SEARCH

SEARCH BY CITATION

Introduction

  1. Top of page
  2. Introduction
  3. Industrial Ecology, Evolution, and Sociotechnical Systems
  4. Complexity and Industrial Ecology
  5. Overview of the Special Issue
  6. Conclusion and Outlook
  7. References
  8. About the Authors

Both in nature and in our current industrialized society, we can discern complex, layered, and dynamic systems that interact with their environment and perpetually affect one another. Over time, these systems evolve, and complex structures, such as the Amazonian rainforest or New York City, emerge. These systems are characterized by diversity, multiple interactions both within and between layers, feedback loops, and emergence—all characteristics that lead us to consider them as complex.

Industrial ecology focuses on the sustainability of anthropogenic systems. As researchers have explored the ecology metaphor, the industrial ecology body of knowledge and tools—life cycle assessment (LCA), material flow analysis (MFA), and substance flow analysis (SFA), among others—have enabled a thorough analysis of industrial systems and anthropogenic metabolism at various scales. The field has allowed identification of causes of unsustainability and barriers to sustainability and continues to provide recommendations for more sustainable configurations, albeit principally on a case-by-case basis. A crucial contribution of these studies, however, is that the results are fed back to the multiple stakeholders who are involved in and responsible for the systems that bring into being a particular product or service. This contrasts with a more traditional engineering paradigm on the design, construction, and operation of industrial facilities or artifacts. This series of activities, although it involves many teams from different engineering and related disciplines, ultimately is coordinated by a single principal based on a defined but generally narrow set of (design) objectives.

In complex systems theory, evolution is a central paradigm; this calls for a different approach—away from centralized coordination and based on a philosophy of guidance rather than control. Under the banner of complexity science, a range of methodologies and associated tools have been developed—including artificial life, multiagent systems, and agent-based modeling. Tools developed in other fields, such as system dynamics, originally from organizational management, have found a new lease on life. Because the configuration of the industrial system beyond single artifacts or industrial facilities is created by many stakeholders or agents under a variety of coordination mechanisms—organizations, markets, policy, and regulation—we propose that complexity theory and its tools have the potential to shift the frontier of industrial ecology. They could do so by enhancing the quality of systems analysis and by underpinning recommendations for redirecting industrial development toward sustainability.

This special issue is an attempt to capture and illustrate the state of the art in complexity and industrial ecology and to build on the success of the set of sessions on complexity at the 2007 conference of the International Society for Industrial Ecology and the associated Complexity Workshop held immediately thereafter (17–20 June and 21 June 2007, respectively).

To provide context for this special issue on complexity and industrial ecology, this editorial presents a sociotechnical systems view as a framework for positioning the articles in this collection. In addition, it summarizes some general themes and tentative conclusions that have emerged on complexity and industrial ecology in discussions that have occurred at the conference, at the symposium, and during the development of the articles presented in this special issue.

Industrial Ecology, Evolution, and Sociotechnical Systems

  1. Top of page
  2. Introduction
  3. Industrial Ecology, Evolution, and Sociotechnical Systems
  4. Complexity and Industrial Ecology
  5. Overview of the Special Issue
  6. Conclusion and Outlook
  7. References
  8. About the Authors

A system may be defined as “a structured assemblage of elements and subsystems, which interact through interfaces. The interaction occurs between system elements and between the system and its environment” (Asbjørnsen 1992, I.1.2). Thus, industrial ecology is a multidisciplinary,

system-oriented concept [which] suggests that industrial design and manufacturing processes are not performed in isolation from their surroundings but rather are influenced by them and, in turn, have influence on them. (Graedel et al. 1993, 18)

According to Erkman (1997, 1), industrial ecology endeavors to

understand how the industrial system works, how it is regulated, and its interaction with the biosphere; then, on the basis of what we know about ecosystems, to determine how it could be restructured to make it compatible with the way natural ecosystems function.

But how is this to be done? In his 1999 book Industrial Ecology: Policy Framework and Implementation, Allenby takes a deep look at how the aspiration of industrial ecology, sustainability, could be realized through a policy framework. In his industrial ecology intellectual framework, he defines the industrial ecology infrastructure to include

developing and implementing the legal, economic, and other incentive systems by which desirable behavior can be promoted, as well as the methodologies, tools, data and information resources necessary to define and support such behavior. (Allenby 1999, 13).

To begin to answer this question and determine its methodological and practical implications, let us view the industrial system as a dynamic sociotechnical system that interacts with a continuously changing world (Bijker et al. 1987; Nikolić et al. 2009). Therein, the technical network and the actors and bodies that are involved—the social network—together form an interconnected, complex network (see figure 1). In the technical network, material, energy, money, and information are exchanged over suitable interfaces. Much like ecosystems, our industrial system can be seen to constantly evolve because individuals, organizations, and governments decide on, for example, consumption, industrial design and (dis)investment, and coordination and regulation, respectively, in response to and in interaction with their respective environments. Activities in the technical network are subject to extensive governance mechanisms, including regulatory regimes.

image

Figure 1. A sociotechnical network—an interconnected complex network.

Download figure to PowerPoint

In the social network, relations, transactions, and decision making are influenced, for example, by access to and property rights on resources and physical and financial assets and are coordinated via markets. The “social” nature of the network should not go unrecognized, with aspects such as kinship, trust, and loyalty playing an important role at the human level of these systems. The interplay between technical and social systems is dynamic, and with time these systems coevolve. The acknowledgment of the dynamic connections between the social domain (including humans; their perceptions, values, behavior, and relationships; organizations; and governance structures) and the technical or physical systems in which they operate and that they continue to bring into being (e.g., industries, infrastructure, buildings, financial markets) is core to the idea of sociotechnical systems.

Thus, although evolution continuously shapes and reshapes natural, living ecosystems in response to changing external conditions, a distinct characteristic of our industrial system is that it evolves as a result of the intentions and decisions of sentient beings related in a layered social network. Some of the actors have a limited span of control and make merely operational decisions to achieve local, short-term objectives in response to change. Other strategic decision making by individuals, companies, or governments may interface with larger systems, countries, or continents. Decisions at this level should facilitate the achievement of some vision of the performance of the system as a whole over a long time horizon.

Recognizing the interactions within and the interplay among the technical network, the social network, and the biophysical world is crucial to understanding and shaping our anthropogenic systems to let them “flourish forever” (Ehrenfeld 2007). Beyond the scale of a single facility, the industrial system is subject to distributed control—no single actor determines the configuration and operation of the system. One may see that this sociotechnical systems perspective can be used at each layer beyond a single facility to frame eco-industrial parks, industrial regions or sectors, mega-cities and urban areas, energy infrastructure, agriculture, fisheries and related industries, material cycles, and so on.

Complexity and Industrial Ecology

  1. Top of page
  2. Introduction
  3. Industrial Ecology, Evolution, and Sociotechnical Systems
  4. Complexity and Industrial Ecology
  5. Overview of the Special Issue
  6. Conclusion and Outlook
  7. References
  8. About the Authors

The preceding discussion suggests that a complex system consists of many agents that interact with each other at multiple physical or social interfaces—material, energy, information—and across layers. With time, dynamic patterns of system behavior emerge from these local interactions between system elements (Kauffman 1993; Newman 2003). As with all types of modeling, modeling of a sociotechnical system requires some degree of abstraction, yet core to this effort is the ability to model interactions and associated (often mutually interdependent) dynamic behavior. Modeling a sociotechnical system as a complex system thus requires selection of the system boundary, identification of an appropriate set of possible system elements, and definition of the span and nature of their interactions. On appropriate definition of initial status and external world conditions, over time the model of the system can start to grow through the web of interaction. With time, a system structure assembles, and total system behavior emerges (after Nikolić et al. 2009).

The ecology metaphor implies that robust system elements and configurations are to be preferred, that system resilience to incidents and change requires a diversity of system elements that populate the technical and social subsystems, and that systems should not overstress their environment's carrying capacity. Thus, when material and energy flows in the industrial system or parts thereof have been inventoried and their effects assessed, it is critical not only to translate “that information into a form that is useful for designers and others who will actually embed the knowledge into the economic system” (Allenby 1999, 14) but also to address the question “What is to be designed or steered to foster sustainability?” considering that sustainability is an emergent property of a complex, global sociotechnical system (Allenby 1999; Ehrenfeld 2007).

Although anthropogenic systems are complex, layered, and dynamic, the majority of the work published by the community to date appears to focus on the metabolism, structures, and relationships abstracted as single-layer, static technical systems. At the time of writing, some 650 articles have been published in the Journal of Industrial Ecology, the majority of which concern the metabolism, structures, and relationships of our technical systems (Ehrenfeld 2007). Only 6 articles specifically relate to agency, complexity, and complex systems theory and its tools in industrial ecology.

Spiegelman (2003) and Ehrenfeld (2007) advocate moving “beyond the central metaphor of industrial ecology” (Spiegelman 2003, 17). On the basis of the founding work of Holling (1986) and Kay (2002) in complex adaptive systems theory and the recognition of ecosystems as complex, self-organizing, hierarchical, open systems, Spiegelman suggests that complex systems theory “could add depth and sophistication to the field of industrial ecology” (Spiegelman 2003, 17). His work revolves around the understanding of the technical subsystem and its interaction with natural ecosystems. Ehrenfeld suggests that “complexity models of living systems can ground alternative normative models for sustainability as an emergent property” (Ehrenfeld 2007, 73). Ehrenfeld examines the relationship between industrial ecology and sustainability without articulating a precise systems view but implicitly defines industrial ecology as an engineering discipline, hints that to date industrial ecology has failed to include equity and economy, and states that “nature never does anything intentionally. It just changes as conditions change” (78).

Positioning these articles in the sociotechnical systems framework, one may see that Spiegelman (2003) and Ehrenfeld (2007) explore the ramifications for industrial ecology when dealing with the interaction between the technical subsystem and the environment at large. Andrews (2000) conjectures economics will help us unravel interactions within the social network and their relation to shaping sustainable technical networks: In “Building a Micro-Foundation for Industrial Ecology,”Andrews (2000) explores how industrial ecology could benefit from economics, notably the core of institutional economics: property rights, agency, and transaction cost theory. Two similarly titled articles address agent-based modeling and industrial ecology (Axtell et al. 2001; Kraines and Wallace 2006). Limiting their scope to the social subsystem, these authors introduce multiagent models as a means to simulate social networks. Whereas Kraines and Wallace (2006) emphasize the possibility to study interactions in human societies, Axtell and colleagues (2001, 12) spell out the analogy to and use of object-oriented programming and suggest the models “will give industrial ecologists a needed test bed for safe, low-cost management and public policy experiments.”

Overview of the Special Issue

  1. Top of page
  2. Introduction
  3. Industrial Ecology, Evolution, and Sociotechnical Systems
  4. Complexity and Industrial Ecology
  5. Overview of the Special Issue
  6. Conclusion and Outlook
  7. References
  8. About the Authors

The articles contained in this special issue range from addressing the future of the industrial system and humans to the use of agent-based modeling on relatively small elements of the industrial system. Let us briefly discuss and frame the articles in order of appearance.

In an invited column, Ehrenfeld (2009) provides clarification on his argument that industrial ecology is a normative discipline (e.g., Ehrenfeld 2007). Drawing on the work of Kay and Regier (2000) and Kay (2002), he elaborates his argument that sustainability is an emergent property of a system and expresses his hope that a deeper engagement with complexity will attract a wider set of researchers and practitioners and, ultimately, expand the reach of industrial ecology.

Allenby (2009) entwines a sociotechnical earth systems view with a deep exploration of the making of the “anthropocene.” He argues that technology clusters continue to shape our industrial system—in the 19th century, railroads evolved with the steam engine, then came electricity and engineering and, from 1930 to 1990, oil, the automobile, and the aircraft. He presents a provocative yet underpinned vision of what may lie ahead for humans and the earth due to what he dubs “the Five Horsemen”: nanotechnology, biotechnology, robotics, information and communication technology (ICT), and applied cognitive science. These may bring us into an age where the sociotechnical “design space” expands beyond the industrial system to include the design of humans. Outlining “the implications of this socio-techno landscape for industrial ecology,” he “suggests profound theoretical challenges as well as important new areas of research” lie ahead for industrial ecology.

Rotmans and Loorbach (2009) start from a normative notion of transition management as “fostering sustainability transitions” and conjecture that understanding the functioning of the social system provides useful insights for directing the industrial system toward sustainability. Not only do they explore this notion of transition management and ground it in complex systems theory rather than presenting “a new policy framework” (Allenby 1999), they offer a transition management framework. This can be seen as a form of adaptive management (Kay 2002), or the organization of a national innovation system (Lundvall 1992; Freeman 1995; OECD 1997) from which innovations and novel regimes then evolve. In this respect, the article provides a link to innovation studies, notably innovation systems research (e.g., Hekkert et al. 2007).

Batten (2009) explores the linkage of complex systems science and participatory modeling, role play, and games, which could be used to build networks of trust in small-scale social networks. Summarizing an agent-based simulation of an electricity sector developed from an adaptive, complex sociotechnical systems view, he emphasizes that such simulations “help to identify potential elements of an industrial ecosystem that could work together to achieve more eco-efficient outcomes … and can explore various ‘what-if’ scenarios under different eco-efficiency goals.”

Baynes (2009) presents a historical overview and opportunities arising from the application of complexity science to urban development and management for the industrial ecology community. The author postulates that “cities are certainly heterogeneous entities comprising many specialized components … it appears that cities possess some if not all of the attributes of complex socio-technical systems”; a summary of modeling methods and advancement over the last 30 years is given. With an emphasis on the technical subsystem and environment, he outlines avenues for industrial ecology research toward understanding the morphology and dynamics of city growth and function.

Ashton (2009) presents a framework to assess the structure, function, and evolution of a regional industrial ecosystem in Puerto Rico. Therein, she integrates insights from industrial ecology, economic geography, and complex systems theory. She focuses on the social network, and her empirical research reveals changes in the system's institutional context, composition of the industrial community, and, consequently, its resource flows and industrial system functionality.

DeLaurentis and Ayyalasomayajula (2009) link yet another emerging body of knowledge, systems-of-systems engineering, to industrial ecology. They present a thorough comparison of the vocabulary used in complex adaptive systems, systems of systems, and industrial ecology and illustrate the use of the common foundations and tools in their simulation and analysis of air transport network evolution. Recognizing that this is a true sociotechnical system, they conclude that there is much to be gained from a synergy in systems-of-systems engineering and industrial ecology.

Wood and Lenzen (2009) have completed a thorough input−output analysis of the Australian economy over its evolution from 1975 to 1999 and explain how their aggregate results can be used as a proxy for complexity of an economy. Instead of employing a bottom-up approach, they take a snapshot of an entire economy and focus on indicators for the evolution of that sociotechnical system and its interaction with the environment.

Kempener, Beck, and Petrie (2009) have developed a multiscale modeling framework that combines top-down optimization to establish globally “best” system configurations and agent-based simulation to explore bottom-up assembly of a bioenergy network. The framework is setup to enable analysis of the effect of different policy frameworks and business strategies. It implicitly uses a sociotechnical systems perspective and focuses on the combined effect of technical−social network interaction.

Davis, Nikolić, and Dijkema (2009) have created a method to construct an LCA to provide environmental information for an industrial system while it evolves. They demonstrate how agent-based modeling and LCA can be connected to provide environmental information to the actors or agents in the social networks for decision support, and they illustrate this in a proof of concept in which an evolving bioelectricity system is modeled and life cycle greenhouse gas emissions are analyzed.

Andrews and DeVault (2009) introduce a multiagent simulation framework for investigating the emergence of niche markets for environmentally innovative products. Their focus is on elucidating the interplay of the social network of consumers, businesses, and government. Instead of focusing on technological product innovation alone, their work illustrates that not only diffusion into the market but the very emergence of such a niche market is crucial for the success of any green product invention.

Conclusion and Outlook

  1. Top of page
  2. Introduction
  3. Industrial Ecology, Evolution, and Sociotechnical Systems
  4. Complexity and Industrial Ecology
  5. Overview of the Special Issue
  6. Conclusion and Outlook
  7. References
  8. About the Authors

The study of the nature of sociotechnical systems and their interrelation with biophysical systems is core to industrial ecology. Achieving our sustainability aspirations requires a systems approach that is conscious of the limitations of system decomposition—what the system elements are—as well as a recognition of system links—intra- and interconnections. The four articles in the Forum section (Allenby 2009; Batten 2009; Baynes 2009; Rotmans & Loorbach 2009) all provide some vision of the relevance and application of complex systems theory in industrial ecology. Picturing our global earth system to be composed of technical and social networked subsystems and recognizing their mutual interdependency on the environment allows us to think beyond the metaphor or analogy of natural ecosystems when addressing the sustainability challenge.

Whereas Ehrenfeld (2007, 77) argues that Allenby (1999) is incorrect in referring to industrial ecology as the “science of sustainability,” we conjecture that climate change, the world economic crisis, and inequity illustrate that we must establish some science and engineering of sustainability (Nikolić et al. 2009, 156)—where the “science” is encompassing of the social sciences and the “engineering” is not the realm of traditional engineering but a call to venture beyond analysis to action (after Nikolić et al. 2009). Indeed, “sustainability is an emergent property of the whole [complex] system” (Ehrenfeld 2007, 77)—our key challenge, then, is identifying how to influence this emergence. Complexity science tools combined with traditional industrial ecology tools and the social sciences may help elucidate the quantitative effects and qualitative experience of industrial system evolution and hence allow the assessment of policy frameworks for their role in bringing into being (future) sustainability. The six articles in the Research and Analysis section demonstrate that substantial work is already has already been done to use and expand the body of knowledge of complexity and industrial ecology to this end, and they highlight the challenges and the excitement ahead.

Industrial ecology is inherently interdisciplinary, bringing together many disciplines and exploring what we can gain from venturing beyond merely meeting at the interfaces into informed and creative synthesis in a quest to gain wider, deeper, and ultimately transformative understanding. On the basis of the overview provided here and the articles presented in this issue, we conjecture that complexity science can contribute to the future of industrial ecology by extending its interdisciplinary foundation and scope. Complexity science can provide a conceptual and practical framework for the exploration and integration of ideas and tools from other (some as yet underexplored and possibly unexplored) disciplines and newer frontiers for industrial ecology, such as nonlinear, nonequilibrium thermodynamics. The aspiration of “bringing systems into being” through guided evolution based on the notions of evolution instead of design and of emergence instead of prescription, inspired by complex systems theory, has much to offer industrial ecology and may help bridge the gap between halting at initial efforts at system understanding and making the leap toward system shaping—a transformation from valuable analysis to deeper understanding and, ultimately, to responsible action.

References

  1. Top of page
  2. Introduction
  3. Industrial Ecology, Evolution, and Sociotechnical Systems
  4. Complexity and Industrial Ecology
  5. Overview of the Special Issue
  6. Conclusion and Outlook
  7. References
  8. About the Authors

About the Authors

  1. Top of page
  2. Introduction
  3. Industrial Ecology, Evolution, and Sociotechnical Systems
  4. Complexity and Industrial Ecology
  5. Overview of the Special Issue
  6. Conclusion and Outlook
  7. References
  8. About the Authors

Lauren Basson is a lecturer at the Centre for Environmental Strategy of the University of Surrey in the United Kingdom. Gerard P. J. Dijkema is associate professor of energy and industry, Faculty of Technology, Policy and Management, within Delft University of Technology, the Netherlands.