projects at the airport) may have facilitated the creation of a common focus and a better-planned process. While we could go on building on such hypotheses, the fact remains that the airport could not have anticipated a change in directorship and with that, a change in the strategic direction of airport management as well. Organizational components (i.e., who is in charge; who defines the strategy; and what is the responsible person’s area of influence) are very difficult to control in a 168 project. Change of management often means changes in direction and goals – and this often creates chaos. The particular way a project team deals with unforeseen events demonstrates the team’s ability to cope with contingencies. How the Terminal project team managed contingencies is the focus of the next section. 4.4. Coping with Changes and Unforeseen Events The Terminal project has been influenced by multiple decisions and events, which increased the complexity of the project as the process moved forward. In addition to changes in scope originating with the client organization, unexpected and noncontrollable external constraints such as steel price increases and the El Niño weather phenomenon also influenced the project. This section reviews the processes used by the project team to cope with these issues as they arose. 4.4.1. Coping with Economic Changes Throughout the Terminal design process, the price of steel rocketed to the point that the design team realized that the client would not be able to afford a building whose structure was based on a huge amount of steel. The main issue for the designers was to figure out how to remove some components of the steel structure, while still preserving the design concept that had won the design competition. There was no solution for the design team other than to go back to the drawing board. The challenge was mainly one of engineering. Fortunately, since structural engineers had been involved from day one in the conception of the building, they were 169 fully aware of the design intent. Their task, therefore, was to redesign a structural engineering solution that preserved the original architectural idea. The solution was found to replace the foundation design by a construction method called base isolation, which isolates a building from the ground in case of earthquake and lets the building move separately from the ground. At the time, the Terminal was the largest project in the U.S. using base isolation. This solution helped to reduce the cost of the building, while allowing the design concept to survive. Thanks to the integration between the engineering and architectural design subsystems, the design joint venture was able to develop an innovative structural solution that minimized the impact of the changes on the project performance. 4.4.2. Coping with Mother Nature Mother Nature is not always on the side of construction projects: the Terminal is a case in point. As El Niño was expected to rage during the winter of 1998, the client asked the general contractor to work selective overtime and accelerate the construction of the curtain wall so that the building could be enclosed; thus, work on the interior finishes could be started in the eventuality of bad weather. That request alone cost the client an additional $4 million. The curtain wall contractor agreed to take on the risks and work selective overtime on the basis that, if they made the December deadline, they would get their acceleration fee. While this was a huge risk, the incentive was strong enough to push the curtain wall contractor to work to the limit of its capacity. While this was an expensive solution, it nevertheless helped to minimize the impact on the project schedule, which was the critical goal for the airport client. 170 4.4.3. Coping with Fast-Tracking and Program Changes Due to fast-tracking and late program changes introduced by the client, construction drawings were as not coordinated as they should have been. For instance, as Kaechner pointed out, many design details were shown differently on different drawing sheets and some drawings referenced the wrong details. Early in the construction process, the client authorized the architects to issue new drawings for every change, thus providing the general contractor with a continuous supply of clean, up-to-date drawings. However, because of the frequency of the changes made and the associated cost of continuously updating the drawings, the client eventually asked the architects to stop issuing new drawings for each change. Consequently, the general contractor was forced to rely on paper-based Requests For Information (RFIs) and sketches. Considering the number of sub-contractors involved in the project and the scope of their responsibility on site (there were 70 to 80 major sub-contractors and vendors with contract budgets varying from $5 or $10 million to $15 or $20 million; a couple of contractors had contracts valued at $30 million), relying on RFIs was very cumbersome. Kaechner pointed out that sub-contractors were asking from ten to twenty questions daily about design details. Some of these questions could be addressed and resolved fairly quickly with the designers; others involved design changes that had cascading repercussions on other building details and spaces. Keeping track of all the changes was very difficult both for the architects and the general contractors. To give an idea of the amount of information to manage, Kaechner indicated that at the time of my interview with him, he had sent 3,500 RFIs – 25 percent of which had already been assigned a revision number because the first answer from the architects had 171 been incomplete. (This meant that he was managing around 5,000 different RFI forms.) In addition, there were 500 field change revisions and 400 pending change orders (each one including multiple design changes). There was no other means to manage such a monumental volume of design changes and RFIs than to coordinate work, on a daily basis, in face-to-face meetings with the architects, the construction manager the client and the sub-contractors in order to resolve issues and orchestrate trades on the field. To further complicate the process, it was critical to maintain airport accessibility during construction – a constraint that required extensive daily coordination with the airport staff. Meetings were held every day at 7:00 am between the contractors and construction managers responsible for each section, as well as with airport staff representing both the air side and the land side of the airport to discuss what could be accomplished that particular day and who would do what as a priority. Daily field coordination was an attempt to increase the information flow among project participants and to facilitate a quick feedback loop between the time an issue was discovered and its resolution on the field. Nevertheless, the daily coordination of field work was very difficult. The construction schedule was so dynamic that the client admitted that formal scheduling methods were actually not useful and in some cases, even got in the way of making progress (this seems to confirm the limits of conventional project management methods in the face of the dynamic nature of the project). However, not being able to maintain a construction schedule meant that there was no clear work flow and that work was “hop-scotching based on where work [could] be done.”7 Activities on the field were therefore very chaotic as no process was in place to counterbalance the amount of uncertainties faced by the participants. 172 It seemed that discussions around construction issues went on and on, leaving the project team with a lack of clarity on whether a particular issue was indeed resolved. The general contractor specifically complained about the indecisiveness of the construction manager; his inability to resolve issues in a timely fashion; and the slow processing of field change orders in general. Interestingly, someone pointed out, in a partnering survey form, that it appeared that problems were resolved more quickly in times of crisis than under normal circumstances. This may be because a field crisis spurred the adaptive creativity of the team, forcing those individuals working on the field to specifically frame the problem and focus on the particular issue that had to be resolved to the exclusion of all else. Conversely, there seems to have been a low level of cooperation between the management of the main firms involved – especially between the general contractor and the construction manager. This seems to indicate that, in this case, individuals at the bottom of the organizational structure were more willing to cooperate than the management of the firms for which they worked. As we discussed in Chapter 3, firms are controlled by their institutional framework and legitimate system which, in this case, clashed with one another. Moreover, the lack of cooperation and trust seemed to have been amplified by the differences in the personalities and behaviors of the general contractor and the construction manager, as well as the two individuals’ inability to reconcile their differences. (This is discussed in mode detail in Section 4.5.) The already complex coordination process between field trades was further complicated by the fact that, because the construction market was booming at the time, the contractor had difficulties finding qualified manpower. To make things even worse, the San Francisco Airport Commission had to formally stipulate how many hours per 173 week each trade was authorized to work on the project beyond the agreed maximum of six to eight hours of work per day or a total of 48 hours per week. The time the Airport Commission took to providing letters of authorization, which created additional schedule constraints for field coordination. Working in such a chaotic environment, participant dissatisfaction was high. The architects complained that things were built incorrectly and that the owner did not get quality construction. In addition, they felt that the contractor took advantage of the aggressive schedule to request acceptance for variances in the ways the work was done. The general contractor complained that sub-contractors were not paid on time, which decreased their willingness to perform the work as required. In addition, the contractor accused the owner’s inspectors of not using the correct drawings with the latest field change orders.8 The accumulation of errors of different types certainly made an accurate assessment of the quality of work very difficult. The combination of schedule pressures, limited manpower, difficult work conditions and low morale for some trades had consequences on the quality of work. The issues around safety were even more troublesome. Security was the subject of concern for the project team that recorded some minor incidents. The pressure on the schedule forced multiple trades to overlap their activities on a very congested site, thus creating greater safety risk for workers. Even though the general contractor provided all subcontractors with safety training, it seemed that the on-site situation was more reactive than preventive. As the general contractor pointed out: “[P]roductivity, efficiency, safety, quality and morale aren’t just numbers; rather, they are all intertwined.” Indeed, the Terminal project fell into a vicious cycle in which poor performance undermined the 174 morale of the workers and project team, which influenced the quality of work that in turn affected the confrontational behavior of the participants. The project team was under the influence of unmatched positive feedback processes that exacerbated issues, while participants were struggling with the different aspects and constraints of the project. As a result, the team had much difficulty reversing this vicious cycle and maintaining a somewhat positive balance between productivity, quality, cost and safety. Managing the budget under these conditions was not a simple matter. To avoid going to court and incurring additional legal expense at the end of the project, the client made the business decision to use partnering to arrive at a global settlement with the general contractor on the $60 million of changes that were added to the project scope. Bittenbender admitted that he had never been involved in a project of this magnitude. Because of the large amount of money involved, the client felt that a global settlement was a much better solution than continuing to negotiate the cost for each individual scope change. This formal negotiation process was a means to avoid the day-to-day battles due to the inability of the project team to manage complexity. 4.4.4. Difficult Adaptation The Terminal project team was not always successful in coping with environmental changes and unforeseen events. The poor relationship between team members was an important variable in influencing the lack of adaptive capacity. This factor prevented the project team from maintaining the control parameters within the level required for an organization to self-organize and enabling the team to operate as an integrated network of communication. 175 4.5. Unsuccessful Integration of the Project Team This section discusses the issues around the professionals’ culture and behavior, as well as their influence on the team members’ ability to effectively communicate and cooperate with one another. In a project as large, complex and highly visible as the Terminal, personality and ego can certainly get in the way of making steady progress. Many participants acknowledged that relationships between some individuals were difficult and that ego issues had sometimes been hard to control. Day after day, participants struggled to improve relationships and the quality of interactions between members of the team. Trust and cooperation between firms was very hard to achieve. Interestingly, the component of trust was never mentioned by anyone I interviewed except Michael Willis, who spoke about it in general terms, not specifically applied to the context of the Terminal project. While I do not have much information about the quality of the relationships between the three design firms, my perception during the interviews was that there had not been major conflicts within the joint venture team. Conversely, there might have been some ego management issues between the architects and the client, since, as I questioned Bittenbender about this issue, he replied that, while “it [ego] had been a matter of criticism,” he did not want to elaborate further. He did acknowledge, however, that: [B]y nature of training, architects need to realize that if they don’t take control of that [ego], this is going to control them… The last thing the client wants to deal with…on the architectural side of things is someone’s ego. As Bittenbender explained, architects should remain open to their client’s needs and demonstrate concern by fully considering the available budget and respecting the 176 project schedule as much as possible. This kind of service seemed to have been delivered by the design firms in the Terminal project. Good participant relationships appeared to have been more problematic during construction even though quarterly partnering sessions with an external facilitator were in place to help resolve conflicts between the parties.9 The partnering sessions were not all successful. While it was fairly easy for team members to agree on the ultimate goals of the project (e.g., the timely completion of the project on budget), it was very difficult for them to agree on how this would be done and at what cost. Participants’ feelings about the benefits of the partnering sessions differed greatly. While the sessions had their successes and failures, Bittenbender still believed that partnering was a great way to solve problems. For Kaechner, it was an opportunity to meet the team players; define who was in charge; and agree on the goals for the project. It was also a way to bring some level of civility so that daily fights could be avoided. However, John Draguesku had a very different take on the partnering process. To Draguesku, partnering is something that project participants should set up internally among themselves. In contrast, the facilitated sessions seemed to him to have been super-imposed on the project team. Because of this, Draguesku believed the sessions were not as productive and genuine as other methods he had used in other projects where the teams worked on building mutual goals and therefore, developed a sense of teamwork on their own. These previous experiences represented a truly collaborative effort among the participants. As Draguesku pointed out: To me, it seems that a person [the facilitator] that comes in every three months or four months and is involved in so many other projects is sort of…removed from what’s going on here…and, sure, he can look at his folder for what the project is and listen to what people are saying, but to me, there is something that is sort of 177 disingenuous about his real concern for the project. To me, this has to be coming from the players who are involved in the project…To me, cooperation and mutual goals [are] not things that need to be facilitated at all. To me, there is an essential problem if there is a thought that there needs to be some facilitation. It is important to recall that Draguesku got involved late in the process, being the third construction manager for the project.10 He mentioned that when he first participated in the project, he was quite surprised to see how bad the relationships were at that point in time. Not surprisingly, the relationship between Draguesku and the general contractor Kaechner was very difficult. Kaechner acknowledged that both of them were still working out their relationship. Indeed, as he joined the project, Draguesku became more forceful than his predecessors had been with respect to enforcing the contracts. Because of differing goals and objectives related to the budget, conflicts emerged with the contractor. Draguesku acknowledged: I was the bad guy because….they [the general contractor team members] were not getting anything their way anymore, anything was not free anymore at their request…so there was a real tightening up of the relationship here, because of…my involvement…And not that there was ever any intent of being the bad guy. My intent, regardless of how people are taking it, is one thing: it is to finish the contract on time and on budget. It will be just like the other ones [projects] I have done. It would be very surprising to me if this one doesn’t [finish on time and on budget]; it would be the first time ever. Changes of perception from each party over time were made very explicit in results from the surveys that were sent to the participants prior to each one of the partnering sessions. At the session scheduled a short time after Draguesku got involved in the project, results of the survey showed improved satisfaction from the client and the designers and a lower level of satisfaction from the contractor. Draguesku explained that: The reason was that they were having it their way…here you see: designers finally are saying: My god! There is somebody there that is enforcing the contracts and the documents. The owner is saying: Gee, that’s great! We are 178 getting better…we were in a real dip back there…The contractor is saying: This is the worst thing that ever happened to us! At the time of my interview, i.e., approximately one year after John Draguesku got involved in the project, satisfaction on the part of the project team was improving slightly. The feedback reported in the evaluation form of June 1998 is revealing of the project team issues. In a transmittal to the team members, the client acknowledged that “[t]he primary cause for many of the concerns can be traced to the extensive changes to the project, the time it has taken to resolve them, and the stress this process has imposed on the team.”11 The client went on: “The major changes such as the BART Link and Concession and the Airline revisions represent about 20 percent of the extra work and about 80 percent of the cost. The many minor changes represent about 80 percent of the extra work and about 20 percent of the cost.” This situation understandingly created an environment that was prone to conflict. Methods of communication between team members greatly influenced levels of cooperation and trust. For instance, the client and construction manager jointly stated in the survey form that communication was still an issue, with some individuals not listening and tending only to hear what they wanted to hear. Outrageous and threatening letters were often sent between parties, which of course impacted the amount of trust between firms and individuals. On his side, the general contractor Kaechner stated that when communicating with the construction manager, both verbally and by letter, the contractor received “indirect code language responses” that required “interpretation” from the contractor side.12 The inability of some team members to “listen to” their teammates during meetings, their inability to “allow others to finish speaking,” as well as their use of 179 “verbal abuse” against some individuals, reveals a profound lack of communication skills; and the inability of some individuals to manage themselves and their egos. Such a lack of awareness and capability, as the architect Michael Willis explained, was an impediment to the swift resolution of issues. By refusing to acknowledge others’ opinions, an individual becomes more than a roadblock; he may actually jeopardize the team’s ability to discover a solution that may add value to the overall project. Such a behavior is counter-productive; furthermore, it impedes adaptation and self-organization. As a consequence of the aforementioned circumstances, the level of trust and cooperation among project administrators was perceived to be rather low. As mentioned before, cooperation seemed to have been better in the field than between administrative offices most likely because resolving field issues could not have simply been achieved with endless negotiations and cumbersome paperwork. Field issues that were more straightforward and required immediate actions were resolved via teamwork. In fact, teamwork seemed to have occurred at the lower levels of each organization, as individuals learned how to deal directly with each other, thus developing trust on a person-to-person basis. In contrast, in the management offices, the contractor criticized the construction manager’s indecisiveness, while the construction manager considered the general contractor to be inflexible. Similarly, cooperation between the owner’s and general contractor’s organizations was difficult because, after having spent so much effort reaching settlements about costs without either being fully happy with the final decision, both entities felt that they had been exploited by the other. The feeling of “losing the game” might have cemented the animosity between the two organizations. 180 Other issues that might have impacted the relationship between the general contractor and the construction manager were related to different management styles and focus; as well as the differences in the ways the two individuals valued and used knowledge. Kaechner, in particular, criticized the construction manager’s apparent lack of knowledge of what was going on in the field. As Kaechner pointed out: I am used to people having more knowledge and having their hands dirty. I am not afraid to go to the field and help…I don’t have time for it; I don’t do it anymore; I don’t run any equipment; I don’t build it; but I know what I am talking about. For Kaechner, getting “his hands dirty” by handling issues in the field is what brings value to the project. In contrast, Draguesku seemed to manage “from the office,” with a clear focus on enforcing the contract. These two different approaches likely aggravated the relationship between the two individuals. Kaechner admitted that he had fought all his life to gain more experience, especially “hands-on experience,” which he seemed to value the most. He acknowledged that even within his organization at Tudor Saliba, his management had some doubts about whether he had enough experience to take on this huge and challenging job. As he pointed out, experience takes time to build and one never seems to have enough of it. He also acknowledged feeling frustrated at not seeing enough people in the Terminal project with both experience in resolving technical issues and, at the same time, the ability to see the whole picture. In contrast to his difficult relationship with Draguesku, Kaechner explained how much he enjoyed working with Keith Boswell, the executive architect at SOM: 181 Keith is good on specifics and we like to deal with each other on specifics. He is very gifted in the knowledge of his [profession] as [compared to] engineers or architects I have been dealing with in the past. That’s a definite pleasure. Having more people with more experience is what is lacking here…I invited him, I forced him into meetings, everything from structural to architectural to painting, finishes, signage, just because I know that he is more focused and he is more experienced in what we are actually talking about, whether it be skylight or curtain wall. He may in fact have the engineers in his office more focused on that package or that submittal or that issue or that item…but he is the one who will actually bring in reality or the experience of the overall plan or the overall project. I find that much more conducive to actually getting the job done. Kaechner emphasized that, in general, projects needed more architects and engineers with field experience. He pointed out that he had taken an engineer out to a fabrication shop and the engineer had had no clue about what was going on; yet it was he who had designed and drawn the piece that was being built in front of his eyes. If the engineer did not understand how the shop was putting the piece together, how could he understand what he had drawn? The value that Kaechner associated with professional knowledge relates to handson experience and practice: it is “field knowledge.” Such practical knowledge contrasts with the more theoretical knowledge of an engineer who works from the office. Kaecher’s personal understanding of what practice is influenced his behavior, especially when confronted with individuals whose knowledge type did not match his own ideal and norm. That clash prevented effective communication and the development of trust between Kaechner and other professionals such as the construction manager and some engineers who were operating from different cognitive perspectives. While such divergences were not so critical in “normal” times, they contributed to the communication gap during times when the team had to resolve conflicts. 182 The failure of the Terminal project team members to operate as an integrated communication network hindered the team’s ability to openly share information essential to developing a common understanding of the issues to solve. Poor communication, in turn, hindered the team’s ability to resolve issues in a constructive way. Issues were solely resolved through formal negotiations as opposed to reciprocity. The lack of cooperation and trust forced the participants to fall back into relative isolation, thereby missing out on opportunities for developing solutions that may have contributed value to the project. Each party was trying to resolve issues from its own perspective, ignoring the overall impact their disciplines’ solutions could have on the other project components. The lack of trust became self-perpetuating and this vicious cycle only made things worse. The lack of reference to the overall project goals impeded integration and alignment. During the construction phase, the Terminal project team was a dysfunctional living system, unable to operate as a unified whole. 4.6. Missed Opportunities? The Terminal now stands in all its glory. Every year, millions of international travelers enjoy their time spent waiting for their turn at the check-in counters by admiring the light and space created by the facility’s gigantic roof structure. Very few of them, with the exception perhaps of architects and other construction-savvy individuals, are aware of the amount of effort and brain power it took to erect this elegant giant. Very few know about the challenges that had to be overcome by the project team to complete the work. For those who were part of the process, the memory of many years of hard work at the airport may provide the basis of another professional experience. At the end of the 183 project, while claims and other budget issues were being settled, perhaps some of the members of the project team thoughtfully asked themselves: Could we have done it differently? Could we have created a better “project atmosphere?” Could higher trust among the participants have facilitated the negotiation processes? Could conflicts have been resolved in a different way? Could less struggle and pain have been generated? Could we have done it for less money? Or, could we have done more for less money? Did we generate high enough value for the client? Would the client hire us again if they had to start it all over? Have we used this experience to build strong, long-term, professional relationships that may be leveraged on other project opportunities? Are we proud of ourselves, professionally, as well as personally? Our theoretical framework based on the living systems theory is driving us to argue that, indeed, there is a better way to manage a project confronted with such adaptive challenges and deal with uncertainties. Professional behaviors based on trust and mutual respect can facilitate adaptation and self-organization. Cooperation among firms and individuals is critical when operating in complex environments, even though much is at stake for each professional entity. Value can be generated by AEC professionals if they are willing to challenge traditional professional models that limit them within the boundaries of old and obsolete professional rules. Finally, professional firms involved in a project can make a fair profit that does not come at the expense of other participants. By illustrating the use of a holistic approach to project management and cooperative behaviors, the next case study presents an alternate approach to the way a project team may manage complexity and adapt to uncertainties. 184 CHAPTER 5 San José Martin Luther King Library San José Martin Luther King Library (henceforth referred to as “King Library”) is oneof-a-kind in the United States, since never before has the merger of two such different institutions as a university library and a city library been attempted.1 Allowing a community of patrons as diverse as young children, teens, senior citizens, university students and faculty to meet under the same roof is a revolutionary experiment that redefines a library’s mission as one of supporting a whole community, with all its diversity. The idea of a joint library was born in 1996 from the minds of two visionaries – former San José mayor Susan Hammer and former San José State University president Robert Caret – who successfully broke the pattern of indifference and disdain, which had been representative of the relationship between the two entities over decades.2 Indeed, on the one hand, the city of San José had been struggling over the years with urban decay downtown; the university, on the other hand, faced internal cultural changes and had become very insular. Very little was shared between the two institutions. However, with 185 Caret and Hammer, things began to change. On her side, Hammer viewed the urban university as a main “player” in restructuring downtown. On his side, Caret considered San José State to be a “metropolitan university” and started to develop partnerships that benefited both the university and the community. Both the city and the university had inadequate libraries and were in need of more space; neither of them, however, had a budget that permitted them to expand on their own. Both Hammer and Caret realized that by bringing their resources together, they could build a landmark building that neither could afford alone. The result was a $177.5 million library jointly funded by the university and the city.3 King Library opened its doors on August 1, 2003. It is located in the heart of downtown San José, at the corner of San Fernando and 4th Streets, just a block from the future $343 million civic center currently under construction that will be surrounded by new retail and housing projects. The design is a collaborative effort of Carrier Johnson from San Diego, California; Gunnar Birkerts Architects from Bloomfield Hills, Michigan; and the San José firm Anderson Brulé Architects (ABA). King Library is a massive, 477,148-square-feet structure including eight stories above ground, which is built on a tight 80,000-square-feet lot belonging to the university. It is composed of two blocks that intersect one another: the first block is a four-story structure that constitutes the base of the building; the second block is a heavy, eight-story structure. Because of the difficulty of fitting the library in such a small lot, the building looks bulky and for this reason, the library has been dubbed by design critic Alan Hess, journalist at the San José Mercury News as “[a] wonderful gift in a bad package.”4 186 The library “package,” due to the somewhat austere exterior glass-and-concrete detailing, may seem more representative of a bureaucratic organization than a library. However, the design of the interior of the library, which offers a broad range of humanscale details that make the space very user-friendly, is the “gift” offered to library patrons. Individual desk spaces next to the outer wall, for instance, allow for privacy while the window lowered to the eye-level of someone seated at the desks provides balanced light. The work on natural lighting is particularly well done: the building connects downtown San José with the university campus through an internal “main street” covered by a large atrium that has already been celebrated for the rainbows created on the walls and walkway inside the building by the prism-like glass ceiling. Wells on the ground floor let light pour into the basement. Patrons using the reading room on the top floor can enjoy lighting from two curtain walls that shows striking views of the city, the campus and Silicon Valley. Thirty-three whimsical art installations by New York City-based conceptual artist Mel Chin have been randomly dispersed throughout the eight-story building to surprise, if not intrigue, library patrons. Some of them have been designed as puzzles to be solved, others simply as objects for contemplation. The library has been a great success from the first day of its opening. The estimated use rate of 8,000 users a day has been largely exceeded and is closer to 12,000. One just needs to wander throughout the building to witness the cosmopolitan nature of the library. King Library that hosts a broad diversity of patrons from all ages, styles and backgrounds provides much opportunity for cultural blending and social exchange. The 187 building has clearly been designed with the users in mind. As Patricia Breivik, Dean of the San José State University (SJSU) library could not help, but notice, “There is a feeling of happiness in the library.” This chapter analyzes King Library project from a living system’s perspective and assess the adaptive capacity of the project team throughout the project life cycle. Section 1 introduces the case study. Section 2 describes the adaptive challenges faced by the project team. Section 3 analyzes how the client team successfully created a collaborative environment that supported its adaptation during the development of the integrated program and operation plan for the joint-library. Section 4 discusses the strategies used by the project team to keep the momentum going through the design and construction phases. Section 5 concludes this chapter with some final remarks about the innovations that added value throughout the King Library project life cycle. 5.1. King Library Case Study Introduction This section first introduces the case study methodology. Next, we present the project timeline and the project organizational structure. Finally, a table of project specifics that describes people interviewed, their company and their role on the project is included. The reader may easily refer to this table throughout the reading of the case study. 5.1.1. Case Study Methodology The interviews for the case study took place over a few weeks during the fall of 2004. As with the Terminal case study, the individuals I interviewed belong to organizations that formed the core project team, i.e., the client, the architects, the construction manager and 188 the general contractor. Participants were identified based on the critical role their firm played in the project; they were themselves directly involved in the project. Interviews were conducted at a place the participants suggested, usually their office. The interviews lasted from 60 to 90 minutes. I met only once with the participants, except for Sam McBane Mulford at Anderson Brulé Architects, whom I interviewed a second time to understand in greater detail the databases she had developed for the project. I had a few follow-up telephone and email conversations with some participants to confirm particular points of the project story. At the time of the interviews, I considered project organizations to be living systems, but I had not yet fully developed the performance model presented in Chapter 3. My interest, as in the previous case study, was to understand the impact of organizational variables (i.e., project context; organizational structure; communication and coordination processes; technologies; as well as the professionals’ culture and behavior) on a project team’s ability to flexibly adapt to contingencies. As with the previous case study, I asked open-ended questions, recorded interviews, took notes and then described the project participants’ narratives about the organizational variables. The transcripts served as the document source of my analysis. Based on this data, I wrote a draft of the case study, showing how the project team organized its environment in order to achieve high-level performance. Once I completed the development of the theoretical model, I revised the writing to show how the project organization variables influenced the control parameters in a way that allowed the project team to adapt to its challenges. 189 5.1.2. Project Timeline and Organizational Structure Similarly to the previous case study, this section provides a high-level project timeline (Figure 5.1). The King Library project idea emerged in 1996 as former San José mayor Susan Hammer and former San José State University president Robert Caret “dreamedout” the concept of the joint library. Hammer and Caret signed an official intent to collaborate in early 1997. By February 1997, multiple library committees were formed and given the task of assessing the feasibility of the project. The San José Redevelopment Agency (SJRA) was chosen as the project leader and main decisionmaker on the client side. Soon after the committees were formed, Anderson Brulé Architects (ABA) was hired by the client to facilitate the feasibility study, and then, later on, the development of the operation plan. Jane Light, Director of the San José Public Library (SJPL), hired Pamela Anderson-Brulé, ABA president, because her company had gained recognition within the San José community, as well as the region, as being a very good facilitator. Both Anderson-Brulé and Sam McBane Mulford, an associate at ABA, became personally involved in the facilitation process with the client team. Later on, ABA was hired as a member of the design team for the project. While ABA was still facilitating the development of the library operation plan, the SJRA initiated the design team selection process. Seven design teams comprising nationally-recognized architectural firms were invited to present their design solutions to the client. In June 1998, the panel selected a design team consisting of Carrier Johnson, in association with Gunnar Birkerts Architects and ABA, to design the King Library. Gordon Carrier, a Principal at Carrier Johnson, was a student of Gunnar Birkerts and he 190 had had a long relationship with him. Moreover, Carrier Johnson and Gunnar Birkerts had done other projects together over the years. So they decided to forge a partnership with one another for this project. ABA, who had been working with the client for approximately two years doing the initial feasibility and operational planning of the project, was asked by five out of the seven design teams that were interviewed for the project to join their team. ABA interviewed each of the executive architects to make sure that the firm would have a meaningful role in the project and that there was a cultural fit between the design firms before agreeing to team up with Carrier Johnson and Gunnar Birkerts. The division of responsibility was clearly established among the design team members. Carrier Johnson was the executive architect responsible for documenting the entire project and the firm was mainly responsible for the design of the public interior spaces; the development of all construction drawings; and the supervision of library construction. Kevin Krumdieck, architect at Carrier Johnson, was the main leader for the design team. Gunnar Birkerts was responsible for the overall massing and external aesthetic of the building. ABA, in addition to its role as the facilitator for the development of the library program, was the local associate architect. ABA facilitated the interface with the client team; provided the interior design for the operational areas of the library (i.e., non-public spaces), in addition to areas related to special collections and the multi-cultural center; and assisted Carrier Johnson in the design, specification and procurement of the furniture, fixtures and equipment (FF&E). Soon after the strategic operation plan was completed in March 1999, Dolores Montenegro, project manager at the SJRA, got involved as the main project manager on 191 the client side. Montenegro was given overall project management responsibility. She was the final decision-maker on issues related to design changes and budget. Montenegro’s counterpart at the University was Art Heinrich, campus architect. Very early in the process, while schematic design work was still going on (i.e., early 1999), the client team hired Gilbane Building Co. as the construction management firm. Two construction managers from Gilbane participated in the project before Jim Klingensmith became involved a year later. Klingensmith remained on the project until the completion of the library. In August 1999, Dr. Breivik arrived in San José to replace the former SJSU library dean. Her arrival in the middle of the process could have been a major project disruption. (As we will see, this was not really the case.) Hensel Phelps Construction Co. had been invited to participate in the bidding process and had the lowest bid; the firm was hired at the end of 2000 as the general contractor for the project. Greg Overhage, the project manager, oversaw the construction of the library until its completion. The construction of the new library started in July 2000 and lasted three years. The building was completed in July 2003. The moving of books and the library staff was completed only a week before the King Library’s grand opening on August 1, 2003. 192 1996 Hammer and Caret “dream-out” the vision for a joint library. 1997 Hammer and Caret officially sign “intent to collaborate” document. Feasibility study committees are formed. ABA is hired to facilitate feasibility study. P. Anderson Brulé and S. McBane Mulford get involved. 1998 ABA facilitates feasibility study. Library operational issues are discussed. Partners sign Memorandum of Understanding. Partners sign Operating Agreement. Design team (Carrier Johnson – Gunnar Birkerts – ABA) is hired. K. Krumdieck is the design PM from Carrier Johnson. D. Montenegro from SRJA becomes PM representing client. Design drawings are completed. Construction drawings are completed. Gilbane is hired as CM. Hensel Phelps is chosen as General Contractor for the project. G. Overhage is the GC PM. J. Klingensmith from Gilbane gets involved. King Library construction starts. 2000 1999 The Strategic Operational Plan is completed. Patricia Breivik takes on the responsibility of dean of the university library. Staff and equipment from the old Wahlquist university library are moved out of the site. 2001 2002 2003 King Library construction is completed. Books and staff are moved to the new library. King Library’s grand opening. Figure 5.1: King Library Project Timeline 193 5.1.3. Project Specifics Participant Name Jane Light Patricia Breivik Dolores Montenegro Art Heinrich Kevin Krumdieck Pamela Anderson-Brulé Sam McBane Mulford Jim Klingensmith Greg Overhage Company San José Public Library San José State University Library San José Redevelopment Agency San José State University Carrier Johnson Anderson Brulé Architects Anderson Brulé Architects Gilbane Hensel Phelps Role SJPL Library director – client SJSU Library dean - client Project manager University campus architect – Project manager Design project manager Feasibility study facilitator Associate local architects Feasibility study facilitator Associate local architects Construction manager General contractor Table 5.1: King Library Project Participants 5.2. Miracle on 4th Street The challenges faced by the King Library project team were daunting at many levels: contextual, operational, organizational, cultural, procedural, technical and architectural. The project was very high profile and much was at stake as patrons from both the campus and the city watched the merger of the two major institutions. The clients had never worked together in the past and had the reputation of being insular. Because there was no precedent for a joint library, they had to innovate to combine their respective operational needs into a single set of requirements, while maintaining the integrity of two completely different cultures. 194 Hammer had defined the main library challenge as making the library operationally smooth, efficient and user-friendly.5 The library had to provide “seamless services” to its users. Because of the novelty of the concept, many believed the experiment would fail. Opposition to the project quickly developed, in particular on the side of students and faculty, who worried that the availability of the university materials to the general public, i.e., 800,000 potential library users, would impact student access to required materials. The difference of users’ needs – from a graduate student working on a thesis to a toddler playing near-by – was perceived as a gap that could not be filled. The merger of two institutions with different missions and goals was believed to be a huge mistake and the very reason why it had not been done before. On the city side, some worried that city patrons would not be welcome on the campus grounds and that the joint library would cost more to operate.6 Like most projects with complex programs, the budget was tight ($177.5 million for an approximately 480-square-feet structure and a large program). More importantly, it was provided by two different institutions with different agendas – a situation that had the potential for creating conflicts between them, for instance, at a time when one of them might have requested a design change that the other may not have liked or wanted. Who would pay in such a case? How would decisions be made? The risk of the project coming to a halt was great. Procedurally speaking, the project was highly complex. The project was governed by the State of California and any design changes required its approval. In order to secure funding from each side of the ownership and realize these funds on time, it was critical that the decisions be made in a timely fashion in order for the project team 195 to keep making progress and deliver the facility on schedule. This was particularly critical during construction because the team utilized a fast-track process. Operationally, the project complexity was somewhat similar to that of the Terminal project. King Library itself involved thousands of individuals whose work had to be coordinated effectively. The huge amount of information generated in defining the program, creating the library’s operational agreement and facilitating design, procurement, construction and logistics had to be managed and communicated between participants so that all could remain fully informed. King Library was built on the same lot as the old university library. The entire staff and thousands of books had to be relocated from the old facility to another location for the duration of construction of the new building and then, moved back in time for the opening. On the design side, Krumdieck, project manager for the design of the library at Carrier Johnson, recalled that the biggest design challenge was to fit the huge joint library program onto such a small lot. In addition, the design team included three different firms in three different locations, which could have been an impediment to coordination and collaboration. Despite the complexity of the project and the inherent uncertainties faced by the “two-headed” client and the project team, the King Library project was relatively smooth and an overall success: the facility was delivered on time, below budget and without a single claim. Months later, the two clients were still highly satisfied not only with the quality of the building itself, but also and perhaps more importantly, with the joint 196 operation plan they have developed. Even early opponents are now ready to admit the library provides what the city and university patrons needed. Each one of the key project participants I interviewed was very proud of the final product and adamant that the project had been a very rewarding experience on a personal basis. For the firms involved, King Library has been a great project to add to their portfolios and a useful marketing and business development tool. Keeping the project context in mind, should we consider this project the “Miracle on 4th Street,” as some people have dubbed it? Or, if it is not a miracle, what factors, then, contributed to the project’s overall performance? If one considers the King Library project’s delivery method at a macro level, there is nothing really “new” or “cutting edge” about the approach used. The project is based on a traditional design-bid-build delivery method. It leveraged partnering; used best-inclass facilitation methods; and followed highly complex, yet standard procedural methods. What then made this project so different from similar projects? Its success seemed to reside in the project team’s ability to operate as a complex adaptive system. As we shall see in the next sections, the project team was able to modify its organizational variables in order to influence the control parameters previously defined (see Chapter 3, p. 95). In doing so, the project team was not only able to migrate to the edge of chaos, but was also able to maintain the required tension between the forces at the origin of strange attractors – forces that facilitated its creative adaptation and self-organization when faced with uncertainties. 197 5.3. Client’s Transformative Challenge The main problem faced by the “two-headed” library client at the early phase of the project was to figure out how to develop a joint operational plan that would allow the two organizations to successfully co-habit in the same facility. The problem was illstructured: there was no precedent joint library in the U.S. and there were neither readymade nor simple answers to the problem. In contrast to technical problems, i.e., problems for which one already knows how to respond to them (e.g., dealing with patients in a hospital emergency room), the library problem was adaptive. As Heifetz (1994) mentioned, adaptive challenges require organizations and individuals to learn new ways of doing things; learning, in such a case, occurs through experimentation and discovery. Consequently, the “two-headed” client had to embark on a transformative journey that provided an opportunity for the librarians to reflect on their respective organizations in order to discover the purpose and identity of the joint library. In the sections that follow, we shall discover how the King Library client re-organized its environment in order to facilitate its transformative adaptation. 5.3.1. Attractor of Meaning The bold decision of the San José mayor and of the San José university president to pool their resources to build a common library based on the concept of “seamless services” created opposite forces that resonated as a strange attractor. On the one hand, by agreeing to collaborate, Hammer and Caret broke the pattern of insularity that characterized the historical relationship between the university and the city and challenged the conventional order. By reaching out toward one another, they 198 demonstrated leadership at the highest level of the project organizational structure – and demonstrated to each individual on the project that King Library was, indeed, a very high-profile project. The concept of “seamless services” provided a clear vision that the librarians could embrace and on which they could focus their efforts. Hammer and Caret’s attitude of “Let’s make this work!” became the example that all the project participants would follow if they were to remain on the project.7 Because there was no precedent for a joint library in the U.S., this attitude was critical for establishing an environment prone to cooperation, creativity and thinking outside-the-box. Moreover, as both Light and Breivik noted, the union between the city and the university libraries was not a merger, but a “marriage,” since both libraries still existed as different and independent entities. However, the idea of a “marriage” was a powerful metaphor to drive behaviors. As Breivik pointed out, “[I]n a merger, someone loses their personality….This is really a marriage – two strong entities that have come together and are stronger and can accomplish more because they chose to be together.”8 Krumdieck, the executive architect, noted that the marriage of the two libraries had a positive effect on the project because the two clients were “joined at the hip.” This situation forced the clients to cooperate, walk in the same direction and make decisions that benefited both of them. As with two individuals in a marriage, the leaders of both libraries had to figure out how to overcome their differences and arrive at a consensus that was acceptable to both parties, even when they were not able to identify win-win solutions. On the other hand, however, the unique marriage between the two different libraries also created much uncertainty. The novelty of the concept of “seamless services” was far removed from the comfortable zone within which each library staff had 199 been used to operate. There was much doubt in the minds of the librarians (as well as in the minds of many patrons) that the idea of a joint library was operationally sound. At the outset of the project, there was much uncertainty about the outcome since no one knew for sure whether the experiment would be a success or a complete failure. Thus, resistance to the concept and pessimistic attitudes of “It will never work” and “We should not go there” arose and came to challenge Hammer and Caret’s attitude of “Let’s make this work.” Had opposition to the King Library concept overpowered Hammer’s and Caret’s vision, the project would have been aborted; had the librarians been too complacent or had they not worked with one another, the client team might have never been able to successfully create an integrated operation plan between the two libraries. While Hammer’s and Caret’s vision was very powerful and served to counterbalance pessimistic attitudes, it was, however, not sufficient to ensure the success of the project. Indeed, one may argue, what is the power of a vision when one does not know how to make it happen? A vision without a supportive process and work environment is not sufficient to spur innovation and creativity. The true meaning of the vision for King Library did not resonate in the librarians’ minds until they actually began interacting with one another. It is thanks to the repeated real-time conversations that the vision that seemed, at first, to be unachievable eventually started to make sense and became meaningful to the librarians. Thus, while the vision of “seamless services” was defined by the San José mayor and university president, its “true” meaning was co-created by the librarians during the feasibility study and the development of the operation plan phases. 200 Such an attractor of meaning gave the client team a sense of purpose and reinforced all participants’ willingness to give their time and energy to help the process move forward. The more the librarians interacted with one another, the more positive feedback was created, which reinforced the meaning of their collaborative activities. Each library organization was able to transcend its own individual organizational identity and enlarge its boundaries to include a larger, emerging organization: King Library. As the librarians started to refer to the joint-library concept, their behavior shifted from being driven by self-interest toward a common pursuit of the success of King Library. As Peter Senge (1990, pp.234-235) notices, when shared goals and a commonality of purpose emerge, “[i]ndividuals do not sacrifice their personal interests to the larger team vision; rather, the shared vision becomes an extension of their personal visions.” Such an extension of personal vision seemed to have occurred in the King Library project. While the attractor of meaning pushed the librarians to the edge of chaos, it did not completely destabilize the client team. Indeed, with ABA’s help, the librarians were able to successfully re-organize their team and create a work environment that allowed them to adapt to the situation. 5.3.2. Creating Stability by Focusing on the “How” At the early stage of the planning process, King Library clients had engaged an architect experienced in library design to do a program analysis of needs for the joint library. Strangely, that architect held a very negative attitude toward the project and was openly skeptical about its success. Based on interviews with the client advisory committee that 201 had been created to assess the feasibility of the project, the architect, nevertheless, developed a preliminary program that established the square footage of the new building. King Library is indebted to SJPL Library Director Light for her leadership in realizing at the outset that the project could not be successful if participants did not believe in its potential success. By hiring ABA to facilitate the feasibility phase, Light created the opportunity for a completely new approach to problem analysis and resolution – an approach conducive to collaboration and learning. When Light hired ABA, she entrusted the firm with a very powerful charter. Light told Anderson-Brulé: “I am hiring you to be the optimist. I am hiring you to never ask the question if, but to ask the question how.” Focusing on the “how” question was very inspiring for Anderson Brulé and her team; this charter became central to ABA’s method of facilitation. When, at the earliest stages of the project, the question in everyone’s mind was “What if…,” ABA challenged the client team to answer questions such as: “How would you do this?” “What gets in your way?” “What would stop you?” “How could you move past that thing that stops you?” The focus on “how” provided a structure to librarians’ discussions and ensured that the team members were maintaining their focus on important questions. It thus helped the librarians to hold their level of anxiety at times when they were uncertain about whether they would succeed in achieving a common solution. The focus on “how” counterbalanced the project uncertainties and helped the librarians to overcome their fear of failure. The “how” questions helped the project team to create a project “philosophy and personality...that was so contagious that it ended up feeding out the seven long years that 202 it took to build up the library.”9 This attitude toward resolving issues by figuring out how to make things work was re-emphasized at each main project phase, as new participants got involved. The project philosophy became so pervasive that it spread out to all the participants, down to the general contractor and subcontractors. The focus on “how” was used in partnering sessions, for instance, to help team members, subcontractors included, align their goals with the project vision. 5.3.3. Fostering a Collaborative Environment As ABA was getting involved in the feasibility study, the firm quickly realized that the previously-formed client advisory committee had not reached out to the staff or involved the librarians in the process. Because the library experts were not engaged, the client team did not have “representative, leadership and real-world experience associated with building the library.” Moreover, ABA believed that the definition of the library problem went beyond the “simple” definition of the joint-library program requirements. The marriage between two libraries was unique and it was critical for the “two-headed” client to undergo a transformative process in order to redefine its purpose and identity as a joint library. ABA believed that the solution to the library problem resided within the client team and with the librarians. Thus, ABA needed a process, adapted to the King Library client’s context, which would be able to facilitate the interactions between and the integration of the two library teams. ABA redesigned the client team organizational structure in such a way that the new structure would support collaboration between librarians from both organizations. To do so, ABA considered the components of the project system, i.e., the participants, 203 their cultural and organizational roles, as well as the project deliverables, and tested these components against each other until a new project structure and process emerged. The final outcome was a process map of 140 meetings or so necessary to achieve the goals – each of which was two to three hours long. Meeting participants were “technical teams” that included librarians from both the university and city sides; these teams were knowledgeable about the different technical areas of library operation (e.g., technical services, collection development, users’ services, circulation services and so on). In hindsight, the design of the technical teams seems to be a fairly straightforward idea. However, one needs to recall that the two library organizations had never worked with one another before and that there was much doubt among their staff about the viability of the project. Considering the divergence of opinions that existed within both teams, the prospect of having to collaborate with one another did not seem highly promising to the librarians. The attitude of “This will never work!” that dominated each client organization and was reinforced by the opposition to the project from libraries’ patrons was a force that supported the maintenance of the status quo; this force could have potentially succeeded in driving the project to a halt. In contrast, the crossfunctional and cross-organizational meetings designed by ABA pushed the librarians to move out of their comfortable organizational silos and to begin entertaining the idea that the marriage between the two libraries was not only possible, but also a potential benefit for the two libraries, as well as the entire San José community. The new collaborative process operated as a strange attractor, maintaining the two client organizations at the edge of chaos. The process not only encouraged the librarians to work in a creative way, but also provided a structure that allowed the librarians to 204 slowly adapt to their transformative challenge. In the next four sections, we shall discover how ABA used this strange attractor process as a means to maintain the client at the edge of chaos. 5.3.4. Unleashing the Distributed Intelligence in the Client Team The merit of ABA’s method of facilitation was its inclusive nature: the technical teams blurred the boundaries between the two librarian organizations and transformed the client team into a communication network in which information flowed among the diverse experts in the room. The cross-functional and cross-organizational teams’ meetings provided a platform to tap into the distributed collective intelligence that resided within the city and university librarian organizations, thus leveraging the diversity of knowledge, expertise and opinions, which resided within the client team. Librarians from both organizations had an opportunity to share their respective points of view with their counterparts and identify issues they needed to resolve for the library to operate successfully. The co-location of the library experts seemed to have been critical at that early project phase to facilitate collaboration. Repeated, face-to-face interactions between the librarians facilitated knowledge exchange among the participants. Real-time communication was essential to facilitate the integration within the client team because short feedback loops helped to reduce the latency between the time an issue was identified and the time needed for someone on the team to provide insight on how it might be resolved, thus saving precious time to the team. Moreover, ABA used the meetings with the libraries’ technical staff as a springboard for influencing positive behavior. Indeed, trust between the participants was 205 not really present at the onset of the process and latent negative attitudes of “This will never work!” were quick to come back to the surface. However, negative attitudes most often originate from ignorance or from old habits and mental norms, rather than from intention; such attitudes can be changed by individuals who are willing to influence their peers. As McBane Mulford of ABA pointed out, in order to influence people, “you have to be yourself an example of positive behavior.” To become such an example, ABA ensured that a lot of work was accomplished during the meetings. The meetings were not simply used as a forum to communicate general information, but also as a place where issues were raised and decisions were made. ABA asked the staff to do some homework and be highly interactive during the meetings. No matter what, ABA’s role as facilitator was to find ways to move the meetings forward “with good humor and positive attitude.” Attitudes and behaviors improved over time, meeting after meeting. Admittedly, it first required a leap of faith from the participants, but then this leap of faith was leveraged and the client team’s organizations were turned around. Eventually, as the participants became familiar with the process and method of working, peer pressure emerged that was associated with being prepared for a meeting and not allowing some individuals to disrupt the momentum, because they did not do what they were supposed to do. Meeting after meetings, as ABA led the team and encouraged the librarians to explore how they could overcome their differences, a sense of teamwork and trust started to develop between the librarians, which reinforced the relationships among the team members. As McBane Mulford described: If you realize you’ve got 15 people in a room who don’t really like each other… they are from different organizations and do not respect each other, [but] they are from the same profession. In the end [of the process], you really have not only resolved the conflicts, but you have created a culture of collaboration… they, they 206 have created a culture of collaboration and they have created an attitude of ‘can do’ rather than ‘why? Why do?’ And they planted the seeds for the catalyst of actually achieving this [project] seven years later. The cross-functional, cross-organizational meetings provided an environment where librarians could be “hard on the issues, [but] soft on the people.”10 As positive behaviors and trust between the librarians slowly emerged, the cohesion within the client team improved, thus facilitating discussions on issues that might have otherwise been marred by conflicts. The librarians had then an opportunity to fully contribute to the process and leverage each other’s knowledge within an environment that facilitated learning and creativity. 5.3.5. Documenting Everything At the early stage of the facilitation process, it became quickly obvious to ABA that their firm would need to find a way of effectively managing the information captured during the 140 or so meetings with the technical teams. McBane Mulford quickly realized that a Word editor program would not be very effective in managing the vast amount of information that needed to be captured. In addition, meeting minutes captured in a static word document format had to be read either linearly, online, or in hard copy. Linear reading, however, would have forced the reader to figure out by him- or herself the evolution of the team thinking across consecutive meetings – a cumbersome mental activity, considering the amount of information that would be generated. In addition, McBane Mulford did not know at the onset of the project how information would need to be organized and how it would be used down the road. What the team needed, instead, 207 was a database that would record the information in such a way that information could be retrieved by using queries. While she had very limited knowledge of database design, McBane Mulford learned the concepts behind relational database design on her own by reading manuals and taking apart databases that were available to her. The database McBane Mulford created provided a means to capture information in an incomplete or unfinished state at times when problems had not yet been resolved. Issues discussed in meetings were captured as different records in the database. All discussions around a particular issue across all the meetings were related to the issue ID so that one could create a report that identified all the discussions related to that particular issue. By capturing the rationale of the decisions made during the meetings in addition to the final decisions, ABA was able to ensure both the integrity of the outcome and the development of a shared understanding among the team members. As McBane Mulford pointed out: We [the team] could not lose the richness of the discussions because we hadn’t come to a resolution. And, even if we had come to a resolution, why, outside of the attendees inside the room, would anyone buy it if they hadn’t been exposed to the richness of the discussion? Such a complete and rigorous documentation process was critical to facilitate the flow of information among team members and to keep different individuals working on different technical areas informed of each other’s progress. This knowledge management method improved the richness of the connections within and between the technical teams and facilitated the management of interdependencies among the diverse operational components of the library. The results of the database queries were more informative than reading a static document’s narrative because the queries could be built around a very specific issue and 208 its development over time across consecutive meetings. Since the rationale for the decisions made was also captured, it was easy for someone to understand the evolution of a particular idea up to and through the final decisions. The database also saved ABA staff much time when the firm needed to create reports because these reports could be created automatically as a result of one or more queries. 5.3.6. Embracing Complexity through a Systemic Process ABA encouraged the librarians to fully embrace the complexity of the King Library’s program as a whole. Anderson-Brulé described the overall process ABA went through with the technical teams as one of “peeling the onion.” It is a very iterative process in which problems are identified in a collaborative manner with the experts in the room, first at a high level and then, in meetings upon meetings, in a more detailed manner. This method is founded on systems thinking, i.e., it is a method that takes into consideration all the components of the project from both library organizations’ perspectives, simultaneously and holistically. At first, the goal of the discussions with the technical teams was not to solve problems, but to understand what the problems were and to start identifying the strategic steps that needed to be taken in order to solve them. Indeed, trying to find solutions to problems that were not yet fully understood or even acknowledged could have been overwhelming and a complete waste of time. Once the pressure of making immediate decisions was removed, the team gained increased freedom to think outside-the-box. ABA’s process created an environment conducive to learning. Learning, in this context, is an open-ended process that supports information and knowledge exchange 209 with the goal of defining problems, as well as creating and testing different ways to resolve them. It is an iterative process that requires much interaction between the people involved and therefore, much flexibility in terms of communication structure. ABA’s method of facilitation helped the librarians to “reflect-in action” or – to apply Schön’s (1983) concept to the specific collaborative context of King Library – to “reflect-in-interaction” (Fruchter & Swaminathan 2005). As the librarians interacted in real time on diverse library operational topics, they started to make sense of new information that was provided by their peers and use their own experience to reflect on the issues at hand, not only from their own organizational perspective, but also from the King Library’s perspective. Through repeated real-time interactions, the librarians came to grasp the differences and similarities between their libraries’ operational styles. They could uncover a nuanced picture of their needs and requirements, which provided a rich framework to test diverse alternatives and solutions. In some cases, they realized that the two library organizations faced similar issues, but that they had applied two completely different approaches to resolve them. In other cases, their problems were different and they came to explicitly acknowledge their differences, thus enlarging their understanding of King Library’s operational needs. The team kept issues open as long as possible until the problems were understood to the point that their resolutions became obvious to all participants. Anderson-Brulé and McBane Mulford tirelessly asked questions and challenged the participants, session after session, so that they would discover solutions for themselves. As Anderson-Brulé and McBane Mulford had never built a library before, they made no assumptions: “Listened to every word and asked a lot of questions.” 210 It is interesting to contrast ABA’s process with the way the first library program had been developed. The first library expert had defined the outcome (i.e., the library program) and had hoped that the librarians would figure out by themselves how to organize their operation to make the program work. In contrast, ABA defined a collaborative process in which the librarians could define their new joint operation plan and the library program requirements simultaneously. Obviously, one would expect the two different methods to deliver two very different outcomes. As Anderson-Brulé explained, ABA “defined a structure and a discipline to the process that was enough to allow the creativity thing to happen.” While no one in the team knew what the “true” King Library problem was and even less about its solution, ABA was confident that both the definition of the problem and its ultimate solution (both organizational and programmatic) would eventually emerge from the process itself. ABA considered the process to be an “exploration” because, as McBane Mulford acknowledged, “Especially when you are undertaking an exploration [of the problem] like this, you don’t know what the end looks like. Therefore you can’t begin with the end in mind; you don’t know what the outcome is.” By considering the process as an exploration, ABA could spur the creativity of the client team members. Each meeting was a “safe” place where the librarians could learn from each other and “play” around with ideas. ABA’s questions were catalysts: they kept the team focused on the issues, while each question was an opportunity for the librarians to learn about each other organizations and tap into each other’s expertise to devise means to bridge the gap between their different needs. As the client team explored diverse options, the focus on “how” was a means for ABA to control the process and 211 bring back the discussion on track at times when the librarians started to wander off to non-value-generating discussions. By embracing the whole complexity of the problem they had to solve and facilitating the cross-pollination between the diverse requirements of the two libraries, the client team was able to fully integrate their respective needs into an innovative operation plan and a new program that could sustain their joint activities for the long run. The integrated program eliminated waste (e.g., duplication of space); maximized utility (e.g., the same space could be used for different functions without conflict between the two organizations); and maximized the value delivered by the King Library team within the allocated budget. In addition, by the end of the process, the “two-headed” client had had an opportunity to define its joint identity and was therefore fully able to operate King Library. One may argue that a participative and iterative method such as the one used in the King Library project is very costly to a client. While such a process may take longer and is resource-intensive, it, nevertheless, ensures that all aspects of the problems are taken into consideration and are understood by all the stakeholders such that the program developed will have effectively resolved the problems identified. Therefore, the process maximizes ownership and minimizes the risk of conflicts down the road. Moreover, while such a process requires the investment of more resources at the early phases of the project, it has the potential for saving time and money for the client in the long run. Indeed, iterations that occur early in a project are positive iterations (Ballard 2000), i.e., they save time and effort at the design phase because designers can work on specifications that are comprehensive and that had been fully worked out by the 212 client. In contrast, in many projects in which requirements are not comprehensively defined, negative iterations occur late in the process due to design changes requested by the client, which impose rework requirements on the design team. All project contexts are different, however, and not all projects require a transformative work as intensive as the one that took place for the King Library client. 5.4. Keeping the Momentum Going Through Design and Construction The interactive and collaborative processes that supported the feasibility study and development of the operation plan were sustained through design and construction. The project team was able to so by: (1) facilitating a creative dialogue between programming and design; (2) involving project participants (i.e., designers and construction managers) early in the process; (3) implementing flexible, decentralized communication and decision-making processes; (4) building the right project team chemistry; and (5) balancing the degrees of power differentials among the professionals involved in the project. 5.4.1. Creative Dialogue between Programming and Design The King Library project benefited from the early involvement of the design team in the project, while the librarians were still working on the operation plan and the library program; this allowed program, operational plan and design development activities to inform each other. The decisions made by the librarians during the early phases of the project became “little fixed points [i.e., program-related decisions] around which other things [e.g., design strategies] could gel.” Furthermore, they supported a “wonderful 213 creative dialogue” between the operational, programmatic and architectural aspects of the building.11 Throughout the feasibility study and the operation plan development process, ABA became extremely knowledgeable about the client and its environment. ABA became such a “highly developed part of their [the client] organization” that the client team started to fully trust Anderson-Brulé and McBane Mulford and empowered them to be very proactive. Since ABA was a member of the design team as well, the firm operated as a conduit between the client and the design team and became a catalyst for systems thinking across the different activities. The benefit of ABA’s early involvement with the client team was not only to have become knowledgeable about the library’s requirements, but also, perhaps more importantly, to fully understand the rationale of the decisions made. That is, the firm understood the relationships between the decision criteria related to the library program; the different alternatives proposed by the librarians; and the reasons why the librarians had made a particular choice over another. As ABA began to work on the design of the library, ABA started to gain knowledge about design constraints as well. Thus, Anderson-Brulé and McBane Mulford could see the “domino effect” of a particular design decision on the library program and operation plan and vice versa. When interacting with the executive architect, ABA could facilitate the sharing of information between the client and the design team via very short feedback loops. This enabled the executive architect to make informed decisions at the early stages of the design process – decisions that would not need to be re-evaluated at a later phase – thus, down the road, saving time and money. As Anderson-Brulé explained: 214 We [ABA] were going to meetings with the purpose of [doing] X, and we [the meeting participants] were adding Y to that. So, we would get it [X and Y] done in about the same time... We learned how important it is to let the architectural elements, the urban aspects of something, the site, its context, sustainability issues, lighting...you know, where infrastructure is, all of the form aspects of buildings – to let it begin to live and dominate and have an opinion, while you’re formulating the operational and functional, and cultural and spiritual aspects of something, and let them go at it, let them be that creative battle and push and pull between the two, with a full understanding… at each layer of the development, of the impact that they have on each other. The idea, before, is [that] usually one would win out over the other. There wasn’t this wonderful creative dialogue that would happen between those two needs. And it certainly wouldn’t happen in this sort of real-time environment, where people are highly charged and educated and beginning to put their arms around something. That was a very, very, different effect. The process that went on between the client team and the design team was very dynamic. On one side, the project context and library requirements informed the design decisions. At the same time, the client was creating a new identity that was formulated around a new operational, functional and cultural personality – a personality that also interacted with the design of the space such that, through a creative dialogue, the client’s needs and the design solution eventually fully aligned with one another. Moreover, the creative dialogue helped to manage the interdependencies among the architectural components of the building. As Anderson-Brulé recalled ABA’s interactions with the executive architect during the design phase: [W]e would be working on a whole area of the building and Carrier Johnson would discover that they had to move the whole front end, [the] whole elevation of a wall, seven inches, in order to miss an infrastructure issue that came up, that would save $100,000 if we do xyz and...we would go: Okay! Got to move it! We knew enough of the impact on the operation, [we] knew enough about the design concepts and theories that we had already developed to do the [staff] spaces, we would redraw, redesign whole pieces of the floor plan, if not the whole floor plan. But that was okay! That was okay because, every time, we were checking into the domino impact of that on the other aspects of it. We could push back and they would let us. So there was really a fabulous sort of push and pull that went on throughout the process to keep checking into… 215 The “creative battle” that took place between programming and design activities was representative of the synergy that occurs between competitive and cooperative forces that operate in the fitness landscape of a project ecosystem at the edge of chaos; this creative battle operated as a strange attractor. In the King Library project, the program requirements did not fully dictate the direction of the design nor was the design fully imposed by the design team as sometimes happens in projects in which architects emphasize architectural form over building function. Instead, in the King Library project, design and program teams interacted with one another without the function of one group winning over the other. As a result, each activity evolved in parallel with the other, with the performance of one activity influencing the performance of the other. Since so much information was already flowing between the project subsystems, the negotiations required for each subsystem to reach a satisfactory level of performance was relatively simplified as those negotiations took the shape of a creative dialogue between the project’s components. Krumdieck remembered that there also were a lot of feedback loops among the three design offices. At times, each design office was trying to influence the others to make changes on a particular design area they were working on so that the project as a whole could benefit from it. For instance, Carrier Johnson would explain to Gunnar Birkerts why some element of the exterior design was not the best solution from a library operation perspective, program perspective, or stacking perspective. And after multiple “push and pulls” to test out the idea, Gunnar Birkerts would eventually agree to the changes because enough had been learned from peers that the interdependencies between the components could be understood. Thanks to these near real-time interactions 216 between the members of the design teams, delays that were related to the identification and the resolution of specific issues were significantly reduced. Thanks to such a dynamic and iterative approach, the design team minimized the waste of resources and much rework was eliminated – in particular, rework that was usually due to a lack of available information at a critical time; the failure of someone to ask something; the absence of an important member at a meeting and so on. As Anderson-Brulé explained, in a traditional, linear design process (see Figure 5.2a), when someone in a team hits a problem that s/he has not anticipated, the team has to figure out how to bring the project back on track. The time between when the issue is identified and when the issue is resolved represents the rework cycle that can be represented by a series of loops. In a traditional process, the length of those loops is very large, i.e., the rework cycles take a lot of time and cost enormous amounts of money. They are horribly frustrating are emotionally draining. As a result, they create an environment where trust usually breaks down. In contrast, in a concurrent, dynamic and information-rich environment (see Figure 5.2b), the time between when an issue is identified and when the issue is resolved is very short and the impact on the process relatively minimal. The time saved may be used for creative work, developing multiple alternatives or simply, reducing the overall project schedule. The series of small loops are representative of an adaptive process, whereby professionals working on different, interdependent, yet parallel activities, continuously gather information about each other’s activities so that they may more easily mutually adjust their disciplinary solutions to one another’s and adapt to a given situation. This interactive work coordination method increases the number of feedback loops between 217 the project team members. However, the amount of “push and pull” that occurs throughout the process serves as a check-and-balance tool to keep the project on track. By interacting in real time and by being extremely knowledgeable about all aspects of the project, the King Library design team was able to identify and resolve issues quickly, thereby reducing the duration of the many rework cycle loops and thus, improving the performance of the project. There is no doubt that the ABA’s facilitation provided great value to the client in the development of the Library. However, the real performance impact on the project can be found in the fact that ABA was not only involved in the program phase, but was also, more importantly, a member of the design team. This provided continuity in terms of the knowledge that was essential for the architectural team to design a facility that met the needs of the clients. In contrast, when a design team is not involved in the programming phase (in a situation in which a client wants to assess the cost of a project before engaging the executive architect), the richness of the information that is generated during the development of the program cannot be fully communicated from the program team to the design team. Because this knowledge gap is rarely filled, the designers who later get engaged are then forced to re-interpret the programs, potentially wasting some of the resources invested at the program phase. Moreover, when programs are completely developed outside of the design process, the opportunity for clients to let their programs evolve through a creative dialogue with design activities is greatly reduced. In the King Library project, the push and pull between program and design ensured that the facility was fully aligned with the needs of the librarians. 218 Moreover, the huge amount of knowledge ABA gathered about the client environment gave the San José firm a sort of power: not a positional power, but a leadership power that Anderson-Brulé and MacBane used to operate in the background. Quietly, day after day, ABA’s facilitators and designers subtly interjected “drops” of critical information that was needed, at certain times, by the client and the design teams to move the process forward. Each “drop” saturated with rich and meaningful pieces of information they had gathered through the facilitation process was delivered at the right time when Anderson-Brulé and MacBane saw the other project participants needed it to make a decision. By remaining in the shadow of the lead architect, ABA intelligently helped the project move forward, without challenging the respective roles and positions of each participant. It is interesting to note that, while much communication occurred between the client and the design team during the design process, the librarians were somewhat “shut off” from the process during the development of construction drawings so that the designers could focus on their work without being distracted. Art Heinrich suggested that this was a good decision as it helped Career Johnson remain productive and keep the project on schedule. Participation of all project participants in all the phases may not be an effective approach. There is a limit to the amount of information that can be absorbed by each professional in a project at any given time; when this limit is exceeded, productivity may decrease. While client input in the design phase is essential to support a creative dialectic between the requirement definition and the development of the solution, it is critical that requirements be frozen when the construction drawing packages are being developed. 219 TRADITIONAL LINEAR PROJECT LIFE CYCLE CREATIVE PHASE PROGRAM DEVELOPMENT PRODUCTION PHASE FEASIBILITY STRATEGIC OPERATION PLAN SCHEMATIC DESIGN CONSTRUCTION DRAWINGS Rework Issue CONSTRUCTION Figure 5.2a: Traditional Linear Project Life Cycle CREATIVE PHASE PROGRAM DEVELOPMENT PRODUCTION PHASE SAVINGS FEASIBILITY STRATEGIC OPERATION PLAN SCHEMATIC DESIGN Rework Cycle CONSTRUCTION DRAWINGS CONSTRUCTION Client gets knowledgeable about itself Client fully prepared. New client’s personality is forming. Client “shutoff” from the process Bid Process CM engaged Dean Breivik’s late arrival Figure 5.2b: King Library Project Life Cycle 220 5.4.2. Benefits of Participants and Process Continuity As mentioned before, the early involvement of the designers in the project facilitated the alignment of the design solution with the project requirements and library operation plan. The same synergy was created when Gilbane, the construction management firm, was hired by the client prior to the bidding process. This gave the construction manager an opportunity to gather enough knowledge about the project, the client vision, and design intent to effectively manage the bidding process and later on, facilitate the transfer of this knowledge to the general contractor. Moreover, the minimal degree of discontinuity in terms of participation of the key players along the project life cycle helped to maintain the high level of shared goals among the project team members. Indeed, most of the professionals involved in the early phases were still engaged when the library was delivered to the client. Key players, therefore, knew the history of the project and the rationale behind decisions made. Such knowledge facilitated the integration of the project team. One critical individual’s late arrival, however, could have created a major disruption in the project. Patricia Breivik arrived in San José in August 1999, during the design phase, to take on her new function as Dean of the San José University Library. After reviewing the design of the librarians’ office spaces, she immediately requested some design changes. Fortunately for the project, two parameters helped to minimize the disruption. First, procedural rules had been created at the beginning of the project, which were driven by the need to manage potential conflicts between the two client organizations and to minimize the impact on the project budget. The clients had agreed on sharing the cost of all design changes that were perceived as providing a mutual 221 benefit (even if the changes were requested by one side only). However, in the eventuality that the changes were not providing benefits to one party, then the party who had originated the changes would fund them. This rule was implemented twice: first, when Breivik requested changes that did not benefit the city librarians, the university had to agree to fund the changes from its own budget; second, when the design change was requested later on by the city librarians who asked to add a Teen Center to the library. The university librarians did not consider this change a benefit to their operation: thus, the city had to find the money to fund this change. The second parameter that prevented disruptions to the project was Breivik’s own understanding of the vision of the project and her personal interest in making sure that the momentum already established would persist. A project team can only influence, but certainly not control, the behavior of its members. Whether or not a project participant will cooperate and align with the overall project vision, as opposed to emphasizing its own interest, is a critical factor that affects overall performance. AEC professionals who understand that the overall performance of a project depends on the creative negotiation process between the level of performance of a project’s subsystem and the level of performance of the overall project will be able to achieve a high level of efficiency. 5.4.3. Flexible, Decentralized Communication and Decision-Making In a traditional design-bid-build project of the size of the King Library project, communications among team members usually follow a hierarchical tree. As Dolores Montenegro mentioned, in the King Library project, all communications from and to the architect, construction manager and/or general contractor should have been addressed to 222 herself since she was the client’s project manager and was responsible for making final decisions on the client’s behalf about design changes and budget-related issues. As the central point of communication, Montenegro should have then taken the responsibility to forward each communication, with her additions, to its recipient, as appropriate. Such a centralized communication process is, of course, slow and cumbersome. To avoid having the communication flow bogged down, the project team decided to decentralize communication among the architects, the construction manager and the general contractor teams by letting them communicate with one another (either face-to-face, by phone, or email) as they deemed necessary. Face-to-face and phone conversations were then recorded in a formal transmittal sent to Montenegro for her information and the team was including her in all email correspondences. While Montenegro did not give up all her decision-making power and was still responsible for making critical decisions related to design changes and budget, the direct flow of communication among the other participants facilitated the resolution of many smaller day-to-day issues. The decision to go against the traditional communication rule in order to enhance the flexibility of the process added tremendous value to the project. The flat, flexible and decentralized communication flow increased the number of interactions among the team members and let them resolve issues in real time. It allowed them to test out ideas before final decisions were made. For instance, Jim Kligensmith, the construction manager, recalled that people were calling each other to ask questions such as “Hey! What did you mean by this?” They were able to quickly verify their understanding about an issue, instead of making assumptions that would have been too difficult or too costly to verify using a more formal process. Another type of question asked was: “I am thinking of 223 answering this question this way. Can you do this?” This question was posed to determine whether the solution to a problem from one discipline’s perspective could be implemented without creating a problem for another discipline. The effectiveness of such a flexible and decentralized communication system is not due mainly to its pure implementation, but to level of cohesion among project participants – a cohesion that was self-reinforcing and ensured team members’ cooperative behaviors and professional integrity. Without such cohesion, decentralized communication and decision-making may, in fact, create chaos. Fortunately, the King Library greatly benefited from: (1) rich relationships (i.e., great amount of trust) among participants; and (2) a balanced degree of power differentials among the professionals involved. These project characteristics are discussed in the next two sections. 5.4.4. Building the Right Project Team Chemistry Everyone I interviewed agreed with the fact that there was no “ego trip” on the King Library project. So I could not help asking: “Was the King Library project team very lucky to have assembled a group of individuals with a similar culture based on collaboration, cooperation and trust?” All participants agreed that, in the end, it all comes down to the kinds of people involved, their personalities and behaviors; it is, nevertheless, obvious that the team used different kind of strategies to influence and nurture professional and personal integrity throughout the life of the project. Strategies used were: (1) carefully choosing the team members; (2) leveraging partnering sessions as an effective communication and problem-resolution environment; (3) continuously nurturing the team and reinforcing good behavior. These are explained next. 224 5.4.4.1. Cherry-Picking Teammates It was not serendipitous that the firms chosen to work on the Library had similar cultures and attitudes. Indeed, the client took special care in building up the team by “cherrypicking” the professionals they wanted to work with toward the goal of getting the best possible firms for the task at hand. For instance, ABA was chosen by the client not only because the firm provided facilitation services, but also because Anderson-Brulé had been recognized for her skills in facilitating a community process that was built on public participation, which demonstrated “integrity as opposed to achieving someone’s unwritten agenda.”12 The client team pre-screened the general contractors that were invited to bid by looking not only at their work experience, but also, again, at their attitude towards projects and clients. Hensel Phelps was one of the best candidates. Indeed, client satisfaction, getting a sense of achievement, maintaining a good reputation, and developing the prospects for further work are Hensel Phelps’ main goals when getting involved in a project. It is not surprising that Greg Overhage, the general contractor manager on the project, described his firm’s corporate culture as being built around the following principles: [T]rying to find better ways to build the project, to save us [Hensel Phelps] and the owner money, to give them a quality project, to give them a project which is a safer project and improve the satisfaction of our employees or the owner. Similar to the way that the King Library’s client chose the general contractor firm, Hensel Phelps pre-screened the sub-contractors before hiring them and chose them not only on the basis of their price, but also on their ability and willingness to collaboratively work out issues. This method, as I heard from other project participants, might have 225 forced Hensel Phelps to “swallow” some of their own profits for the sake of a better project. Here again, one may assume that Hensel Phelps’ reputation in the industry and the quality of service it provides to its clients are two business goals that carry at least as much weight as the simple goal of making of a profit, as critical as the latter may be from a business perspective. Managing relationships at a personal level was also critical in this project. Some good working relationships had been established prior to this project, as was the case for Montenegro and Art Heinrich. Krumdieck acknowledged that as the design team was getting involved, ABA helped to “smooth the bumps” because they already knew the client team and thus, helped to “break the ice” and to facilitate the process at times when situations might have been “sticky.” In some cases, key personnel changes were made. Jim Klingensmith remembered that at times, [I]f it wasn’t the right person in the right place…the group members called each other on that and the group was willing to recognize that it wasn’t the right fit and personnel changes were made to the good of the group. It didn’t become an opportunity to say: I don’t like you! It was an acknowledgement that if there wasn’t value brought to the table and a willingness to be part of this team and if you were to persist in kind of outlaw behavior, then you should not be here. Similarly, Greg Overhage noted: The key to having a successful project is being able to get along with the key players or [at least] a majority of them. Sometimes there are just personality conflicts and that doesn’t happen. But to be successful this has to be one of the major goals. As a manager at any level, whether you are a project superintendent, or a project manager, or a vice president of the company, you have to establish those relationships so you can work through your problems. By carefully choosing the Library project firms, the project team minimized the level of professional cultural diversity, thus facilitating cohesion within the team and improving the richness of the relationships among the individuals involved. 226 5.4.4.2. Making Partnering Work At the beginning of the construction phase, the King Library team members agreed to use partnering and to have recurring sessions on a quarterly basis, while no budget had been reserved for such activities. (In contrast, in many projects, a partnering session is held once and then the team considers the activity complete and moves on.) The financial constraint forced the team members to self-manage the partnering sessions (Jim Klingensmith became the partnering session facilitator). As the Terminal project case study has demonstrated, while partnering is a technique used in most large projects, it does not always generate the desired outcome (i.e., increased cohesion between project team members and an effective platform for issues resolution). If one compares the process that took place at the King Library project to the one that took place in the previous case study, it seems that the attitude of the participants toward the process accounts for one of the reasons partnering was effective on one project and not really on the other. Self-managed partnering sessions are an unusual approach since project teams usually prefer to hire a “neutral” consultant to facilitate the process. However, the simple need for a consultant suggests that frictions between participants are expected and that the project team needs an intermediary to help facilitate the negotiation process and conflict resolution. In contrast, by using a team member to facilitate the process, King Library participants acknowledged that they would be willing to work out the issues collaboratively. Behavior rules were established at the onset of the partnering sessions: participants agreed that each professional would be allowed to speak freely about any issue and that no comments or questions (as inappropriate as they might have sounded to 227 the other participants) would remain unanswered. The behavioral rules helped to set the tone of the partnering sessions; reinforce positive behaviors; and successfully support the integration of the construction team into the overall project team. In fact, the selfimposed and recurring partnering sessions created a blueprint for how the team would conduct day-to-day business. Library participants used the partnering sessions as a communication structure among the construction team, the design team and the client representatives; it was also a platform for the subcontractors to demonstrate their talents and their abilities, as well as a place for team members to explicitly recognize each others’ contributions. As Klingensmith pointed out, the team did not consider the process as a panacea, but as an accountability tool supported by all members, which shaped the team members’ behavior. This confirms that partnering may effectively support teamwork as long as none of the parties abuses its privileges and imposes its rules on the others in the name of collaboration among professionals. When participants are not willing to cooperate, partnering becomes only a “patch” that covers a project team members’ inability to work out issues collaboratively. 5.4.4.3. Nurture the Team and Reinforce Good Behavior Montenegro pointed out that human-based emotion and interest have to be continuously fed. While there was, eventually, a high level of trust among the project members, this trust had to be earned by the participants. Klingensmith remembered that at a critical time in the project, when Hensel Phelps acknowledged that the firm had made a mistake, the client agreed to pay for it when, in fact, they were not required to do so. Thanks to such behavior, trust started to build. Trust became a springboard for mitigating and 228 expediting other issues that could otherwise have become much larger problems. Through healthy dialogue throughout the project, the team was capable of resolving differences in opinions. As Klingensmith remarked, if people have the “fortitude and the courage to do that [communicate openly]…it becomes an atmosphere of expected ability to resolve the issues.” Montenegro noticed that there was an interesting dynamic in the King Library project in the sense that each individual’s professional integrity and trust was mutually reinforcing. The team somehow created an environment in which dissent from any one of the participants was made difficult due to each participant’s high commitment to the project vision and goals. The one who would risk dissent (for instance, by protecting his own interests at the expense of the overall project’s interests) would have needed to contend with all the other project members. But no one really wants to be in such an uncomfortable situation in which he is the team’s “black sheep.” The positive environment created by the team members was self-maintained over time by encouraging people to creatively work toward resolving problems, rather than escalating problems to the level of conflict. Cooperation was particularly critical because one defector in the team could have influenced the other members to start protecting their own individual interests. The consequences of multiple defectors could have had a snowball effect, thus reinforcing negative and confrontational behaviors. 5.4.5. Balancing the Degree of Power Differentials Among Professionals The King Library team was able to successfully balance the degree of power differentials among its participants, thus facilitating the coordination of activities throughout the 229 project life cycle. The power that was exercised by the team members, at different times, was never too strong or too weak for the team to successfully overcome the different challenges that arose. While the collaborative personality of the project team helped to minimize power imbalances, enabling power structures were established to ensure that decisions would be made in an appropriate time frame. Power differentials originated from contractual agreements; procedural rules; the clear division of roles and responsibilities; the empowerment of key individuals; trust; and leadership. The marriage between the two client organizations, which, as pointed out earlier, had the positive effect of “joining them at the hip,” forced the librarians to make decisions that benefited both of their organizations. While the libraries’ marriage made the decision-making process more complex, it, nevertheless, influenced the level of cooperation between the two organizations. The financial rules agreed upon by the two clients about who would fund potential change requests were also a means to guarantee that one client would not overpower the other. Such rules created flexibility by allowing changes to be made if necessary, while ensuring that the budget would be fairly allocated between the two clients. The project team also adopted a flexible attitude toward the use of contractual documents. Greg Overhage explained that contractual documents developed during the bidding process are usually developed using very restrictive language because their ultimate goal is to enforce the terms of the contract should the general contractor need to do so. But in this project, Hensel Phelps made a point of never using the documents against the owner. Problems were worked out collaboratively. As he explained: 230 [Y]ou have some projects where the owner goes right by the letter of law and the contractor…has to do right the same things and it becomes a struggle, it becomes a battle throughout the whole job and that’s not fun anymore. Klingensmith confirmed this by explaining that in the King Library project, he might have written perhaps 20 to 30 letters over the entire project life cycle, while a construction manager might write a similar number on a daily basis in a typical $100 million job. In this project, it was the norm that no letters would be written by an individual before that individual had taken the opportunity to pick up the phone and talk to his teammate. Letters were only written as a means to confirm conversations. This attitude shaped the expected behavior and as Klingensmith pointed out: You [as a team member] couldn’t hide behind that letter-writing campaign and you couldn’t hide behind a piece of paper. You had to deal with it and you had to confront it face-to-face with a person. Such an attitude toward contractual agreements demonstrated the power of trust and reciprocity among project participants; while trust and reciprocity may not eliminate the need to develop contractual agreement in a project, it may, nevertheless, alleviate the need to enforce contracts through complex and expensive legal procedures. Moreover, trust and reciprocity facilitate the expedition of issues that would otherwise require a long, paper-based negotiation process. In addition, roles and responsibilities were very clearly defined to ensure that decisions would be made in an appropriate time frame in the eventuality that the team members were unable to reach a consensus. For instance, Montenegro was recognized by the other participants as the person who would give final say on some issues such as design changes and budget on the client’s behalf. The project team also agreed that the architects would make the final decisions regarding anything related to the external 231 aesthetic of the building. Final decisions regarding internal spaces, however, because of their critical functional aspects, were left to the librarians. Key individuals on the project had been empowered by his/her own organization so that s/he could make decisions “that stick.” For instance, individuals at Gilbane and Hensel Phelps had enough decision-making power that they did not need to constantly check in with the management of their own organizations: they could expedite the process by making a decision and move on to the next issue. Individual empowerment was critical to support real-time communication and decision-making. Without it, it would have been very difficult for the project team to minimize communication latency and delays. However, empowerment requires key individuals to have enough project management experience and skills to take on such a responsibility. Depending on the project size, decision-making responsibility may need to be delegated to lower levels in a project hierarchy, to individuals who may not have the expertise required to make the right decisions. In such a case, empowerment may have a negative impact of the project performance. Very strong leadership was demontrated by many individuals involved on the project. As Anderson-Brulé and McBane Mulford stated, there was “leadership from every step of the ladder.” Leadership changed hands along the project life cycle based on the particular involvement of a professional; the need of the project team; and the particular issue to solve at a given time. Top-level leadership was demonstrated by Hammer and Caret at the outset of the project. The vision they created inspired the entire project throughout its life cycle. The support at the highest step of the ladder was extremely critical because the project 232 complexity and political environment had a great potential for failure, which would have been a disaster for the future of San José and the university. Light demonstrated leadership when she hired ABA and asked Anderson-Brulé to be the “optimist” and to focus on the question “how.” Light clearly understood the risks involved in marring two very different library cultures and operations. She knew that she had to find a way to influence the team to think outside-the-box beyond what seemed possible. The focus on “how” forced the team to think creatively and helped them to remove roadblocks. ABA’s leadership was shown at multiple levels. By proactively re-defining the process the team would follow in the feasibility phase and in the development of the operation plan, ABA created a structure that allowed solutions to emerge as the team members interacted with one another. Leading by example, ABA taught the project team new methods of working, which improved productivity, while creating a friendly, collaborative environment. Gaining the trust and respect of the client at the early phase became critical for ABA when, later on, during the design phase, the firm was empowered by the client to provide necessary knowledge to the executive architect, thus influencing Carrier Johnson to make design decisions that did not need to be re-evaluated later on. While ABA was only a small component of the design team, the knowledge it had gained about the client gave the firm the leverage to keep the project moving forward. The team, as a whole, demonstrated leadership in many instances. Recall, for instance, their voluntary decision to use self-managed and repeated partnering sessions even though no budget for them had been allocated, thereby, improving the relationships 233 among the firms involved. The partnering sessions facilitated conflict resolution during the construction phase, while helping to maintain a sense of team building among the participants. Finally, leadership was displayed in the way team members called upon each other when they realized someone was not compatible with the work. For the sake of the project, the management of the organizations involved agreed to swap personnel as necessary. These examples demonstrate the importance of moving beyond traditional project management processes and behaviors, which are usually based on formal authority and positional power. The King Library project participants’ power was exerted through the development of a shared vision; continuous and proactive communication; collaborative problem solving; and empowerment. Such processes enabled the project team to flexibly overcome obstacles and to adapt to contingencies. 5.5. Adaptive Capacity through Innovations The analysis of the King Library project demonstrates how the project team was able to create an integrated environment for adaptation to take place. The King Library’s project vision and goals were driven from the top and reinforced by key players all along the way. The project philosophy and personality that was created by the team helped to overcome many obstacles and enabled the alignment of the goals of hundreds, if not thousands, of people involved in making the library a reality. ABA’s facilitation process provided a platform for key stakeholders to have a voice in the process. The process stimulated individual participation and good relationships between stakeholders; supported collaborative problem solving; facilitated learning; and encouraged the 234 development of a cooperative culture. Once trust was established, the participants’ positive behaviors were mutually reinforcing, thus creating an environment supportive of cooperation. The tools that were specifically developed to support the project processes ensured the integrity of the information being captured; facilitated communication; and constructive problem solving. Ground rules provided a framework within which decisions could be made effectively. The adaptive capacity of the team was exemplified by innovations at multiple levels: • Process innovation through an inclusive process that facilitated interactions between the librarians. • Programmatic innovation through the active collaboration of the librarians of both the university and the city. • Organizational innovation through the development of a new operational plan for the “two-headed” client. • Technical innovation through ABA’s initiative in developing the necessary tools to manage the knowledge gathered throughout the feasibility and operation planning phases. • Design innovation through the continuous flow of information and knowledge among members of the design team and the members’ ability to make the right decisions at the right time. These innovations were not foolproof recipes; they were devised within and for the particular context of King Library. They enabled the project team to deliver superior 235 value to the client: a facility delivered under-budget and on-schedule for a completely transformed client that was capable of fully operating the library from day one. 236 CHAPTER 6 Adaptive AEC Project Management and Leadership The Terminal and Library case studies offer a depiction of two living systems that are thriving to achieve their goals within complex and uncertain environments. Each project organization responded to its respective challenges with a different adaptive capacity that was influenced by diverse environment and organizational factors, which affected its ability to learn and self-organize. The analysis of the case studies indicates that the higher the adaptive capacity of a project organization, the more value is delivered throughout the project life cycle and the less waste is generated. Thus, one may ask, what can be learned from the implementation of adaptive practices? Complexity and living systems theories have many implications for the management of AEC projects and the evolution of the AEC industry as a whole. Therefore, this chapter is mainly prescriptive. Our goal, here, is to provide AEC professionals and project managers with a new philosophy for managing AEC project organizations in complex and uncertain environments. Indeed, as we must recall, 237 complexity and living systems theories do not provide recipes for success. They only provide a new mental model to understand project behavior and a new approach to assess complexity. The role of adaptive project management is to influence the adaptive capacity of a project organization so that it is able to self-organize and adapt to its challenges. AEC project performance and the value delivered by AEC project organizations are the emergent properties of adaptive project management. This chapter is organized as follows. In the first three sections, we discuss three critical activities for adaptive project management. First, we argue that the pattern of behavior of a project organization and its overall performance can be understood only by conducting an analysis of the project systemic structures, that is, by taking a holistic approach in assessing project complexity. Second, in contrast to conventional project management whose goal is to bring a project organization back to equilibrium when the project activities are threatened by disturbing forces, adaptive project management focuses on unleashing the creativity that exists within the project organization. Third, project complexity and uncertainty imply that the strategic fit between a project organizational structure and the project context cannot be designed at the planning phase; the fit is an emergent property of the strategies and actions undertaken by a project organization as it self-organizes. Therefore, adaptive project management needs to facilitate dynamic contingency and planning. In Section 4, we apply the adaptive capacity concept to the AEC industry as a whole. First, we discuss the role of adaptive leadership for supporting AEC projects’ adaptive challenges. Second, we discuss the adaptive skills that AEC professionals and 238 firms need to develop in order to improve their adaptive capacity. Third, we look at the implications of living systems theory for educational and professional institutions; indeed, the AEC Industry as a whole may overcome its adaptive challenges only when adaptive practices are institutionalized. Finally, we discuss the need to redefine AEC contracting models in order for these models to better support adaptive project management. Section 5 discusses the implications of the analysis of AEC project organizations from a living systems perspective on further research. Finally, Section 6 provides some considerations on the potential value that may be delivered by adaptive AEC project organizations. 6.1. Assessing and Embracing Project Complexity Considering a project organization as a complex system embedded within its ecosystem implies that the project organization’s behavior is influenced positively, as well as negatively, by the interactions among the project’s subsystems. The influence is positive if the project organization is able to grasp the intricacy of the relationships among the diverse subsystems and devise strategies to facilitate mutual adjustments among them. For instance, the simultaneous interactions among programming, development of the operation plan and design at the King Library project facilitated the alignment of disparate project requirements and helped the project team to resolve conflicts in a timely manner. The interactions among a project’s subsystems may also influence the project organization negatively. Recall, for instance, how the complex nonlinear interactions 239 among leadership, late program changes, fast track schedule and economic changes created non-anticipated events with which the Terminal project organization had to cope in order to move forward. Consequently, a critical role for adaptive project management is to make sense of the structural complexity of the project environment and get insights on the many variables that may affect performance. By doing so, a project organization can anticipate issues early and proactively develop strategies that facilitate adaptation. To analyze the relationships between all the diverse project subsystems, a holistic project management approach must be taken. One may use many different approaches to assess project complexity. Three methods are discussed below: (1) using “what if” scenarios and/or story-telling for analyzing the environment; identifying potential disrupting events and their potential consequences; and facilitating decision-making and mitigating risk; (2) scanning the environment for discontinuities and identifying conflicts all along the project life cycle; and (3) keeping decisions open as long as possible, also known as deferring decisions until “the last responsible moment” in Lean Project Delivery System (Ballard 2000, p.5), to facilitate mutual adjustments among a project’s subsystems, thereby improving performance and increasing value to clients. 6.1.1. Using “What If” Scenarios and/or Storytelling While project performance may be influenced by dozens or more variables mutually influencing each others, the mapping out of such a complex performance systems is a virtually impossible task for the human mind. Indeed, it is well known that humans cannot effectively hold more that two or three variables in their minds simultaneously. 240 Techniques such as the use of scenarios at the planning phase provide an effective tool for anticipating adaptive challenges and mitigating risks. The building of scenarios aims at providing an organization with a greater understanding of the environment within which decisions are made and the impact of these decisions on the future of the organization. In The Art of the Long View, Peter Schwartz (1991) describes how scenarios can help an organization to assess the impact of its actions; understand better the environment in which actions will take place; and identify the forces, trends, attitudes and other influences, which will impact the fit of the actions taken over the long-term. As Schwartz (1991, p.xiv) points out, [t]he point [of scenario building] is to make strategic decisions that will be sound for all plausible futures. No matter what future takes place, you are much more likely to be ready for it – and influential in it – if you have thought seriously about scenarios. The use of scenarios may be a critical AEC project management activity and a powerful tool for analyzing and mitigating risks and investigating potential project roadblocks or contingencies. One may argue that the Terminal client team, should it had used scenarios and story-telling during the planning phase, could have better anticipated events such as the change in the airport leadership and the consequent program changes; economic changes and their impact on construction activities; and, finally, the coordination requirements created by distributing work among three design firms involved in the design and construction of the International Terminal, in addition to Terminals G and A. 241 6.1.2. Scanning the Environment for Discontinuities and Conflicts Second, AEC professionals need to become skilled at continually scanning their project environment for potential discontinuities. By remaining alert to and focusing on what may go wrong, as opposed to exclusively focusing on achieving pre-established performance goals, a project organization may anticipate and/or detect events early and thus, will be more prepared for initiating an adaptive response. The project will be able to proactively plan and update the plan as project conditions change, thereby continuously learning through feedback loops. Focusing on discontinuities requires AEC professionals to be willing to expose and deal with issues and conflicts when they arise and take the risk to temporarily destabilize a seemingly smooth project operation. Some professionals may hesitate to do so because of the potentially negative reactions from other project participants who may resist the need to question the existing process or project direction for fear of jeopardizing their performance. Moreover, current contractual arrangements and professional liability often deter professionals from interfering with a process, even though it may benefit the overall project. However, the failure to illuminate the existence of conflicts may lead to two negative consequences: first, important issues may be overlooked, which may amplify a problem down the road; second, limited attention to adaptive challenges limits a project organization’s opportunity to innovate and consequently, deliver increased value to a client. Thus, adaptive leadership – that is, the willingness of project participants to challenge the status quo and force a project organization to migrate away from its equilibrium – is a critical professional skill in complex project environments. (The role of adaptive leadership is discussed in more detail later in this chapter.) 242 6.1.3. Deferring Decisions until the Last Responsible Moment Since one cannot optimize the performance of a given subsystem without impacting the performance of the other subsystems in the system and interactions among subsystems are not linear and often take place within different timeframes (as the Terminal project demonstrated), a focus on short-term performance can only fail. Because professionals cannot know the impact of one decision may be over time, project organizations need to keep decisions open until the “last responsible moment” within the projects’ constraints. Doing so gives AEC professionals a chance to discover all major interdependencies and resolve interdisciplinary conflicts early, thus preventing unnecessary expenditures that often occur with the late discovery of problems that require changes. As the interactive process used by ABA and King Library client demonstrated, AEC professionals who work simultaneously on diverse project subsystems should facilitate a creative dialogue among the subsystems so that mutual adjustments among the project subsystems take place over time as the project team gains a better understanding of the issues at stake. When a decision has to be finalized, professionals need to be cognizant of the fact that the decision may have unexpected long-term consequences that may create future adaptive challenges for other subsystems. 6.2. Dynamic Contingency and Planning In the preface to the first edition of Strategic Organizational Diagnosis and Design: The Dynamics of Fit, Burton and Obel (2004, p.xviii) state that “[o]rganizational design is a normative science with the goal of prescribing how an organization should be structured in order to function effectively and efficiently.” The paramount theory of organizational 243 design – contingency theory – is based on the idea that there is no one best structural form of organization under all conditions and environments. Organizational performance depends on the fit between both the structure and the context; ultimate, the design of an organization should encompass the definition of the particular structure that is best adapted to a specific context (Galbraith 1977; Mintzberg 1979). Organizational theory provides the tools to diagnose whether a particular organizational structure fits a given organizational context and thus, whether an organization will be able to effectively and efficiently achieve its strategic goals. When there is an incompatibility between organizational structure and its context, organizational theory may be used to design (i.e., prescribe) a more adapted organizational structure. Contingency theory assumes that, based on an assessment of the characteristics of an organization’s context and of its existing structure, it is possible to prescribe a new organizational structure. This theory may be effective for organizations operating in relatively stable environments; however, AEC project organizations are one-of-the-kind complex adaptive systems that operate in highly dynamic environments. Traditional planning and rational decision-making strategies are limited in their effectiveness in such contexts. The strategies, structures and processes defined by a project organization at outset, during the early planning phase, quickly become obsolete as the project evolves and the environment changes. As we may recall, the strategies that the Terminal client team had thought to be the best ones at the planning phase (e.g., give work to as many design firms as possible in times of economic recessions) in fact generated coordination problems at a later stage. Similarly, the preliminary structure of King Library’s client team was not well-adapted to the task of transforming the attitudes and perceptions of the 244 librarians. However, as the librarians started to better understand the type of problem they had to solve and as ABA influenced the development of the adaptive process, the Library’s client team started to take actions that transformed its communication processes and organizational structure. Eventually, a new joint-identity emerged, which was not anticipated by the librarians. Thus, the fit between a project’s structure and its context cannot be predesigned; the fit is an emergent property of the strategies and actions taken by the project organization as it learns and self-organizes. In other words, while traditional contingency theory focuses on designing organizational fit, a dynamic contingency theory focuses on improving the adaptive capacity of the project organization (by influencing the control parameters): the fit emerges as the project organization learns and self-organize. Therefore, in a dynamic environment, learning and adaptation leads to self-organization (structural changes); self-organization, in turn, leads to the development of a good fit between the organizational structure and its environment, thus improving organizational performance. Focusing on the capacity for learning and adaptation, as opposed to the fit between structure and environment, is especially critical in AEC project settings, because of the intrinsic heterogeneity of these projects. Indeed, since project organizations typically consist of networks of firms, it is likely that different project participants will perceive the fit between the project structure and its environment differently based on their level of development, past experience, sensitivity to environment’s characteristics, professional goals and so on. In similar situations, different individuals or organizations may react differently. Consequently, what may be considered to be a critical alignment 245 component for one firm may not even be noticed by another firm. However, in spite of their differences, project participants need to progressively influence the control parameters toward the required control points, thereby increasing the adaptive capacity of the project organization. Yet, the control points themselves are not static and evolve based on the project context. Therefore, influencing the control parameters toward the required control points needs to be an on-going, iterative process that takes advantage of continuous learning; a focus on learning allows a project organization to continuously adapt to change and update its strategy as new situations require it. In complex and uncertain environments, organizational fit is a characteristic that is always at risk of becoming obsolete. Organizational adaptive capacity, on the other hand, is a characteristic that maximizes an organization’s chances of evolving and ultimately, surviving crises. While the fit is limited in time, adaptation is a continuous process that allows an organization to reach new states of order. By focusing on improving its adaptive capacity, a project organization is improving its chances of successfully managing contingencies and maximizing its overall performance. 6.3. Developing the Adaptive Capacity of AEC Project Organizations The assessment of the control parameters (i.e., the rate of information flow; agents’ schemas diversity; the richness of connections; the level of contained anxiety; and the degree of power differential) that was performed in each case study supports the hypothesis that they need to reach critical points in order for a project organization to creatively respond to unforeseen events and overcome adaptive challenges. As we may recall, in the construction phase of the Terminal project, ineffective communication 246 processes; the lack of trust; divergent professional cultures and values; and the inability of some participants to hold their level of anxiety – as demonstrated by the tension in the relationship between the general contractor and construction manage – were factors that influenced the project team’s inability to flexibly re-organize as needed when disruptions such as program and economical changes arose. Faced with external pressures, the project organization lacked the creativity, leadership and contractual structures required to cope with its environment. The lack of adaptive capacity of the project organization was costly for the client, as well as for the professionals involved, both in monetary and psychological terms. In contrast, the King Library project team was able to develop a project culture that was based on collaboration, cooperation, trust and creative problemsolving. There was enough diversity in the library technical teams for different perspectives and ideas to be considered when discussing the library’s emerging operation plan. The rules and protocols that were defined about who would make final decisions on diverse types of issue provided some structure in case of conflict; the rules, however, did not impede the team from reaching most decisions through negotiations and consensus. The level of reciprocity among the professionals was high and facilitated the resolution of issues in a timely fashion. Improving the adaptive capacity of a project organization requires a completely new project management philosophy based on living systems theory. Such a philosophy adopts a holistic approach towards project management and considers learning as one of the most important factors influencing project performance. Consequently, adaptive project management should focus on transforming a project organization as an integrated 247 communication network. To do so, a project organization first needs to diagnose its adaptive capacity and then devise strategies to improve it as necessary. 6.3.1. Diagnosing the Adaptive Capacity of a Project Organization Similar to a physician who questions her patient about physical symptoms in order to diagnose the patient’s level of illness, project organizations need to inquire about the level of each control parameter in order to assess their adaptive capacity. Such an assessment may be done at the early phases of the project, as well as on a regular basis, along the project life cycle. Since the control parameters are influenced by organizational variables (i.e., the causes of the symptoms), it is through the analysis of these variables that a project organization will determine its adaptive capacity. Diverse methods may be used to qualitatively and/or quantitatively assess the levels of the control parameters. Four possible methods are discussed below. First, Figures 6.1a through 6.1e present a list of questions project organizations may consider for building a diagnostic of their adaptive capacity. The questions are designed around the understanding of the influences of typical organizational variables on the control parameters. (Organizational variables’ influences were identified in Chapter 3 through a review of the literature on organizational theory.) The list of questions provided is not meant to be exhaustive; it is a starting point for project organizations to start reflecting on their own environment and adaptive capacity. Second, based on a review of organizational theory literature, Burton and Obel (2004) have developed a computer-based expert system that organizations and/or consultants may use to assess an organization’s structural fit. The expert system includes 248 a knowledge base for building a diagnostic of organizational fit to context and design rules in order to develop a multiple-contingency model. The system offers recommendations for the design of new structures that are better aligned with the organization environment. One could take a similar approach for assessing the adaptive capacity of an organization. A knowledge base of influence rules may be designed based on the impact of diverse organizational variables on the level of the control parameters. An example of influence rules may be: “If the level of interactivity among project participants is low, then the rate of information will be low.” Such rules may be built, based on the review of organizational theory and tested with case studies for accuracy. The objective of such a knowledge base is to provide information about the level of the control parameters; it would not be designed to prescribe a particular organizational structure. However, one may be able to offer general recommendations on how a project should be managed to influence the control parameters. A word of caution needs to be made here: each project context is different and there is no recipe that can be applied from one project to another without being re-evaluated, based on the specific project context. (This limitation is further discussed in the next section). Third, a control parameter such as the rate of information flow may be assessed by using organizational modeling techniques. Jin and Levitt (1996) have developed a computational model of Virtual Design Teams (VDT) based on information processing and contingency theory perspectives. In VDT, design teams are modeled as “information-processing structures that are composed of tasks generating information to be processed, actors processing and communicating information, communication tools linking actors for communication, and an organization structure that constrains actors’ 249 information-processing and communication behavior” (Jin & Levitt 1996, p.174, emphasis as shown is theirs). The VDT simulations only model routine tasks. In order to be used to model project organizations facing complex adaptive challenges, the model needs to be modified to include non-routine project teams’ activities. Fourth, social network analysis are used to uncover the patterning of people’s interactions and to visually map network activities that are usually invisible. They may be used to map and measure the relationships and flows among people, groups and organizations. In the last few years, many computer programs have been designed to facilitate social network analysis and visualization.1 For instance, IKNOW, a program developed by Contractor, Zink and Chan (1998) can assist the study, creation, maintenance and dissolution of knowledge networks. IKNOW can help an organization to visualize the patterns of knowledge and information flow through the organization’s informal network. InFlow, developed by a company called Orgnet.com, may be used to assess the location of knowledge in a network; workforce diversity; leaders’ influences; power relationships; influences on decision-making; and so on.2 Thanks to the above methods, a project organization may gain a good understanding of its adaptive capacity at a give time. The next step is to improve such a capacity, if necessary. 6.3.2. Influencing Control Parameters From the diagnostic of its adaptive capacity, a project organization may be able to determine which control parameters need to be influenced and which organizational variables have the greatest impact. Influencing control parameters is not an exact 250 science; it is more like an art. When using this process, one needs to carefully consider the context of the project organization and particular constraints. Not all organizational variables may be changeable. Moreover, change may not happen overnight: it takes time to build a project organization’s culture and influence adaptive behaviors. Individuals and organizations need time to absorb new methods of working. Trust is slow to develop. Transforming a project organization into an integrated communication network requires tinkering and experimentation, as well as the evaluation of the outcome of experimentation, since strategies that may be implemented to increase learning will not always deliver the expected results. Project teams should not be afraid of making mistakes. Learning organizations are the ones that encourage experimentations and recognize that mistakes are not only unavoidable, but also necessary to the learning process as long as those mistakes are not repeated over time (e.g., the auto-industry leader, Toyota, which has implemented a lean production system, is an example of a company that has embedded learning practices in all its activities).3 Failures may in fact provide very useful information about the sensitivity of the organization to changes. There may be many reasons why a project organization is resistant to the pursuit of an adaptive path: • Changes may have been made too quickly for individuals to have an opportunity to cope with the pace; • • • The benefits of adaptive work may not be well-understood; Defensive behaviors may counteract efforts to transform the organization; and The project’s legitimate system may not be supportive of an adaptive structure. 251 Control Point _ + Rate of Information Flow Organizational variables Examples of assessment questions* - Project structure allows for integration of downstream activities - Project management procedures allow for open communications - People have the information they need to do their job, when they need it - People have an opportunity to interact with one another when they need it - People meet face-to-face or via teleconferences on a regular basis - People are overwhelmed by too much information - When issues are raised, the response is (1) extremely slow or (2) prompt and appropriate - Web-based technology is used to support project communication - Level of connectivity and interactivity - Rigid, hierarchical structure versus flat, network structure - Project delivery method - Informal communication network - Information bottleneck - Information quality - Information technologies - Information management system - Asynchronous/synchronous communication - Feedback delay - Formalization, i.e., type of communication/process procedures (rigid versus flexible) - Mutual adjustment - Others, as identified by further research * Questions may be assessed on a scale of 1 to 5 unless otherwise specified. Figure 6.1a: Assessing the Rate of Information Flow 252 Control Point _ + Agents’ Schemas Diversity Organizational variables Examples of assessment questions* - Project team members share the same project vision - Professionals share the same professional culture - The project involves participants from different cultures and languages - There is some homogeneity among the firms involved in the project in terms of size, power, scope of work, and level of influence - Project rules, protocols, standards, and explicit rules of conduct have been defined between the team members and agreed upon - All required project stakeholders are involved and participate to the process - The project team keeps the external community engaged and well informed - Diversity of culture, beliefs, norms, and values - Diversity of language - Diversity of practice - Types of organizations and institutions involved - Professionals’ personalities - Project culture - Professionals’ differences in term of codes of ethic - Experience in working in a heterogeneous environment* - Degree of individualism - Goal alignment - Shared vision - Level of agreement on project rules, protocols and standards - Explicit and implicit rules of conduct - Others, as identified by further research * Variables that apply to the individual firms and professionals involved in a project and not to the overall project organization * Questions may be assessed on a scale of 1 to 5 unless otherwise specified Figure 6.1b: Assessing Agents’ Schema Diversity 253 Control Point _ + Richness of Connections Organizational variables Examples of assessment questions* - Communication with participants is (1) difficult with frequent disagreement or (2) open, honest, and free-flowing - Concerns and problems are acknowledged (1) only when they cannot be ignored or (2) at first sign - Concerns and problems are (1) swept under the rug or (2) dealt with quickly and directly - Cooperation between participants is (1) non-existent or (2) characteristic of all phases of work - Project participants follow a “one team” approach - Project participants are respectful of one another - Project participants exhibit a positive attitude and have fun - Project participants hold each other accountable - Nature of personal and organizational ties (i.e., personal, arm’s-length ties, weak, strong) - Cooperation - Trust - Reciprocity - Integration - Self-interest* - Altruism* - Sharing of strategic information - Rivalry - Defection - Transparency of action, clear talk and straightforward actions - Tolerance for paradox - Others, as identified by further research * Variables that apply to the individual firms and professionals involved in a project and not to the overall project organization * Questions may be assessed on a scale of 1 to 5 unless otherwise specified Figure 6.1c: Assessing the Richness of Connections 254 Control Point _ + Level of Contained Anxiety Organizational variables Examples of assessment questions* - Project participants are willing to raise issues and take risk - Project participants are willing to engage in transformative work - Some project participants are willing to/able of helping their teammate through crisis and uncertainties - Project participants find the project goals and activities meaningful - A sense of purpose supports the project team when the project environment become incoherent and chaotic - Some project participants are able to exercise adaptive leadership - Project participants are willing to help each other resolve issues, even when their own activities are not directly impacted - Project participants accept constructive feedback and are willing to let others challenge their own ideas - Project participants exhibit defensive behaviors when their own activities are threatened - Mistakes are not considered as professional/individual failures - Internal tensions - Level of frustrations* - Ability to hold ambiguities and paradoxes* - High level of integration, mutual control - Routines, standard procedures - Level of conformity to professionals norms - Defensive mechanisms, resistance to changes* - Quality of professional relationships - Individual and team self-reflection - Leadership - Methods used to exercise power - Others, as identified by further research * Variables that apply to the individual firms and professionals involved in a project and not to the overall project organization * Questions may be assessed on a scale of 1 to 5 unless otherwise specified Figure 6.1d: Assessing the Level of Contained Anxiety 255 Control Point _ + Level of Power Differentials Organizational variables Examples of assessment questions* - Some project participants/firms are more skilled than others at influencing others in their choices or decisionmaking process - Project leaders influence other to (1) make decisions that support their own vision or (2) define for themselves what need to be done - Some professionals/firms have technical skills and expertise that would be very difficult to replace if needed - Everyone in the project feels that everyone else is needed to complete the job - When a critical decision has to be made there is always someone in the project who has the final decisionmaking power - Most decisions are made through consensus - Budget is always the main criteria taken into consideration when making critical decisions for the project - Distribution of power - Level of interdependence - Ability to influence others* - Exercise of power - Position in the organization* - Roles and division of labor - Reputation in the industry* - Economic base* - Technological influence* - Expertise* - Trust - Legitimacy* - Others, as identified by further research * Variables that apply to the individual firms and professionals involved in a project and not to the overall project organization * Questions may be assessed on a scale of 1 to 5 unless otherwise specified Figure 6.1e: Assessing the Level of Power Differentials 256 Moreover, the adaptive capacity of a project organization will greatly depend on the adaptive capacity of its members. Indeed, as we discussed in Chapter 2, a complex adaptive system is a nested system in which components interact with one another at multiple levels in the system. In an organization, the adaptive capacity of members – i.e., their collaborative behavior; their ability to trust one another; their ability to handle stress; their ability to learn; their approach toward influencing others (e.g., via adaptive leadership versus control and coercion); their method of control; and so on – directly influences the organization’s adaptive capacity. Thus, influencing the adaptive capacity of the organization and of its members may have to be done simultaneously in order to be mutually reinforcing. Therefore, prioritizing which specific organizational variables and control parameters need to be influenced does not matter because project variables are interdependent and interact with one another. For instance, while focusing on the design of an inclusive facilitation process, ABA also modified the client organizational structure through the creation of technical teams. The teams were not integrated at first. However, after numerous meetings that provided the members with the opportunities to interact and collaborate with one another, positive and optimistic behaviors developed and trust started to emerge between the librarians. As the collaboration between the librarians gained some momentum, the process became self-reinforcing as positive feedback increased the learning capability of the team. Momentum, however, can act against an organization as was the case in the Terminal project, which was the victim of unbalanced negative feedback loops. As we may recall, the differences of personalities between the general contractor and the 257 construction manager impeded good communication; this, in turn, influenced the level of trust between the two individuals and diminished the team’s ability to resolve issues in a timely fashion. Unresolved issues in the field impacted the project schedule and resulted in the need for reworking – a situation that increased the level of worker dissatisfaction and, to some extent, undermined work safety. When a project organization is the victim of such negative inertia, changing the patterns of behavior may require drastic actions such as replacing an individual or stopping the process. The context should also be a critical consideration in shaping the adaptive capacity of a project organization. Applying the same recipe from one project to another will not work. What may be an effective strategy for one project may fail in another one. For instance, partnering was a highly successful process for the Library project as it increased the cohesiveness between team members. Conversely, the process had very little impact on team building in the Terminal project because partnering was only a strategy implemented without considering the critical factors: the control parameters. Focusing on the control parameters, as opposed to relying on standard management recipes, will help a project organization to focus on the components that have a real impact on a project adaptive capacity. The focus on context should be an important consideration for those who want to facilitate transformative processes or lead an organization through its adaptive challenges. Indeed, as these individuals gain experience in different project contexts, they may feel inclined to always utilize the same strategies and methods, irrespective of the new organizational context. Such an approach is akin to the application of standard procedures in single-loop learning. Standard procedures are only effective when the 258 environment is stable and does not change. Every project is unique, however, and leaders need to develop their own learning and creative capability so that a fresh look at each project situation is taken and new strategies are developed that facilitate organizational learning in different contexts. 6.4. Building the AEC Industry’s Adaptive Capacity For the AEC industry to develop its adaptive capacity will require changing the inner programming that exists within individuals, professional organizations, as well as educational, professional and legal institutions. However, beliefs, culture and values systems changes are rarely top-down driven. One does not transform an entire industry overnight, but one can certainly influence its members who will then become agents of change. Adaptive skills are first developed at the individual level. As individuals and professionals start interpreting the world from a living systems perspective, their mental models will change. A new mental framework will drive new behaviors. Over time, professionals will embed these new models into their own practices. Eventually, these practices may be institutionalized within firms and professional institutions. In order to build the AEC industry’s adaptive capacity, the following components will need to be in place: first, AEC professionals and other project stakeholders need to become adaptive leaders, i.e., facilitators of change process. Adaptive leadership is particularly critical to the facilitation of the resolution of non-technical problems, i.e., problems that do not have an obvious technical solution, but require individuals to adopt living systems philosophy and revise their own mental models accordingly. Second, to support adaptive leadership, AEC professionals and firms need to acquire critical 259 adaptive skills. Third, educational and professional institutions need to embrace living systems principles and embed them within their institutional frameworks in order to fully support the adaptive capacity of the entire AEC industry. Fourth, new contractual models need to be developed that are better aligned with the adaptive project management philosophy. Each of these components will be discussed individually in greater detail. 6.4.1. Adaptive Leadership The term, “leadership,” has taken different meanings in the literature: leadership often refers to people who hold top management positions; it may also refer to individuals who are able to define a vision and have enough authority (not necessarily related to position) to influence others to follow their vision. Whether the vision is good or bad is seldom questioned when assessing this type of leadership. In contrast to the above definitions, adaptive leadership is not value-free. Adaptive leadership relates to the ability of an individual to influence a community to face its problems (Heifetz 1994). In this definition, the leader is not the one providing the vision or solution to a problem. Rather, the leader is the one who is able to mobilize a community to define its own values for itself, enable it to identify the goals that support those values and implement the actions that need to be taken to achieve them. Both San José city mayor and San José state university president exercised conventional leadership based on their positional power. They defined the vision of the joint library and created the concept of seamless services. While such a leadership was critical to initiate the process, it was not sufficient to rally the librarians around the concept and help them to overcome their adaptive challenge. On the other hand, SJPL 260 director, Light, and the architectural firm, ABA, demonstrated adaptive leadership by influencing the client team to face the reality of the situation and engage in a transformative process, whose outcome redefined the library’s identity. In Leadership Without Easy Answers, Heifetz (1994) discussed at length the concept of adaptive leadership; how leadership is exercised; and the type of skills required for an individual to act as an adaptive leader. Borrowing from Heifetz’s (1994) ideas, the role of an adaptive leader within the context of a project organization includes: • Identifying adaptive challenges: An adaptive leader is able to identify an adaptive challenge when other project participants only see a technical problem.4 Faced with uncertainties and complex problems, individuals and groups have a tendency to turn to experts or to individuals with authority, hoping that these knowledgeable and/or powerful people will provide the right answers and lead them to the right solutions. Unfortunately, knowledge and power are insufficient for overcoming adaptive challenges because these challenges require a transformation of the values and beliefs of the entire organization or community involved – that is, they require learning and creativity. An adaptive leader is able to grasp the whole complexity of the project organization and of its ecosystem and point out the gaps between, on the one hand, the internal values, norms and structure of a project organization and, on the other hand, the organizational context. • Stirring and disturbing a project organization by creating a sense of urgency: An adaptive leader can create a sense of urgency by asking meaningful and tough questions that challenge the beliefs, norms and values of the project stakeholders. 261 In other words, an adaptive leader is not afraid of disrupting the existing organizational equilibrium and pushing the project organization to the edge of chaos. In doing so, s/he will need to pay attention to the adaptive capacity of the project organization in order to ensure that s/he neither pushes too hard nor too little. If s/he pushes too hard, the project organization may be too overwhelmed and may disintegrate; if s/he pushes too little, the project organization will remain in the equilibrium state. • Designing and facilitating an adaptive process for supporting organizational learning and providing structure and stability to counteract instabilities created by the adaptive challenge: An adaptive leader needs to be ready to cope with the outcome of instability that his/her questioning may have created. To prevent the disintegration of the project organization, the leader needs to provide some level of stability, for instance, by designing a process that provides sufficient structure and discipline for the team members to progressively tackle their issues. Process design and management may encompass the design of participative structures to facilitate the development of connections between project participants; the definition of communication protocols and/or rules of conduct; the design and facilitation of decision-making processes; the creation of information management systems; and the orchestration and management of conflicts. • Holding, to some extent, the level of anxiety of the project team: As project participants are forced to confront the reality of a situation, their level of anxiety and stress may rise. An adaptive leader should be skilled at holding team members’ level of anxiety, at least for a while, until the team is able to cope with 262 it. If the leader holds team members’ level of anxiety for too long, members may become disengaged from the process and will not face their own problems. Thus, the leader needs to empower team members to act and encourage them to take responsibility for their own actions. Adaptive leadership is a subtle skill that requires much practice to learn and courage to pursue. As Pascale et al. (2000) argue, one needs to behave like a gardener, as opposed to a mechanic, when faced with an adaptive challenge. A good gardener does not water her plants because her schedule indicates she must do so. A good gardener examines the plants to assess their specific conditions; determine whether fertilizer and/or pesticide are required; and assess the quantity of water to use. The plants’ treatment is always personalized and tuned to what a particular plant needs at a given time. Similarly, adaptive leadership requires a deep understanding of the specific context of the project and needs. Moreover, a leader must always be cognizant of the project organization’s adaptive capacity at any given time and act accordingly. To complete this discussion on adaptive leadership, we add some considerations on the role project stakeholders may have in exercising adaptive leadership. First, one needs to emphasize that everyone in a project, regardless of his/her profession, role, level of responsibility and authority can exercise adaptive leadership. Whether an individual is a professional hired by a client, a member of the client team or of the extended community, or a member of an agency, he or she may take an adaptive leadership role. However, some professionals, because of their training, skill and position within a project may be better positioned for influencing a project organization to embark on an adaptive journey. Our consideration, here, only focuses on the role of architects. Indeed, based on 263 their central position in a project and their role as coordinator, architects have an immense potential for adding value through adaptive leadership. Educated as generalists, architects have a variety of interests and skills, which allow them to get involved in many different aspects of the building enterprise, which are not only related to technical problems, but also cultural, economic and political issues. Due to their training, architects are skilled in analyzing their clients’ requirements and environments; seeing the interrelationships between the different building sub-systems; and grasping the complexity of a project ecosystem. Thus, they have the potential for identifying adaptive challenges. The greatest opportunity for identifying an adaptive challenge and disturbing a project organization may be at the planning and program development phase when the client defines his/her real needs. Indeed, this is the time when a client has an opportunity to reflect on the purpose of its organization, its organizational value, and the role it plays in the community. It is also the time when the project stakeholders can influence and shape the vision and direction of a project; the organizational structure; the coordination processes; the communication protocols; and the behaviors of the participants. In situations where a client or project stakeholders have to go through a transformative process, as was the case for the Library project’s client, the early involvement of architects may facilitate the adaptive journey. Often, however, clients hire architects when the program is already defined. This, unfortunately, limits the potential impact architects could have in ensuring that the program and requirements have been mindfully defined, i.e., that the client has fully considered all the diverse aspects of the project (social, economic, usability, sustainability and so on). When 264 architects are involved only at the design phase, it is too late to challenge the basic assumptions and values that were at the basis of the program development without jeopardizing the project operations and schedule Moreover, the early involvement of architects in the planning process may save a project precious time since the architects can both facilitate the definition and resolution of issues and give shape to the physical aspect of the building simultaneously. Moving back and forth between problem identification and problem resolution creates an opportunity for greater integration between requirements and design solutions. (Recall the value delivered by ABA at the design phase when facilitating knowledge transfer to the executive architect.) Similarly, the continuous active involvement of architects throughout the construction phase – not only in answering requests for information from contractors, but also in participating in the production process design and control – would provide a greater opportunity for learning, especially in terms of the integration of design and construction activities. 6.4.2. Adaptive Skills for AEC Professionals and Firms As AEC professionals and firms begin to embrace living systems principles for managing project organizations, they will need to develop new skills. We discuss four critical skills here: (1) systems thinking; (2) facilitation and process management; (3) self-reflection; and (4) emotional intelligence. 265 6.4.2.1. Systems Thinking Systems thinking is a necessary skill for making sense of a complex project environment and comprehending how multiple variables interact with one another via feedback loops. Systems thinking provides the tools to assess recurring patterns (e.g., virtuous or vicious cycles) and balancing processes. By visualizing such processes and patterns, one may gain insights on what component, strategy, or process needs to change so as to stimulate or decrease the pattern. Thinking in terms of feedback loops requires our mind to be trained in circular thinking, as opposed to linear thinking (Senge 1990). Since for most of us, circular thinking is non-intuitive, this may require some training and practice.5 6.4.2.2. Facilitation and Process Management Skills In order to facilitate and manage the adaptive process, adaptive leaders need to develop the following skills: • Inquiry is a skill that will help professionals to frame and ask questions that challenge a project organization and force its members to face critical issues. It is an important skill to learn since professionals are not necessarily trained in challenging the basic assumptions and values held by project communities. By learning to ask “tough” questions that help to unravel the real issues and interdependencies between project subsystems, professionals will improve their ability to identify adaptive challenges. • Teambuilding is a skill that facilitates the development of relationships among project participants and the cultivation of trust, which are critical to adaptation. 266 • Play within some structures maintains a positive balance between control and freedom of thinking and action. While some professionals consider play to be counterproductive, it is critical to the creative process. • Information management facilitates knowledge sharing and decision-making processes. It is an essential component of interactivity and disciplinary integration. • Holding anxiety will help professionals to create the space in which other project participants can develop their adaptive capacity. In order to be able to hold anxiety without becoming overwhelmed, professionals need to be capable of selfreflection and emotional intelligence (discussed below). • Conflict management ensures that divergences of opinion do not generate animosity and impede creativity and learning. The aforementioned facilitation skills do not depend on technical expertise. Facilitation and process management skills are not usually part of professional training, whose focus is on developing professional technical expertise for resolving technical problems. However, facilitation and process management, which aim to help a project organization cope with its adaptive challenges, require professionals, at least for a time, to suspend their technical expertise and authority to focus on their learning and adaptive skills.6 The ability to suspend their reliance on their technical expertise is counterintuitive to professionals who have been trained to resolve issues with their technical skills. Adaptive challenges, however, cannot be overcome solely with the application of technical expertise. This means that, in some cases, expertise needs to be shared with laymen involved in a project. While the sharing of authority between professionals and 267 experts, although often problematic, is nevertheless inherent in any projects, the sharing of authority with communities of non-experts may be more controversial. However, public participation and client involvement are absolutely critical for overcoming adaptive challenges. 6.4.2.3. Self-Reflection Self-reflection, the hallmark of living systems and human society in particular, is necessary for a project organization to transcend divergent value systems. When professionals are not able to reflect on their own project environment, they cannot gain an understanding of the complexity of the interactions among the diverse project’s subsystems. Without self-reflection, there is no transformation that can take place, no learning. When professionals are active participants in a process, they often get lost in the tumult of the multiples activities taking place because they are focusing their attention on the immediate actions; as a result, they lose sight of the whole picture. Professionals need to learn to stand back and reflect on the entire project ecosystem in order to analyze the whole, as opposed to the parts. Self-reflection is necessary for a project organization to migrate to the edge of chaos when maintaining equilibrium is not the appropriate strategy. Self-reflection is a capacity that needs to be learned at the individual, as well as the organizational levels. Because our beliefs influence our perceptions of the world, we need to examine our assumptions and beliefs to ensure we do not hold internal contradictions. Similarly, a project organization needs to examine the fundamental assumptions at the basis of a project’s vision, goals, power structures and so on in order 268 to assess whether these assumptions are aligned with project participants’ tacit understanding of what the vision, goals and power structures should be. By making assumptions and beliefs explicit, contradictions will arise. Identifying these contradictions will allow the project organization to gain a deeper understanding of the issues at stake. Reflection-in-interaction will create an environment where anxiety can be released and trust developed as issues get resolved. While the drive for efficiency usually translates to the elimination of slack time, AEC professionals need to learn to take the time for self-reflection at the individual and group levels. 6.4.2.4. Emotional Intelligence Emotional intelligence is a set of non-cognitive capabilities, competencies and skills, which influences our ability to cope with environmental demands and pressures (Goleman 1995, 1998). It is both an intrapersonal (emotional) and interpersonal (social) skill “that determine[s] how effectively we understand and express ourselves, understand others and relate with them, and cope with daily demands…Ultimately, being emotionally and socially intelligent means to effectively manage personal, social and environmental change by realistically and flexibly coping with the immediate situation, solving problems and making decisions” (Bar-On 2005, p.4).7 Emotional intelligence has been shown to be directly related to personal and business performances.8 Because of this relationship, emotional intelligence may be one of the most important skills for AEC professionals to acquire in order to successfully manage AEC project organizations at the edge of chaos. 269 By improving their emotional intelligence, AEC professionals will be better capable of holding divergent ideas and opinions and ask tough questions without generating inter-professional conflicts. Emotional intelligence will help them to sustain simultaneous cooperation and competition, which are required for creativity and learning, while maintaining control when faced with high levels of uncertainties. It will also facilitate the management of anxiety as project organizations migrate to the edge of chaos, thus minimizing the risk of organizational disintegration. 6.4.3. Implications for Educational and Professional Institutions In the first edition of Architecture: The Story of Practice, Diana Cuff (1991, conclusion chapter, pp.251) points out the need to “closely align the institutions of the architectural profession with practice itself” and challenge the myth of the “primacy of the independent practitioner working with relative autonomy.” She notes that within architectural schools, such a myth is bred in the studio, which does not sufficiently emphasize the social and collaborative aspects of design. She argues for the improved preparation of students to engage in collaborative practice by challenging them to deal with more complex and therefore, more realistic design problems. Furthermore, Cuff (1991) also emphasizes the need for training future architects in the area of “sense making” by including social, economic and political complexities of buildings in studio projects and developing students’ self-reflective capacity. Finally, she calls for the need to include business and management courses with architectural curriculum in order to reemphasize the synergy between good business practices and excellent design. 270 Given the fact that Cuff (1991) made her recommendations more than a decade ago, it is interesting to discuss whether they have been implemented by architectural and engineering schools. In the following section, the following questions will be addressed: Have Cuff’s (1991) recommendations been addressed by architectural and engineering schools? What additional recommendations should we make for schools to prepare AEC students to become adaptive professionals and leaders? An effort to embed collaborative practices into curriculum can be seen, in the last few years, in the emergence of interdisciplinary courses that bring together students from architecture, engineering and construction management schools, as well as business schools (in some cases). In such courses, students work in interdisciplinary teams and have an opportunity to collaborate around a particular design problem; identify interdisciplinary design conflicts; negotiate design constraints; manage design costs and schedules, while responding to project requirements. Such exercises provide glimpses into the complexity and uncertainty, which are typical of AEC projects. Yet, students get most often involved in these collaborative courses with no theoretical background related to complexity theory and with no understanding of the behavior of adaptive complex systems. Because of this lack of theoretical framework, they cannot fully analyze their own behavior, as well as the behavior of their teammates, and the influence these behaviors have on their studio projects. As architectural and engineering schools begin to include sustainability courses into their curriculum (sustainability being a science grounded in systems theory), one may expect students to be introduced to the basic concepts of systems theory and to become familiar with the nonlinear interactions of components in a complex system. In addition, to facilitate the understanding of the 271 concepts of adaptation and self-organization, computer simulations should also be used to help students to simulate the behavior of complex systems. In addition to acquiring theoretical knowledge, students need to acquire the social skills necessary for becoming learners and adaptive leaders. They need to become skilled at successfully negotiating design solutions with their peers and using the process of creative abrasion as a means of increasing learning and innovation, as opposed to creating conflicts. Some exercises that test cooperative behaviors and reciprocity include teambuilding exercises or exercises adapted from the “Prisoner Dilemma” (Axelrod 1984). They can help students to better assess their personal cooperative skills and understand that self-interest and defensive behaviors, which impede the achievement of high-level individual and organizational performance, could have been reached via cooperation. As young AEC professionals begin their professional activities, they need to be supported by professional institutions that re-emphasize the benefits of adaptive practices. As we discussed in Chapter 3, AEC mental models have been shaped through the process of professionalization of architects. However, the traditional definition of professionalization as a specific form of domination of one set of experts over another is at odds with the living systems philosophy and adaptive practices. To be effective and meaningful, the American Institute of Architects (AIA), as well as any other professional associations in the AEC industry, must be representative of the society in which AEC professionals operate. As society evolves and new needs arise, professional institutions need to lead the transformative journey to redefine the character of the AEC professions. 272 Today, AEC professional institutions need to acknowledge that the complexity of the built environment requires adaptive AEC practices based on collaboration, cooperation, trust, learning and self-organization. However, since institutional change is a slow process that requires the destruction of previous formal beliefs, models and norms, as well as the creation of new ones, changes will be more likely to be influenced by the AEC professionals who have embraced living systems theory. These professionals have the responsibility and the right to initiate change in their professional institutions so that the professional body of knowledge becomes aligned with the living systems philosophy; the body of knowledge, once institutionalized, can, in turn, influence the entire AEC industry. Examples of changes that need to be made by professional institutions are: revising project delivery methods; revising the project management handbook to include adaptive management process; updating the professional code of ethics and professional conduct as necessary; working with legal institutions to redefine contractual models (this is discussed in more detail in the next section); update continuous professional training curriculum to include complexity and living systems theories and concepts; devise means to facilitate interdisciplinary collaboration; encourage members to experiment with new models of practice and get involved in research projects with universities. 6.4.4. Aligning Contracting Models with Adaptive Management Practices A living project organization is an integrated communication network that thrives on knowledge sharing, cooperation and trust. It requires flexible contracting models that support and facilitate organizational learning, self-organization, innovation and the 273 generation of value for the project participants, as well as for the project organization as a whole. Conventional contracting models, however, limit cooperation and innovation (Matthews & Howell 2005) and impede adaptive project management in various ways. First, contracts are designed to support sequential delivery methods – methods that prevent interdisciplinary collaboration and a holistic approach to project management. Consequently, contracts do not encourage organizational learning and interfere with the generation of value. Second, current contractual approaches are adversarial in nature. Their language is often centered on remedies and penalties for noncompliance. Such restrictive language reinforces the lack of cooperation and trust inherent in many projects. Third, contracts are often focused on the specific division of responsibilities and scope of work. They are designed for one single discipline and thus, do not support multidisciplinary coordination and collaboration around value engineering. Consequently, professionals are driven to defend their turf at the expense of both the client and other AEC professionals. This emphasis promotes local optimization, as opposed to overall project performance. Therefore, legal institutions need to work with AEC professionals and owners to include adaptive project management practices in terms of contracts to produce more cooperation, efficiency and fewer disputes. Adaptive project organizations require the flexibility to extend professionals’ scope of work based on incremental learning and the need to increase value to the client. Thus, contracts should be structured to facilitate incremental deliveries that may not have been defined in the contract itself. Contractual agreements should facilitate collaborative and cooperative relationships built on trust. For instance, relational contracting models 274 (e.g., project partnering agreements, team-based incentives, reward incentives and so on) that are becoming more common in the AEC industry are written to support the coordination of the work of integrated project teams. In these cases, participants are considered equal in terms of the expertise and knowledge they bring to a project and the risks they share (Colledge 2005). Some attempts at developing new contractual models that aim at improving project performance through adaptive management are already being pursued. For instance, Sutter Health, a not-for-profit, community-based healthcare and hospital system headquartered in Sacramento, California, has developed a contracting model to support lean project delivery. The Sutter Health project organization is defined as a design and construction community in which all participants, construction managers, general contractors and trades included participate in all the project phases so as to encourage the sharing of information early in the process. The project team is experimenting with Target Value Design – a similar, yet broader process than Target Costing – that encompasses additional design criteria beyond cost (e.g., time, work structuring, constructability and other similar issues). One of Sutter Health’s contracts goals is to reduce overall project risks by creating a system of shared risks among project stakeholders. Sutter Health has also eliminated the traditional “negligence” standard as the measure of the designers’ financial responsibility. Instead, Sutter Health negotiates a deductible as a percentage of construction costs for “error & omissions,” even those resulting from negligence that Sutter will fund out of the design contingency. The contracts also establish an escalating series of dispute resolution meetings in order to maximize the opportunities for party-controlled dispute resolution. Finally, Sutter Health 275 has developed an incentive program to reward exceptional performance to overcome inertia.9 Further experimentation with innovative contracting agreements must be pursued to support living project organizations’ activities. The law needs to be put in place not only to protect professionals in case of liability, but also to promote collaborative strategies that are based on adaptive practices. Without a synergy between adaptive practices and contractual agreements, adaptive project organizations will not be able to reap the benefits of the living systems approach. 6.5. Living Systems Theory’s Implications for Further Research In this section, we briefly point out further academic research that should extend our theoretical and empirical knowledge of AEC adaptive project organizations. Potential research includes: (1) refining the adaptive performance model defined in Chapter 3; (2) developing computer simulations to model adaptive project organizations; facilitate the analysis of the behavior of AEC project organizations in complex environments; and support experimentations with the implementation of new adaptive processes; and (3) pursuing interdisciplinary research between business schools, law schools and design and engineering schools to increase our understanding of the influence of living systems philosophy on management and law within the context of the AEC industry. In Chapter 3, we have defined a new AEC project performance model based on complexity and living systems theory. This model needs to be tested on case projects and refined so as to gain a better understanding of the influence of organizational variables on the control parameters. The new knowledge may be used to build an expert system that 276 is similar to the one developed by Burton and Obel (2004), which was discussed earlier in this chapter. The expert system could be used by AEC project organizations to assess their adaptive capacity and provide recommendations on how they may improve it. Computational models of adaptive project organizations could be developed that would facilitate the testing of hypotheses related to the influence of organizational variables on the control parameters. Using such models, one could simulate the adaptive behavior of a project organization and assess the impact on project performance. Finally, there should be much synergy between research in the fields of complexity and living systems theory, organizational theory and law within the context of AEC project design and construction. Indeed, as one conceives of a project organization as a living system and takes into account the diverse organizational variables influencing the organization’s adaptive capacity, one will recognize the inevitable need for new contracting arrangements. Different types of contracting arrangements will dictate the extent to which an organization is able to self-organize and adapt to contingency. Through a better understanding of the mutual interactions between organizational theory and law, one will be able to align contractual agreements with adaptive management practices in the AEC industry. 6.6. Delivering Value through Adaptive Project Management Throughout the 21st century, the AEC industry will face the challenge to develop solutions to highly complex problems. Those problems may be due to limited resources; the need to deliver sustainable, energy-efficient and ethical built environments (e.g., built environments that do not place one community at a disadvantage for the benefit of 277 another); the necessity of taking into account the diverse needs of disparate communities; and so on. Such problems will require technical solutions. Yet, most of the challenges will be for AEC project organizations to deliver adaptive solutions, that is, solutions that require a transformation of the value, beliefs and norms of the communities involved, AEC professionals included. These problems will require individual and organizational learning and creativity. Living systems theory teaches us that creativity is a by-product of life: only living things can be creative. Considering project organizations as living systems implies that project organizations should provide an environment where creativity can thrive. Creativity is a social process that operates from the tension between opposite forces, which maintain an organization at the edge of chaos. The outcome of creativity is the development of new social structures that are continuously generated through circular communications between project participants. This means that values and beliefs held within a project community are not static and unchangeable; rather, they should evolve through the interactions between project stakeholders. Thus, one should no longer perceive divergences between project stakeholders’ goals as a threat to project performance; instead, we should consider divergences as critical elements feeding the project team creativity, as long as cooperative and mutually-supportive behaviors and trust are present. The network of conversations between project participants should help each one gain a deeper understanding of the different worldviews and opinions by making assumptions explicit. Tradeoffs, when they occur, should not be perceived as a loss, but as a means of reaching a higher level of order for the overall project. By tapping into the collective intelligence that resides within project organizations and extended 278 communities, and by continuously negotiating between diverse perspectives, project organizations will increase their chance of resolving complex problems in a way that adds value to the entire project community. The value that may be generated through the application of living systems principles and adaptive project management goes far beyond the efficient delivery of a physical artifact (i.e., a facility delivered to its client, on time, on budget and meets specifications). The value is multifold. It includes the emergence of new project stakeholders’ values, beliefs, goals and performance measures; the development of a project organization identity that facilitates integration and cooperation; new adaptive processes that minimize waste and facilitate learning; and, finally, a facility that reflects the emergent vision of the project stakeholders. Indeed, there is a synergy between adaptive processes and the product being created. As the identity, culture and vision of a project organization emerge from the interactions among stakeholders, one may expect the organizational characteristics to influence the design of the final products. Thus, one may wonder what the impact of the application of living system theory and adaptive project management will be on the built environment. What architectural principles will emerge out of adaptive project behaviors? Will the built infrastructure be more adapted to clients and communities’ needs and achieve a better integration between physical spaces and usage? Will architectural design be more sustainable? There is yet much to learn about the implications of living systems theory on the management of AEC project organizations. Nevertheless, as we watch living systems evolve and creatively adapt to their environment, one should be confident that AEC 279 project organizations will also be able to deliver sustainable built environments efficiently and with minimum waste. Therefore, to overcome our adaptive challenges, we need to embrace the living systems approach. 280 NOTES Chapter 1: Creating Value in AEC Project Organizations A striking example of the effects of an unpredictable economy on the construction market is the impact of the terrorist attacks on the U.S. on September 11, 2001 on – among many other things – architectural commissions. The October 2001 issue of the electronic magazine DesignIntelligence (www.di.net) pointed out that many non-profit projects had to be cancelled because their funding depended on individual and corporate donations – donations that had been given to the victims of the attack instead, and that nonprofit organizations would probably not be able to raise enough capital to pursue the improvement or expansion of their facilities. Many retail and hospitality projects were cancelled because they were considered too risky by developers in that economic context. For more information on Foster’s innovative design, see Lambot, I. & Davis, C. (eds.) 1997, Commerzbank Frankfurt, Prototype for an Ecological Highrise, Birkhauser, Basel. A prototype 4D CAD modeling software developed at Stanford University’s Center for Integrated Facility Engineering (CIFE) was used to develop what-if scenarios in the Frank O. Gehry Associates project of the Walt Disney Concert Hall in Los Angeles, California. For more information, see Goldstein, H., 2001. Maestros of design and construction render a virtual masterpiece [online]. Available from: http://www.construction.com/NewsCenter/it/features/01-20010502.asp [Accessed May 2005]. 3 2 1 281 Howell (2001) defines “value” as “meeting the purposes of the client and producers.” Conversely, “waste” is considered as “anything not required to deliver value.” Thus, this definition extends beyond the waste of materials to include defects in products; overproduction; large inventory waiting to be consumed; unnecessary processing; unnecessary movement of people; unnecessary transport of goods; time lost waiting for people or equipment to complete an activity; products designed to wrong specifications or not meeting customers’ needs. As Kuhn (1996, p.111) points out, “Examining the record of past research from the vantage of contemporary historiography, the historian of science may be tempted to exclaim that when paradigms change, the world itself changes with them. Led by a new paradigm, scientists adopt new instruments and look in new places. Even more importantly, during revolutions, scientists see new and different things when looking with familiar instruments in places they have looked before. It is rather as if the professional community had been suddenly transported to another planet where familiar objects are seen in a different light and are joined by unfamiliar ones as well. Of course, nothing of quite that sort does occur: there is no geographical transplantation; outside the laboratory, everyday affairs usually continue as before. Nevertheless, paradigm changes do cause scientists to see the world of their research-engagement differently. In so far as their only recourse to that world is through what they see and do, we may want to say that after revolution scientists are responding to a different world.” In Newton’s home country, England, the belief that universal laws could explain all the mechanisms of the universe was very appealing to the society of the time, which was torn apart by disputes between Catholics and Protestants and monarchists and Republicans. The Laws provided everyone – no matter his religious beliefs or political convictions – with a means to agree on a universal “truth.” For more details on the implementation of self-organization and adaptive strategies in the United States Army, see Pascale et al. (2000). Chapter 2: Complexity Theory and Living Systems The 3-body problem is the following: given 3 bodies (e.g., Sun, Moon, and Earth) and their initial positions and velocities, the problem is to determine the motion of three bodies attracting one another according to Newton’s law of gravity. While the 2-body problem had been solved by Newton, the solution to the 3-body problem had not yet been found. Nevertheless, in line with the deterministic way of thinking, it was believed that given enough information, the 3-body problem could be solved, at least in theory. “A deterministic nonlinear feedback system is a network of agents whose behaviors are determined by a common schema [i.e., rules] consisting of a few rules that are fixed over time, apply to all agents without exception and do not have to do with achieving some 2 1 7 6 5 4 282 purpose. It follows that agents do not adjust their behaviors in the light of their consequences for a particular purpose” (Stacey 1996, p.51). 3 In the late 1950s, the French mathematician Benoit Mandelbrot began to study the geometry of a wide variety of irregular natural phenomena, which, he realized, all exemplify property, which he called “fractal.” Over the next ten years, Mandelbrot invented a new type of mathematics to describe and analyze fractals. In an attempt to answer the question, “Could a machine be programmed to make a copy of itself,” Von Neumann stated that a system able to reproduce itself requires two fundamental things: 1) a program that encodes a description of the system and can be executed to reproduce itself; and 2) the ability to copy the description of the system to the offspring so that the offspring can also reproduce itself. He proved this was the case in at least one cellular pattern that he demonstrated could reproduce itself. For more on cellular automata, see Waldrop 1992, pp.211-240. To visualize the behavior of cellular automata, see: http://math.hws.edu/xJava/CA/index.html (accessed June 2004). The Website includes applets with which a user can play to observe the behavior of one-dimensional automata at the edge of chaos. 6 7 5 4 The Gaia hypothesis in www.oceanonline.com/gaia.htm (accessed April 2005). The term “holarchy” is synonymous to “hierarchy.” However, since the term “hierarchy” is often used to denote rigid human organizational structures of command and control, the term “holarchy” will be preferred when talking about complex systems. For a lengthier list of paradoxes, see Stacey 1996, pp.97-98. 8 9 The concept of fitness landscape first introduced by biologist Sewall Wright in 1932 has been developed further by Stuart Kauffman (1993) as he studied self-organization and evolution at the Santa Fe Institute. This survival strategy is similar to the concept of “simulated annealing” – a technique to find a good solution to an optimization problem by trying random variations of the current solution. 11 10 See Whitaker in Autopoietic Theory: http://www.unikk.ch/course/seiten/lesson3.1.htm (accessed November 2004). Chapter 3: Managing AEC Projects at the Edge of Chaos 1 The idea of a “global brain” has been around for some time now. It was mentioned in the work of 19th century evolutionary biologist Herbert Spencer; science fiction writer 283 H.G. Wells in his book, World Brain; and by the revolutionary mystic Pierre Teilhard de Chardin who introduced the term “noösphere,” meaning the global domain of the mind. Nonaka and Konno (1998) have formulated a process that organizations can use to foster knowledge creation. Their process is based on the concept of “ba.” A ba is a shared space or platform (physical, virtual, or mental) in which individual and/or collective knowledge can be created. A ba is created when knowledge is transferred from one individual to another, so that both persons can share a common idea. Knowledge creation evolves through a four-step process (socialization, externalization, combination and internalization), in which knowledge passes through different stages of being tacit and explicit. Socialization consists of the sharing of knowledge between individuals through joint activities. Externalization consists in the conversion of tacit knowledge into comprehensible forms that can be understood by others. Combination consists in converting explicit knowledge into a more complex and inclusive set of explicit knowledge. Internalization is the process by which explicit knowledge is used by individuals to create their own tacit knowledge. In a survey of large and small companies whose management was based on an organic approach, Lewin and Regine (2001) define three different paradoxes: (1) The paradox of allowing is direction without directives, freedom with guidance; (2) The paradox of accessibility is leaders are mutual but not equal – mutual in respect and ability to affect and be affected by others, but also not equal in power; (3) The paradox of being attuned is knowing and not knowing – knowing intuitively while not knowing everything. 4 3 2 Douglas North (1992, p.9) defines institutions as the “formal rules, informal constraints (norms of behavior, conventions, and self-imposed codes of conduct), and the enforcement characteristics of both.” The phenomenon of path dependence in the economy was demonstrated by the economist Brian Arthur (1990). The demonstration that positive feedback occurs in economics was an astonishing discovery in complete contradiction to the neo-classical economists’ belief that the economy was driven by equilibrium, i.e., that resource allocation was based on the rational behavior of self-interested agents whose actions trigger negative feedback, thus driving the economy to equilibrium. Positive feedback in the economy explains why, for instance, inferior technologies such as the standard QUERTY keyboard layout or VHS videotape format have prevailed over superior competitors. Let’s consider the case of VHS. In the mid- 1970s, VHS competed with another technology called Beta, which had been assessed by experts as slightly superior. However, chance dictated that VHS vendors gained a larger market share in the beginning. Because video stores did not want to stock VCRs in two formats and customers did not want to be stuck with an obsolete VCR, VHS market share increased. The slight advantage that VHS had over Beta grew quickly until VHS became the standard format. This phenomenon is called, in economics, increasing returns. The example illustrates the importance of path-dependent processes, i.e., that historical 5 284 accident, plus increasing returns, can play a critical role in the locking-in of a particular outcome over another that may have rationally been preferable. Axelrod (1984) uses an electronic version of the famous “Prisoner’s Dilemma” game to demonstrate his findings: “In the Prisoner’s Dilemma game, there are two players. Each has two choices, namely cooperate or defect. Each must make the choice without knowing what the other will do. No matter what the other does, defection yields a higher payoff than cooperation. The dilemma is that if both defect, both do worse than if both had cooperated.” In other words, defection yields a higher payoff than cooperation if it is chosen by only one party. Cooperation by both parties yields a moderate return for both, while defection by both parties yields the lowest return for both parties. Among all the programs proposed for the Computer Prisoner’s Dilemma Tournament, the entry TIT FOR TAT won each round. TIT FOR TAT’s strategy is very simple: the strategy starts with cooperation and thereafter does whatever the other player did on the previous move. This strategy shows that cooperation based solely on reciprocity seems possible. Arie de Geus’s argument is based on the works of psychologist D.W. Winnicott (Playing from Reality) from the Tavistock Institute, American teacher/writer John Holt (How Children Learn) and the Media Lab’s Seymour Papert (Mindstorms: Children, Computers, and Powerful Ideas) based at the Massachusetts Institute of Technology. Leith Star and Griesemer (1989, p.393) define boundary objects as “…[S]cientific objects which both inhabit several intersecting social worlds and satisfy the informational requirements of each of them…They are weakly structured in common use, and become strongly structured in individual site use…They have different meanings in different social worlds but their structure is common enough to more than one world to make them recognizable, a means of translation.” (emphasis theirs). Chapter 4: International Terminal at the SFO Airport See more details at SOM’s website: http://www.som.com//resources/projects/1/3/1/printPreview.html (accessed June 2005). I do not have information about the formation of the general construction team and cannot elaborate further on the reasons why Tutor-Saliba Corp. entered in a joint venture with two other construction firms. One hypothesis is that the project was very large and Tutor-Saliba Corp. needed additional resources to complete the job. They may have also tried to share the risks with the two other firms. Tutor-Saliba Corp., nevertheless, remained the leader of the construction team. The concept of frozen accident in the economy has been studied by Brian Arthur (1990). In contradiction to the thinking in neo-classical economics, which was based on the assumption that the rational behavior and actions of self-interested agents trigger 3 2 1 8 7 6 285 negative feedback, thus driving the economy to equilibrium, Arthur (1990) demonstrated that we can find substantial evidence of positive feedback in the economy. 4 In contrast to a traditional cost-management approach that focuses on satisfying a customer’s requirements and considers a product’s cost as a dependent variable that is the result of the decisions made about a product’s functions, features and performance capabilities, target costing builds upon a design-to-cost approach. The latter approach focuses on market-driven target prices as a basis for establishing target costs. The traditional cost management approach is linear and commonly drives products’ costs to be higher than desired. Target costing is iterative and proactively works to achieve target cost during product and process development. For more information abut these two different cost management approaches, see article by K. Crow (2002): http://www.npdsolutions.com/target.html (accessed July 2005). In lean management, the process related to dividing up work and assigning it to different professional firms is called “work structuring.” This process may have added great value if it had been implemented in the Terminal project. 6 5 The role of adaptive leadership is discussed in more detail in Chapter 6. While adaptive leadership can be exercised by anyone in a project organization, not matter his role and/or position, some individuals, because of their specific role and position may have a better chance to succeed in such endeavor. One of the adaptive leader’s main roles is to shake the project inertia and influence the project team to creatively adapt to a situation. See owner’s comments and general contractor’s comments about schedule in partnering form, pp. 5-6. 8 7 See owner’s, designers’ and general contractor’s comments on the quality issue in the partnering form, p. 7. See Chapter 3, p. 125 for a more detailed explanation of the partnering process. 9 Probably due to the difficulty of managing such a complex project and to inherent conflicts between firms, the construction manager on the project changed many times. John Draguesku, whom I interviewed, was the third person in the construction management seat for the project. His late arrival created much disruption as indicated by the drop in the satisfaction level of the general contractor in the charts included in the partnering report. 11 12 10 See Transmittal package dated July 20, 1998. See Transmittal package dated July 20, 1998. 286 Chapter 5: San José Martin Luther King Library 1 2 There are examples of joint libraries in Europe. For more details about the history of the relationship between the city and San Jose State University, see the article “How city, SJSU made peace to become library partners” in San Jose Mercury News, August 10, 2003. The total cost of the project had been funded using the following repartition: the State funding for California State University (CSU) capital projects was 48.5 percent of the total budget ($86 million); the City contribution was 39.4 percent ($70 million); private fundraising contributed 9.3 percent ($16.5 million); and the University fund provided 2.8 percent of the total budget ($5 million). “Awkward building is efficient but boring,” in San Jose Mercury News, August 10, 2003. 5 6 4 3 “Shared Goals, Shared Building,” in San Jose Mercury News, August 10, 2003. For stories on the opposition to the project, see “Public gets look at library design,” in San Jose Mercury News, September 2, 1998; “Protesters target joint-library plan,” in San Jose Mercury News, October 15, 1998. While firms were linked by contractual relationships that ensured their involvement in the project, individuals might have been asked to be replaced in situations in which they did not “fit” in the project team. Indeed, there was replacement of some individuals in the project. “Shared Goals, Shared Building,” in San Jose Mercury News, August 10, 2003. Anderson-Brulé, in interview. 7 8 9 The expression “hard on the issue, soft on the people” is borrowed from Hirshberg (1998, p.168). 11 12 10 Anderson-Brulé’s description. From Anderson-Brulé’s description of ABA’s facilitation process. Chapter 6: Adaptive AEC Project Management and Leadership A list of social network analysis computer programs can be found at: http://www.insna.org/INSNA/soft_inf.html (accessed May 2005). 1 287 2 For more information about InFlow, go to: http://www.orgnet.com/index.html (accessed May 2005). For more information about Toyota’s lean production system, see Liker (2004). 3 4 Difference between adaptive challenges and technical problems are discussed in Heifetz (1994, pp.73-76) Techniques for mapping systemic structures are presented in Senge (1990). For more information of adaptive leadership skills, see Heifetz (1994). 5 6 7 See Bar-On’s paper: http://www.eiconsortium.org/research/baron_model_of_emotional_social_intelligence.pdf (accessed May 2005). 8 See Cherniss, C. ‘The business case for emotional intelligence’, Paper prepared for the Consortium for Research on Emotional Intelligence in Organizations, Available from: http://www.eiconsortium.org/research/business_case_for_ei.pdf (accessed May 2005). Fore more information on Sutter Health’s contracting model, see Lichtig (2005). 9 288 REFERENCES AIA California Council. 1996, Handbook on Project Delivery, Santa Clara. Allen, J. P. 1994, ‘Mutual control in the newly integrated work environments’, The Information Society, vol. 10, no. 2, pp. 129-138. Arthur, W. B. 1988, ‘Self-reinforcing mechanisms in economics’, in The Economy as an Evolving Complex System, eds P. W. Anderson, K. J. Arrow & D. Pines, AddisonWesley, Reading, pp. 9-33. ______________1990, ‘Positive feedbacks in the economy’, Scientific American, vol. 262, no. 2, pp. 92-99. Axelrod, R. 1984, The Evolution of Cooperation, HarperCollins Publishers, New York. Baldwin, C. Y. & Clark, K. B. 2000, Design Rules, Volume 1: The Power of Modularity, The MIT Press, Cambridge. Ballard, G. 2000, ‘Managing work flow on design projects’, in CIB W96 Architectural Management, Atlanta, 19-20 May. ____________2000, ‘Lean project delivery system’, Lean Construction Institute White Paper-8 [online]. Available from: http://www.leanconstruction.org [Accessed June 2005]. ____________2000, ‘Positive vs. negative iterations in design’, in Proceedings Eighth Annual Conference of the International Group for Lean Construction, IGLC-6, 17-19 July, Brighton. Bar-On, R. 2005, ‘The Bar-On model of emotional-social intelligence’, Issues in Emotional Intelligence, [Online], vol. 1, no. 4, pp. 1-28. Available from: 289 http://www.eiconsortium.org/research/baron_model_of_emotional_social_intelligenc e.pdf [Accessed June 2005]. Bartindale, B. 1998, ‘Public gets look at library design’, San José Mercury News, 2 September, p. 4B. _______________2003, ‘Shared goals, shared building’, San José Mercury News, 10 August, p. 3, 21. Burns, T. & Stalker, G. M. 1961, The Management of Innovation, Tavistock Publications, London. Burton, R. M. & Obel, B. 2004, Strategic Organizational Diagnosis and Design: The Dynamics of Fit, 3rd edn, Kluwer Academic Publishers, Groningen. Capra, F. 1996, The Web of Life: A New Scientific Understanding of Living Systems, Anchor Books, Doubleday, New York. ____________2002, The Hidden Connections: Integrating the Biological, Cognitive, and Social Dimensions of Life into a Science of Sustainability, Doubleday, New York. Cherniss, C. 2005, ‘The business case for emotional intelligence’, paper prepared for the Consortium for Research on Emotional Intelligence in Organizations, Rutgers University, [Online], Available from: http://www.eiconsortium.org/research/business_case_for_ei.pdf [Accessed May 2005]. Cole, C. 1993, ‘In microbial microcosm’, In Context: A Quarterly of Humane Sustainable Culture, [Online], Available from: http://www.context.org/ICLIB/IC34/Margulis.htm [Accessed April 2005]. Colledge, B. 2005, ‘Relational contracting – Creating value beyond the project’, Lean Construction Journal, [Online], vol. 2, no. 1, pp. 30-45, Available from http://leanconstructionjournal.org [Accessed July 2005]. Contractor, N., Zink, D. & Chan, M. 1998, ‘IKNOW: A tool to assist and study the creation, maintenance, and dissolution of knowledge networks’, in Community Computing and Support Systems, ed. T. Ishida, Springer-Verlag, Berlin, pp. 201-217. Cross, R. et al. 2004, ‘Six myths about informal networks – and how to overcome them’, in Creating Value with Knowledge: Insights from the IBM Institute for Business Value, eds. by E. Lesser & L. Prusak, Oxford University Press, New York, pp. 47-60. Crow, K. 2002, ‘Target costing’, [Online], DRM Associates, Available from: http://www.npd-solutions.com/target.html [Accessed July 2005]. Cuff, D. 1996, Architecture: The Story of Practice, MIT Press, Boston. De Geus, A. 1997, The Living Company: Habits for Survival in a Turbulent Business Environment, Harvard Business School Press, Boston. Doherty, P. 2001, ‘A rose by any other name: will nested procurement replace design build?’ DesignIntelligence [Online], Available: 290 http://www.greenwayconsulting.com/DesignIntelligence/DI_RoseByAnyOtherName _article.htm [Accessed October 2001]. Dostoglu, S. 1986, ‘On the fundamental dilemmas of architecture-as-profession’, in O.D.T. . Mimarlik Fakültesi Dergisi Cilt 7, Sayi 2 Bahar. Fruchter, R. & Swaminathan, S. 2005, ‘Reflection-in-interaction’, SID2005 Social Intelligence Design Workshop, March, Stanford University. Galbraith, J. R. 1977, Organization Design, Addison-Wesley, Reading. Ginsberg, S. 2000, ‘Stuck in holding pattern’, San Francisco Business Times, [Online], 8 September, Available from: http://sanfrancisco.bizjournals.com/sanfrancisco/stories/2000/09/11/focus1.html?page =1 [Accessed June 2005]. Gleick, J. 1988, Chaos: Making a New Science, Penguin Books, New York. Goldstein, H. 2001, ‘Maestros of design and construction render a virtual masterpiece’, [Online], Available from: http://new.construction.com/NewsCenter/it/features/01-20010502.jsp [Accessed February 2002]. Goleman, D. 1995, Emotional Intelligence, Bantam Books, New York. ______________1998, Working with Emotional Intelligence, Bantam Books, New York. Granovetter, M. S. 1973, ‘The strength of weak ties’, American Journal of Sociology, vol. 78, no. 6, pp. 1360-1380. Green, S. D. 1999, ‘The dark side of lean construction exploitation and ideology’, Proceeding IGLC-7, 26-28 July, University of California, Berkeley. Guinier, L., Fine, M. & Balin, J. 1997, Becoming Gentlemen, Women, Law School, and Institutional Change, Beacon Press, Boston. Hammer, M. 1990, ‘Reengineering work: don’t automate, obliterate’, Harvard Business Review, July-August, pp. 104-112. Harman, W. W. & Sahtouris, E. 1998, Biology Revisioned, North Atlantic Books, Berkeley. Haviland, D. 1994, The Architect’s Handbook of Professional Practice, 12th edn, The American Institute of Architects, Washington D. C. Heifetz, R. A. 1994, Leadership Without Easy Answers, The Belknap Press of Harvard University Press, Cambridge. Heisenberg, W. 1927, ‘ ber den anschaulichen Inhalt der quantentheoretischen cinematik und mechanik’, Z für Phys., vol. 43, pp. 172-198. Hess, A. 2003, ‘Awkward building is efficient but boring’, San José Mercury News, 10 August, p. 4, 23. 291 Hirshberg J. 1998, The Creative Priority: Driving Innovative Business in the Real World, HarperCollins Publishers, New York. Hock, D. W. 1999, Birth of the Chaordic Age, Berrett-Koehler Publishers Inc., San Francisco. Howell, G. A. 1999, ‘What is lean construction’, Proceedings IGLC-7, 26-28 July, University of California, Berkeley. _________________2001, ‘A new operating system for construction: a call for reform’, AEC Vision, [Online], Available from: http://aeccafe.com/AECVision/opinion/new_os.php [Accessed October 2001]. Hugues, E. 1963, ‘Professions’, Daedalus, vol. 5, no. 3, pp. 655-668. Jarvenpaa, S. L. & Ives, B. 1994, ‘The global network organization of the future: information management opportunities and challenges’, Journal of Management Information Systems, vol. 10, no. 4, pp. 25-57. Jin, Y. & Levitt, R. E. 1996, ‘The virtual design team: a computational model of project organizations’, Computational & Mathematical Organization Theory, vol. 2, no. 3, pp. 171-195. Johnson, S. 2001, Emergence: The Connected Lives of Ants, Brains, Cities, and Software, Scribner, New York. Kauffman, S. A. 1991, ‘Antichaos and adaptation’, Scientific American, August, pp.7884. _________________1993, Origins of Order: Self organization and Selection in Evolution, Oxford University Press, Oxford, England. _________________1995, At Home in the Universe: The Search for the Laws of SelfOrganization and Complexity, Oxford University Press, New York. Koestler, A. 1967, The Ghost in the Machine, Arkana Books, London. Kogut, B. & Zander, U. 1996, ‘What firms do? Coordination, identity, and learning’, Organization Science, vol. 7, no. 5, pp. 502-518. Kuhn, T. S. 1962, The Structure of Scientific Revolutions, The University of Chicago Press, Chicago. Lambot, I. & Davis, D. (eds.) 1997, Commerzbank Frankfurt, Prototype for an Ecological Highrise, Birkhauser, Basel. Langton, C. G. (ed.) 1989, Artificial Life: Santa Fe Institute Studies in the Sciences of Complexity, Proceedings Vol. 6, Addison-Wesley, Redwood City. Langton, C. G., Taylor, C., Farmer, J. D. & Rassmussen, S. (eds.) 1992, Artificial Life II: Santa Fe Institute Studies in the Sciences of Complexity, Proceedings vol. 10, Addison-Wesley, Redwood City. Larson, M. S. 1977, The Rise of Professionalism: A Sociological Analysis, The Regents of the University of California, Berkeley. 292 ___________________1993, Behind the Postmodern Façade: Architectural Change in Late Twentieth-Century America, University of California Press, Berkeley. Laufer, A. & Hoffman, E. 2000, Project Management Success Stories: Lessons of Project Leaders, John Wiley & Sons, Inc., New York. Leigh, S., Griesemer, S. & Griesemer, J. R. 1989, ‘Institutional ecology, ‘translations’ and boundary objects: amateurs and professionals in Berkeley’s Museum of Vertebrate Zoology, 1907-39’, Social Studies of Science, vol. 19, pp. 387-420. Lesser, E. & Prusak, L. (eds.) 2004, Creating Value with Knowledge: Insights from the IBM Institute for Business Value, Oxford University Press, Oxford. Lewin, R. & Regine, B. 2001, ‘An organic approach to management’, The Cap Gemini Ernst & Young Center for Business Innovation, Perspectives on Business Innovation Journal, no. 4, pp. 19-26. Lichtig, W. A. 2005, ‘Sutter Health: developing a contracting model to support lean project delivery’, Lean Construction Journal, [Online], vol. 2, no. 1, pp. 105-112. Available from: http://www.leanconstructionjournal.org [Accessed July 2005]. Liker, J. K. 2004, The Toyota Way: 14 Management principles from the World’s Greatest Manufacturer, McGraw-Hill, New York. Lovelock, J. 1979, Gaia: A New Look at Life on Earth, Oxford University Press, Oxford. Luhmann, N. 1990, ‘The autopoiesis of social systems’, in Essays on Self-Reference, ed. N. Luhmann, Columbia University Press, New York, pp. 1-21. Margulis, L. & Sagan, D. 1997, Microcosmos: Four Billion Years of Microbial Evolution, Simon & Schuster, New York. Marshall, T. H. 1939, ‘Recent history of professionalism in relation to social structure & social policy’, Canadian Journal of Economics & Political Science, vol. 5, no. 3, pp. 325-340. Matthews, O. & Howell, G. A. 2005, ‘Integrated project delivery: an example of relational contracting’, Lean Construction Journal, [Online], vol. 2, no. 1, pp. 46-61. Available from: http://www.leanconstructionjournal.org [Accessed July 2005]. Maturana, H. & Varela, F. 1980, Autopoiesis and Cognition: The Realization of the Living, Boston Studies in the Philosophy of Science, vol. 42, D. Reidel Publishing Co., Dordecht. Miles, R. E., Snow, C. C. & Miles, G. 2000, ‘TheFuture.org’, Long Range Planning, vol. 33, pp. 300-321. Mintzberg, H. 1979, The Structuring of Organizations, A Synthesis of the Research, Prentice-Hall, Englewood Cliffs. Mitchell, D. 1979, Control Without Bureaucracy, McGraw-Hill, London. 293 Mohsini, R. 1992, ‘On measuring project performance: some problems of aggregation’, in Evaluation and Prediction in Design, ed. Y. E. Kalay, John Wiley & Sons, New York, pp. 239-250. Monge, P. R. & Eisenberg, E. M. 1987, ‘Emergent communication networks’, in Handbook of Organizational Communication, eds. F. M. P. Jablin, L. Linda, K. H. Roberts & L. W. Porter, Sage Publications, Newbury Park, pp. 304-342.. Morgan, G. 1997, Images of Organizations, Sage Publications, London. Nonaka, S. & Konno, N. 1998, ‘The concept of “ba”: building a foundation for knowledge creation’, California Management Review, vol. 40, no. 3, pp. 40-55. Nonaka, S. & Takeuchi, O. 1995, The Knowledge-Creating Company, How Japanese Companies Create the Dynamics of Innovation, Oxford University Press, New York. Nooteboom, B. et al. 2000, ‘Network interactions and mutual dependence: a test in the car industry’, Industry and Innovation, vol. 7, no. 1, pp. 117-144. North, D. C. 1992, Transaction Costs, Institutions, and Economic Performance, ICS Press, San Francisco. Parsons, T. 1939, ‘Professions & social structure’, Social Forces, vol. 17, no. 4, pp. 457467. Pascale, R. T. et al. 2000, Surfing the Edge of Chaos: The Laws of Nature and the New Laws of Business, Three Rivers Press, New York. Pinto, J. K. 2003, Project Management Handbook, Project Management Institute, JB Jossey-Bass Publishers, San Francisco. Powell, W. W. 1998, ‘Learning from collaboration: knowledge and networks in the biotechnology and pharmaceutical industries’, California Management Review, vol. 40, no. 3, pp.228-240. Ricketts, M. L. 1999, ‘Design-build not draw-build’, The Military Engineer, no. 598, pp. 65-68. Rosenhead, J. 1998, ‘Complexity theory and management practice’, Human Nature Daily Review, [Online], Available from: http://human-nature.com/science-asculture/rosenhead.html [Accessed June 2005]. Russell, P. 2000, The Global Brain Awakens: Our Next Evolutionary Leap, Element Books Limited, Boston. Schön, D. 1983, The Reflective Practitioner: How Professionals Think in Action, Basic Books, N.Y. Schrage, M. 1997, ‘The real problem with computers’, Harvard Business Review, September-October, pp. 178-183. _______________2000, Serious Play: How the Worlds Best Companies Simulate to Innovate, Harvard Business School Press, Boston. 294 Schwartz, P. 1991, The Art of the Long View: Planning for the Future in an Uncertain World, Currency Doubleday, New York. Senge, P. M. 1990, The Fifth Discipline: The Art & Practice of the Learning Organization, Currency Doubleday, New York. Smith, A. 1880, An Inquiry into the Nature and Causes of The Wealth of Nations, 2nd edn, ed. James E. Thorold Rogers, Clarendon Press, Oxford. Stacey, R. D. 1996, Complexity and Creativity in Organizations, Berrett-Koehler Publishers, San Francisco. Taylor, F. W. 1967, The Principles of Scientific Management, Norton, New York. The Haskell Company. 1997, ‘Architectural excellence’, [Online], Available from: http://www.thehaskellco.com/assets/documentation/aboutdesignbuild/whitepapers/Ar chitecture_White_Paper.pdf [Accessed June 2005]. Thompson, J. D. 1967, Organizations in Action, Social Science Bases of Administrative Theory, McGraw-Hill, New York. Thomsen, C. B. 1999, ‘The 21st century master builder’, in The Military Engineer, vol. 91, pp. 40-45. Thorelli, H. B. 1986, ‘Network: between markets and hierarchies’, Strategic Management Journal, vol. 7, pp. 37-51. Tsoukas, H. 1996, ‘The firm as a distributed knowledge system: a constructionist approach’, Strategic Management Journal, vol. 17, pp. 11-25. Uzzi, B. 1997, ‘Social structure and competition in interfirm networks: The paradox of embeddedness’, Administrative Science Quarterly, vol. 42, pp. 35-67. von Bertalanffy, L. 1969, General System Theory: Foundations, Development, Applications, George Braziller Inc., New York. Waldrop, M. M. 1992, Complexity: The Emerging Science at the Edge of Order and Chaos, Touchstone, New York. Weber, M. 1947, The Theory of Social and Economic Organization, Free Press, New York. Wheatley, M. J. 1992, 1994, Leadership and the New Science: Learning about Organization from an Orderly Universe, Berrett-Koehler Publishers, San Francisco. Whitaker, R. 1996, ‘Autopoiesis theory, tutorial lesson 3’, [Online], Available from: http://www.unikk.ch/course/PDF/lesson3-text1.PDF [Accessed April 2005]. Winograd, T. & Flores, F. 1987, Understanding Computers and Cognition: A New Foundation for Design, 1st edn, Addison-Wesley Professional, Reading. 295