Industrial ecosystems as technological niches

Journal of Cleaner Production 17 (2009) 172–180

Contents lists available at ScienceDirect

Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro

Industrial ecosystems as technological niches Emmanuel D. Adamides*, Yannis Mouzakitis Section of Management, Department of Mechanical Engineering and Aeronautics, University of Patras, University Campus of Patras, Rio 26500, Greece

a r t i c l e i n f o

a b s t r a c t

Article history: Received 7 June 2007 Received in revised form 9 April 2008 Accepted 9 April 2008 Available online 23 May 2008

In this paper we consider industrial ecology as a novel and distinct state in the evolution of production systems trajectory. More specifically, adopting an evolutionary institutionalist’s perspective at the sociotechnical system level of analysis, we discuss the transition towards an industrial ecology-inspired industrial production system through the governance approach of strategic niche management (SNM). Towards this end, the paper first develops industrial ecology as a novel state of the industrial production socio-technical system paying particular attention at its technological component, then it discusses in detail the transition of this system through SNM, and examines whether existing implementations of the industrial ecology (industrial ecosystems and eco-industrial parks) are efforts embracing this philosophy and transition procedures. We conclude by stressing the importance of adopting the technology policy and strategic niche management perspectives in eco-industrial projects and we provide insights for accomplishing it in a more effective manner. Ó 2008 Elsevier Ltd. All rights reserved.

Keywords: Industrial ecology Socio-technical systems Industrial production paradigms Transitions Strategic niche management Industrial ecosystems

1. Introduction – Industrial ecology: towards a novel regime of industrial production Industrial ecology can be thought as a vision that inspires technological and organisational innovation towards a novel industrial production paradigm, i.e. a novel system for organising work and technology that is based on the design and operation of industrial processes in series, as interlocking manmade ecosystems interfacing with the natural global ecosystem [1–3]. This perspective suggests that a shift towards industrial ecology constitutes a system innovation [4,5], and implies significant changes in the dominant world production system(s) which extend along the whole array of technology development, production of technological artefacts, and technology use activities, and affect, as they are been affected by, the related social groups and institutions involved [6]. On the basis of the existing evidence of the limited acceptability and practical application of the concept of industrial ecology [7,8], despite a considerable amount of academic research and pilot studies, an apparent question that comes into surface is whether this holistic perspective is articulated and understood appropriately. The aim of this paper is to contribute in this direction by considering industrial ecology in the framework of socio-technical system transitions [9,10]. Transitions are system transformation processes, in which society, or a complex system of society, changes in a fundamental way * Corresponding author. Tel.: þ30 2610 997 231; fax: þ30 2610 997 260. E-mail addresses: [email protected] (E.D. Adamides), ymouzakitis@ gmail.com (Y. Mouzakitis). 0959-6526/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.jclepro.2008.04.003

through a combination of improvement interventions and system innovations [10]. One way to stimulate such transitions is through the formation of technological niches [4,11] which are supported desirable socio-technical system configurations, institutionally and market protected, that act as ‘incubation rooms’, or settings for experimentations. Strategic niche management (SNM) is a strategy for organised policy-driven transition based on the creation of these ‘‘protected spaces’’ [12], with the aim that will eventually lead to market niches and induce changes on the current system configuration. By adopting an evolutionary institutionalist’s perspective, in this paper, we consider industrial ecology and industrial ecosystems as a desired new state for the industrial production socio-technical system, and we investigate the role and contribution of existing industrial ecosystem implementations in the transition towards this state. More specifically, the purpose of this paper is: (i) to reveal and emphasize the need for adopting the socio-technical systems perspective with respect to industrial ecology and industrial ecosystems analyses and policy-making, (ii) to clarify and operationalise the issues related to the adoption of this perspective, (iii) to comment on the process of formation, maintenance and appropriability of existing industrial ecosystems from the educative and paradigmatic perspective of niche development and SNM, and (iv) to provide insights for more effective, locally and globally, niche management towards an industrial ecology-inspired industrial production system. Towards achieving these objectives, in the rest of the paper, we first develop the idea of industrial ecology as a technology-driven novel state of the industrial production socio-technical system.

173

E.D. Adamides, Y. Mouzakitis / Journal of Cleaner Production 17 (2009) 172–180

Then, after discussing the technology component of the system, we discuss the transition process towards an industrial ecology-inspired (IEI) system, underlining the difficulties involved. Then SNM is presented as a means for governing this transition and overcoming the rigidities of the current paradigm. Based on the analysis of existing implementations of industrial ecology, we then develop a conceptual model of the development and operation of industrial ecosystems, which we use for performing a secondary level analysis of three implementations of IE under the prism of technological niches. We conclude the paper by providing some insights for its efficient and effective application for widely diffusing the concept of industrial ecology by adopting a technology policy perspective.

2. Industrial ecology as a novel state of the industrial production system Since the first industrial revolution, the global industrial production system has evolved through distinct stages influenced by the invention of core technological artefacts, production organisation methods, as well as societal developments. Elements and attributes of these three major subsystems of the overall industrial production socio-technical system have defined distinct system stages (frequently co-existing) that can be identified as craft production and the American system of manufacturers, mass production, and later, flexible specialisation, agile and lean production [13,14], which correspond, from an institutional economics perspective, to ownership, managerial and collective capitalism forms, respectively [6]. In order to understand the evolution of the industrial production system for identifying policy instruments that will shift it towards an industrial ecology-inspired state, we have to look closer at its structure and dynamics from the perspective of socio-technical systems. In general, a socio-technical system (STS) consists of two co-evolving subsystems: the development and production of technological artefacts subsystem and the technology and artefacts use subsystem [15]. The former includes interlinked activities that surround the actual development of technology and the production of technological artefacts, and whose purpose is to coordinate the supply of productive factors (transfer of scientific knowledge to the labour force through the education system, development and supply of technological knowledge, supply of capital and industrial

Materials

Energy

Industrial production process technologies system (development and production)

Industrial production management technologies system (development and production)

equipment, and supply of natural resources). Similarly, the use subsystem consists of activities for the adoption and use of technological artefact(s) (products), such as the supply of complementary artefacts, the provision of installation and maintenance services, the attachment of cultural connotations, etc. Relevant social groups at this side include customers, various customer groupings, professional organisations, media, NGOs, etc. Authorities and related organisations operate on both sides and, directly or indirectly, regulate the flow of artefacts and decisions/information between the two subsystems. Historically, transitions of industrial production systems have been the result of both contextual factors, such as the evolution of social structures and institutions, and agency mirrored in economic variables, such as changes in demand and technology and the search for efficacy. There has been a tendency to undermine the latter [16], but recent studies have revealed and stressed the importance of both driving forces [6,17]. In the case of industrial ecology, it is the use context derived from its vision that defines the ‘‘systemness’’ of the technology (i.e. how work and technology are organised), rather than the relations among the technical artefacts, which may be totally unrelated and differentiated by industry characteristics (i.e. specific to the chemical industry, specific to the metallic structures industry, etc.). In a more refined form, the basic structure of the industrial production system can be represented as in Fig. 1. In this, the development and production of technological artefacts’ subsystem consists of two generic co-evolving subsystems: the technology development and technological artefacts (process technologies) production subsystem (material transformation processes, machinery) and the production organisation and management methods and technologies (management and indirect process technologies) subsystem (methods and tools for organising and managing work on material transformation processes, e.g. production planning and control, procurement management, inventory management, labour organisation and development policies, etc.). It is accompanied by the use subsystem (institutional arrangements and social practices) which acts as the selection environment for selecting and putting into work specific configurations of production process technologies and production management methods and tools. The three subsystems co-evolve and influence each other as they are influenced by: as specific institutional arrangements and social practices select and support the development and production of specific process technologies and

Waste, pollution

(Selection and) use of industrial production process and management technologies system

Fig. 1. The basic structure of the industrial production socio-technical system.

174

E.D. Adamides, Y. Mouzakitis / Journal of Cleaner Production 17 (2009) 172–180

production management methods, in the same way, the latter trigger, reinforce or reject specific institutional arrangements and social practices. What distinguishes industrial production system configurations (paradigms, or system states) is the size and flexibility of production units, their relations with suppliers and customers, the organisation of work and labour and the nature of technology and its mode of employment. Based on its vision and the different existing implementations of industrial ecology, what would provide the distinguishing attribute of an IEI industrial production system is the cooperative relations with suppliers and customers of raw materials and components, products, by-products, energy and waste, as well as, the employment of ‘‘connecting’’ (processes and people) technologies. 3. Direct and indirect process technologies for IEI production systems So far, the majority of IE-related technology studies and implementations seem to be more concerned with how to use existing (peripheral to IE) technologies for post-processing and preprocessing waste, constructing physical connections among processes, etc. rather than considering technological innovation as the enabler of industrial ecology. That is, in the context of Fig. 1, there is a one-way asymmetric relationship between the use and the technology systems. Researches, policy makers and developers of IES are predominantly interested in selecting from existing technological configurations rather than stimulating the development of novel technologies and technological artefacts. As a result, the whole issue of industrial ecology becomes an attempt to accomplish small improvements on the existing industrial production systems rather than change them. However, in such situations, what can be done and what must be done are primarily determined by the possibilities and limitations of technology. As industrial ecology direct process technological artefacts can be defined all the devices and systems that are used for making economically and environmentally effective the connection of industrial processes, the re-using and recycling of wastes, the feeding of processes with alternative energy forms, the disassembling of products, etc. Consequently, industrial ecology technology is a usecontext-defined bouquet technology comprising a set of lowerlevel (individual) technologies whose artefacts are used to modify existing production process and to link inputs and outputs of processes. In Kurz’s words they are ‘‘product-cum-process innovations’’ that transform ‘‘the bads of production into goods’’ [18]. The multitude of IE-related technologies and the apparent specificity of application contexts, suggest, first, that there cannot be a major (core) technology and technological artefact which will trigger the transition of the current industrial system towards industrial ecology, as it was, for instance, the electric motor in the case of the transition to mass production [12]. Secondly, the active involvement of users in the innovation and technology development processes is necessary. User-centred innovation and technology development [19] imply that the links between the constituent parts of the socio-technical system, at that particular state, are quite strong and that in the process of transition changes are required across all system elements. In addition to direct process industrial ecology technologies, i.e. technologies for conditioning waste and by-products for (re)use as raw materials and energy sources of other processes, respectively, and the related connecting and transportation technologies, there is an accompanying set of indirect process industrial ecology technologies related to production management methods (the organisation and management of work), which can be defined as the ‘‘appliance of science to the processes which provide, or support, the infrastructure for the processes which directly contribute to the

production of products and services’’ [20]. These technologies are rarely emphasised, or even mentioned, in the industrial ecology literature, despite their importance at both operational (to synchronize processes internally and externally) and strategic (to cultivate inter-organisational relations and social capital) levels (exceptions include the Pfaffengrund case [21] (also see below)). In practical terms, indirect process technology is closely related to the information and communication technologies that support decision making, at various levels of detail and intelligence, at the product–process interface (e.g. innovation supporting, product design, CAD/CAM technologies), the production process per se (e.g. MRP and QC software), and the purchasing and distribution functions which are the contact points of the factory with the rest of the supply network (e.g. collaboration technologies, joint product development and concurrent engineering systems, and supply-chain management software). Some of the production management methods and technologies are bound to specific process technologies, even to specific organisational structures and markets, but overall they can be considered as independent. Undoubtedly, in an environment of collaboration and synchronised operation that characterises industrial ecology, indirect process technologies of all kinds, generic or customised, play a very significant role. 4. The transition towards an industrial ecology-inspired industrial production system The transition towards an industrial ecology-inspired industrial system requires a multitude of changes of varying scale and intensity along all the dimensions of the related current socio-technical system. These changes are fiercely resisted by the inertia forces of the existing industrial production regime, which is the result of dominant practices, logic sets, etc., in either the policy/use domain or the technological one, guiding decision making and inducing stability [4,10,22]. Social and institutional factors contribute to the inertia of regimes directing the locus of problem solving activities of engineers and other technologists, as well as policy makers, towards specific directions [22]. In socio-technical regimes, there is always a core technological framework that is shared by a community of technological and economic actors ‘‘as the starting point for looking for improvements in product and process efficiency’’. It focuses the attention of engineers upon certain problems, while neglecting others [12]. As a result, the selection of technologies and their employment in industrial/economic activities are biased towards activities and means that resemble those currently executed and used, respectively. The IEI industrial production system can be viewed as a future state, a potential new regime, which can be reached from the currently existing systems (mass production, flexible specialisation, lean production, etc.). In the transition towards a new industrial production regime formed around the concept of environmental consistency, there are barriers imposed by path dependencies on the different elements of the current regime, centred around individual production processes and optimising management technologies at the same level with a secondary emphasis on sustainability. In their consideration of transportation technologies, Kemp et al. [12] identified seven types of barriers in the shift towards a more sustainable system. Based on the same logic, and adding an additional type (economic factors associated with the dynamics of diffusion and substitution [23]), Table 1 summarizes the barriers towards an industrial ecology-inspired production technology-driven system/regime. It should be noted that these extend beyond classes of barriers identified for the implementation of individual industrial ecosystems where the role of technology is primarily viewed as supportive for the specific implementation [24,25]. The table is self-explaining, and one can easily argue that, beyond cognitive inertia, since there are immense interests vested

E.D. Adamides, Y. Mouzakitis / Journal of Cleaner Production 17 (2009) 172–180 Table 1 Barriers towards technologies

175

5. The strategic niche management approach industrial

ecology-driven

production

and

consumption

Technological factors - Production processes are not designed with an ‘‘extended-process’’ logic. - Production processes are not designed for being able to absorb waste and energy produced by other processes. - Engineering education systematically promotes technological optimization at the individual process level. - Modular and scalable designs are required to support continuous innovation at product level. - Technologies are complex to manage due to their specificity to products - Engineering education is primarily based on specialization (e.g. mechanical, chemical engineering, etc.). However, a chemical process may use the waste of a mechanical, etc. Government policies - Principally reactive to individual environmental issues. - No incentives for IE technology production and implementation are given. - Regulation is directed towards individual products and processes, not webs of products and networks of processes. - Little risk is taken in promoting IE or producing IE-inspired specifications for public sector projects. Cultural and psychological factors - Industrial areas are isolated ‘‘dirty’’ areas, not accessible by the public. - Perception of high costs for compliance and adoption. - Competition and indifference prevails cooperation. - Preference of more transparent to the public sustainable technologies. Demand factors - Perception of risk in need for cooperation and for adapting something new. - Industrial users may be requested to change practices of years. - Difficulty in deciding ‘‘who pays for what?’’ – Where is the value, in taking the burden of the waste, or in using the waste? Production factors - Competences in existing technologies may become obsolete, engineers and workers should invest in learning the new technologies. - Performance management systems are oriented towards individual process optimization. - There is a tendency towards localized micro-scale production activities. Infrastructure and maintenance - Marketing and distribution of IE technological artifacts are novel and demanding. - Brokerage services must be developed for bringing together users of different processes and technology providers. - Maintenance-providers relations should be changed (may concern more than on owner of the artefact). Undesirable societal and environmental effects - Some jobs in waste collection and damping may be lost, as it may be the case for jobs in raw materials and energy production. - Some loss of autonomy at organizational and regional levels. Economic factors - The economic rationale shifts from the minimisation of consumption to the minimisation of environmental impact at source. - Structure of financial system favours investments at the individual firm level (not operationally cooperating). - It is easier to apply and attract competitive investments for end-of-pipe technologies. - Specificity of industrial processes and micro-scale production reduces the impact of economies of scale.

in existing production technologies, organisation structures and management styles, stemming from economic and political commitments and institutional arrangements [9], the transition towards an IEI production regime is rather unlike to occur on its own, in the absence of long-term policy. However, a transition towards industrial ecology and IEI production may be possible through a gradual participative change in the rules of the game, i.e. in the way the industrial ecology related parameters are institutionalised in the decision making of actors in all the subsystems of the industrial production socio-technical system, who usually hold diverse, even conflicting views, ideologies and interests [26–28]. An instrument policy makers may use to facilitate this process is the approach of strategic niche management (SNM) whose principles in the context of industrial ecology are discussed in the following section.

Strategic niche management (SNM) is a strategy for policydriven regime transition based on the creation of ‘‘protected (from market forces) spaces’’, that is, niches, for the development, production and use of new technologies. Protected micro-socio-technical systems are formed around innovative technologies to act as sites of experimentation and learning about their desirability, their directions of future development and the ways to accelerate their diffusion. It is important to note that SNM differs from other technology development and diffusion approaches in that it brings the knowledge and expertise of users and other actors into the technology development process and generates interactive learning processes and institutional adaptations [12]. In accordance with the choice of the appropriate policy instrument, niches can be created in three principal ways [29]: as a result of intentional central planning (‘‘designed’’) by ‘‘imposing’’ rules and regulations, in a bottom-up market-like way by influencing the behaviour of local agents using instruments such as incentives and tax deductions, or by forming networks of actors which may play a significant role in the innovation process. Kemp et al. [12] distinguished five steps in the creation of niches: the process commences with the choice of a promising candidate technology, continues with the selection, implementation and scaling of the experiment(s), before the dismantling of protection, so that the specific socio-technical system learns to respond successfully to competition. The proponents and users of the SNM approach pay particular attention to a number of contextual factors, distinguished into two classes, which play a significant role on the success or failure of the niche. First, there must be a number of preconditions which include the availability of sheltered spaces for incubation, the possibility for continuous evaluation and incremental improvement of the experiment(s), the ability of the technology for capturing learning economics, i.e. have an inherent learning-by-doing capability with financial returns, the technology should still be open for development in diverse direction, and the technology (in its present form) should be already attractive to be used (and used) for certain applications. Secondly, there must be an appropriate external environment that stimulates experimentation. In such an environment, there may be a dominant regime with inherent instabilities that favours the development of new technologies indirectly (it ‘‘asks for’’ new technologies), which, however, as it was underlined in the previous section, is not a sufficient condition for successful niche development. In addition to a broad public support base, sufficient institutional support, actor skills, knowledge and techniques must also be available in the existing regime to enable its shift.

6. The dynamics of industrial ecosystems and eco-industrial parks The implementation of industrial ecology can take two forms: product-based and geographically confined [30]. In this article, in our discussion of the IEI production system, we pay particular attention to geographically confined industrial ecosystems (eco-industrial parks and districts) because they are more complex entities involving a greater number of organisations of varying activities and roles. In viewing existing industrial ecosystems as niches of a novel industrial production paradigm, these implementations of industrial ecology seem to be better candidates as their variety of actors and functions forms a micro-instantiation of global industrial production systems which are associated with regional development strategies that are easier to study and experiment with.

176

E.D. Adamides, Y. Mouzakitis / Journal of Cleaner Production 17 (2009) 172–180

There is a rich literature of descriptive and analytic nature concerning the development and operation of industrial ecosystems, eco-industrial parks, and industrial symbiosis (for instance [7,21,31–36]), which, however, does not consider them (at least explicitly) as local instantiations of any industrial production system. The literature stresses the notion of embeddedness (frequently expressed in different terms) as the main cause, the result, as well as the necessary condition for the survival of eco-industrial systems, as well as the differentiating factor from intra-firm interprocess connectivity systems, such as the BASF ‘production verbund’ [21]. Embeddedness is a property of inter-organisational relationships that allows a social dimension to exist, influencing the economic behaviour of partners. Embedded relationships are characterised by stronger trust, rich information exchange, and joint problem solving [37,38]. It becomes apparent that existing eco-industrial parks are the result of both deliberate and evolutionary processes, which either induced sustainable economic/industrial activity into ‘‘virgin’’ areas, or transformed areas of non-sustainable economic activity into sustainable ones [39]. In the majority of cases, the development/transition process has been initiated either internally by firms or institutions operating in the specific location (endogenous development, such as in the Kalundborg case [40]), or by external bodies (exogenous development, such as in the Pfaffengrund case [21]). Endogenous development usually starts from an economic or environmental crisis at local level which triggers a re-embeddedness process (formation of new relationships, probably with new partners) along a different competition and cooperation rationale (e.g. from large mature industries to modern small-scale ones). Exogenous eco-industrial development frequently has to overcome the dis-embeddedness (the breaking of social relationships) caused by an outdated or failed model. Re-embedding may take place either under the umbrella of an encouraging and supporting environment, or, alternatively, within an alien one. Eco-industrial systems that have changed the very nature of economic activity of a geographic area (e.g. from manufacturing to service-based) admittedly faced an alien environment (long cognitive distance) as institutions and resources, such as labour, were unfamiliar with the new activities. Obviously, such changes were more difficult to take place and be supported, and required the import of new resources. Nonetheless, in both cases, during the development and growth of eco-industrial systems/parks, additional resources are required to support and maintain the momentum for the development of the technological and organisational innovations required for adapting to a changing environment. Of particular importance to industrial ecosystems is the notion of technology-dependent embeddedness, mirrored on the organisations involved. That is, the mutual investments and commitments made bind organisations together so that reflexive behaviours, particularly on the part of gate-keepers or anchor tenants, may result in disembeddeness and failure. Efficiency-seeking strategies of individual firms may result in relocation, re-negotiation of relationships, or even in process innovation. On the other hand, innovation-seeking strategies may result in novel products and processes, or in association with different partners [38]. In all cases, connections and recycling loops are likely to break, while existing technologies may be incapable of responding to the new situation. Novel products and processes require novel industrial ecology technologies, new relations, even new organisational forms, all involving a certain amount of risk to be undertaken by individual organisations. Hence, appropriate technical, organisational and institutional resources [41,42] need to be a priori built internally or supplied externally for minimising risks (Fig. 2). Industrial ecosystems have been developed following different paths and their analysis can be performed by considering their organisational and technical aspects, as well as their processes of

Environment, (natural, social, economic) and industrial structure

Exogenous initiatives

Supporting or impeding Initial state (non-sustainable)

Dynamics of transition

Endogenous initiatives

Resources to achieve and maintain organizational and institutional innovation Industrial Ecosystem

Resources to achieve and maintain technological innovation

Fig. 2. A conceptual model for the development of an industrial ecosystem.

formation and evolution using the model of Fig. 2. Clearly, this conceptual model serves as an industrial ecosystem-specific model for niche formation and experimentation. Once the resources and processes required for the development and maintenance of an industrial ecosystem as described above are put into place, or at least considered, the system would directly or indirectly require an action agenda similar to strategic niche management. The installation of knowledge diffusion mechanisms and the implementation of policies for the gradual withdrawal of protection will contribute to greater awareness and wider acceptance of the industrial ecology-inspired production model. In examining industrial ecosystems as niches we are particularly interested in their completeness with respect to both structure and process. In a niche, as an experimental micro-instantiation of the new IEI production system, one would expect the involvement of knowledge producers, innovators, technology producers of the direct and indirect technologies, producers of related technological artefacts, suppliers of capital, public authorities at different levels (regional, national and international), users of technology, organisations supporting the use of technology, NGOs and media. These organisations and societal groups will participate and enable the development of IE-related technology, the production of the technological artefacts, the regulation mechanisms for their diffusion, use and lifecycle support, etc. In addition, the availability of these social groups and the existence of contextual factors described in Section 5 would enable the consideration of the conceptual model of Fig. 2 as a niche of the IEI production system. In the process of development and maintenance of an industrial ecosystem, lessons will be learned concerning the overcoming of the barriers depicted in Table 1.

7. Three cases of eco-industrial parks as strategic niches 7.1. Kalundborg, Denmark 7.1.1. General description Kalundborg is the most celebrated example of industrial ecosystem (industrial symbiosis) and serves as a reference case for similar projects. The effort began in 1961 with a project to use surface water from a nearby lake for a new oil refinery, in order to save the limited supplies of groundwater. Today, the symbiosis which consists of six main partners (a power station, an oil refinery, a biotechnology company, a company which produces plasterboard for the building industry, a soil remediation firm and a waste management company) and the municipality of Kalundborg, has developed a networking cooperation based upon exchanges of energy and resources such as steam, heat, water, refinery gas, gypsum, biomass, liquid fertilizer, fly ash, sludge, etc.

E.D. Adamides, Y. Mouzakitis / Journal of Cleaner Production 17 (2009) 172–180

7.1.2. Process of formation, maintenance and technology diffusion Despite its impressive results, Kalundborg was not explicitly designed for demonstrating the benefits of industrial symbiosis. It was an uncoordinated response to groundwater scarcity which is generally claimed to be the motivation force that brought many of the partners together. Hence, by no means Kalundborg has been set up as an experiment. The endogenously initialised development was driven by pure economic rationale (to cut costs) and existed for quite a time without being identified as industrial symbiosis. An institutionalised platform (Center for Industrial Symbiosis) was developed later. No substantial governmental, or whatever direct protection has been offered. There was no master plan and each link between entities in the system was negotiated as an independent business deal and was finally established only if it was expected to be economically beneficial. The symbiosis has not been driven by a technological innovation, or by the use of an innovative technological artefact. The technology employed has been rather conventional, but its use can be considered innovative. Over the years, the degree of experimentation has been low – actors and material exchanges remained unchanged. Although the core of the system remains Kalundborg, the ‘‘inner symbiosis’’ has been extended to include outer companies (treatment of their wastes). 7.1.3. Preconditions – contextual factors The whole project was started from an ‘‘external’’ regime destabilising factor (the scarcity of water) within the boundaries of the Kalundborg industrial area that offered a common infrastructure and helped in cultivating relations (embeddedness) among companies. Kalundborg’s small size and relative isolation have made a community in which employees and managers interact socially with their counterparts on a regular basis. The financial success of the initial implementation of the industrial ecology concepts accelerated the overall effort. 7.1.4. Overall assessment of Kalundborg as a niche Undoubtedly, in terms of citations Kalundborg constitutes the most popular example of industrial symbiosis. Clearly, it is also a successful one in terms of the amount of waste reused and the energy saved. Nevertheless, the lessons it can provide for other would-be industrial ecosystems are limited, as it is limited its role as a niche for an IEI production model. Unless a similar mix of industries already co-exist in a specific area, it is very difficult to persuade heavy industries and power generating companies to move into a geographically confined area. It is also hard today, at least in the developed world, to create from scratch such an area in the framework of a regional development policy. In current development models, mature industries are accompanied by more dynamic ones, whereas on the other hand there is a tendency for renewable energy sources and decentralisation. Although Kalundborg demonstrates the role of embeddedness and social capital in its endogenously-driven development, at the same time it provides no solution to the issues of technology-dependent embeddedness. The participants of Kalundborg are mature industries where the tight coupling of processes and the absence of flexible technologies to facilitate changes are not of great concern. The technologies developed and the related technological artefacts seem to be appropriate only for the specific industrial system. In addition, no innovators and technology developers for commercialising technologies have been involved so far, since there was no need for them to be there. Moreover, no indirect process technology was developed, neither its use has been stressed. The same holds for novel IE-related management practices. As far as performance management systems are concerned, it is clear that in Kalundborg the principal role is played by financial measurements since each project has been principally judged by its economic rationale. As it has already been mentioned, the role of public

177

institutions, including those of learning and research, as well as that of related agencies has been only marginal, confined to tying the price system to pollution control measures. Overall, in our opinion, Kalundborg raises the question of whether market forces can lead industrial ecology-inspired development. Market forces are synonymous to arms-length relationships, whose prolonged maintenance implies a continuing mutual economic interest, a condition that holds only in mature industrial sectors. More dynamic settings require the development of social capital which is a prerequisite for the development of intellectual capital [43] and which, in turn, supports the required dynamism through continuous organisational and technological innovation. 7.2. The Heidelberg industrial estate and the Rhine-Neckar region, Germany 7.2.1. General description This industrial symbiosis effort started as a university-driven research project to test whether a Kalunborg-like symbiosis could be implemented elsewhere. In this way, it provided by its very basic objective a test on the value of Kalundborg as a niche. Eighteen companies, mainly SME’s, participated in the project after paying a small participation fee. For the majority of the participating companies it was the first time that they came so close to cooperate. The local government was supportive but not in financial terms. The initial phase of the project concerned the Pfaffengrund industrial estate where four types of inter-firm relations were implemented: direct (connecting/cascading processes), tender solutions (companies opening disposal paths to neighbours), joint transportation of used pallets, and informational coordination. The role of the latter has been especially emphasised in every phase of the Heidelberg project. After the completion of the initial phase of the project which was geographically restricted to the Pfaffengrund industrial estate boundaries (an area of 93 ha), an attempt was made to extend the eco-industrial activities at the Rheine-Neckar regional level (a 50 km  50 km area where larger companies such as BASF, Roche, ABB, etc. operate facilities [44]) as part of a regional development project. Towards this end, two actions have been implemented: (i) the formation of an adequate network structure between different actors as a means for discussion and preparation of coordinated actions, and (ii) a waste management information system to facilitate data exchange between participants. 7.2.2. Process of formation, maintenance and technology diffusion Clearly, the Heidelberg region project exhibits many of the characteristics of experiments (at least at the initial phases) and was driven by a clear vision whose attainment was implicitly and explicitly associated with technological innovation. Although at the beginning the project was focused on the development of interfirm relations and the use of standardised technology, the latter phases – especially at the regional level – were directed towards the development of indirect process technology (ICT) and human network development, as their contribution to the sustainability and success of the project was appreciated. As it is noted by Sterr and Ottt [21], ‘‘Pfaffengrund represents an indicative case of exogenously-driven change towards sustainable development embracing the concepts of industrial ecology’’. In addition to the university, other public organisations and institutions supported both projects in non-financial terms. 7.2.3. Preconditions – contextual factors There have not been any strong external regime destabilising factors that initiated and supported the Heidelberg and Rhine Neckar region eco-industrial development efforts. Neither there was an internal leading anchor tenant that drove the project. Nevertheless, the existence of a geographically confined industrial

178

E.D. Adamides, Y. Mouzakitis / Journal of Cleaner Production 17 (2009) 172–180

estate that contained a typical (not an ideal) mix of companies and other organisations contributed to the success of the project as a technological niche. The success of the initial phase led to the identification of the challenges and opportunities involved in larger eco-industrial development projects. 7.2.4. Overall assessment of Heidelberg/Rhine Neckar as a niche Clearly, the Heidelberg/Rhine Neckar project has a more technological flavour than Kalundborg and its performance can be assessed along this line. The setting is more typical involving organisations of diverse sizes and activities. The performance objective has always been to increase participation and diversity. It does not provide an example of a ‘‘fully working system’’, but it demonstrates many important elements of an IEI production system: the role of information and knowledge management, the importance of social capital and the role of indirect process technology, an issue much ignored in the industrial ecology discourse. The role of knowledge producing institutions is also emphasised and the drawbacks of technology-dependent embeddedness are revealed (a tender solution with fluorescent tubes stopped after the tube producer left the area). It also demonstrates the need for coordinated policies on the part of governments (as a result of the national policy for free disposal of packaging materials, the byproducts of a film role producer could not be used after regranulation from a plastics producer). Clearly, the absence of the appropriate level of (real) embeddedness contributed to the selfish, purely economic behaviour. Overall, it seems that the Heidelberg/ Rhine Neckar project has still a long way to go. However, it demonstrates some of the characteristics of an IEI production system technological niche and SNM, and it provides important lessons for more ambitious and organised efforts.

7.3. The Landskrona industrial symbiosis 7.3.1. General description Various companies of the Landskrona region had always been involved in sustainability projects, mainly along the lines of cleaner production, before the industrial symbiosis network was initiated in 2002 by researchers from the International Institute for Industrial Environmental Economics (IIIEE) at Lund University and representatives of local companies. The aim was to increase the competitiveness of the region’s firms by cutting costs, developing new products, defining new value streams and improving the environmental image through network collaboration. The funding came from NUTEK (Swedish Business Development Agency) and from the participating companies. Twenty-one companies of diverse size (from 4 to 500 employees) belonging to diverse sectors (metal works, chemicals, printing, auto parts, etc.) have been participating in the project together with three public agencies. Most companies already had certified management systems of quality, environment or both. That is, the environmental awareness of the participating organisations was already high. In addition, as many of the participants went through together and survived the depression years, a community and collaboration spirit was already present [45]. Through a designed process involving data collection and meetings, the network was established and the potential synergies were identified. The role of inter-organisational collaboration in technological innovation became apparent from the early stages of the programme and appropriate actions were taken to strengthen it [32]. The Swedish government realised the potential of industrial symbiosis for innovation too, and financed specific projects along this line. So far, the results from the Landskrona industrial symbiosis project are limited to the use of waste for district heating, to collective waste management, cooperation in transport and logistics, use of unprocessed waste, and use of

renewable energy technologies.

technologies

as

input energy efficiency

7.3.2. Process of formation, maintenance and technology diffusion The Lanskrona project represents a case of a combination of endogenous/exogenous industrial ecosystem development. As it is emphasised in the related literature, the initial priority of the Landskrona project was to develop the necessary organisational and inter-organisational infrastructure to develop and support the innovations necessary for the embedding of the local/regional industrial system into its natural and social context, i.e. to develop and sustain the system through the supply of appropriate resources. Particular emphasis has been given to the activities that facilitate the development and deployment of local knowledge for environmental innovation, namely collective problem definition, searching for technological innovation opportunities at the intersector interface (between collaborating firms that belong to different sectors), and increase inter-organisational interactions in environmental problem solving networks. So far, however, no innovative environmental technologies developed at Landskrona were reported. 7.3.3. Preconditions – contextual factors The Landskrona industrial symbiosis network exemplifies a typical case of planned, brownfield ecosystem of firms that are not collocated. The driving forces for the creation of the network were: (i) the declining economy and natural environment of the area as the result of an outdated development model, and (ii) the pressure exerted by the regulating authorities towards a more environmentally benign industrial activity. The process of design was highly structured and its implementation seems to be orderly and controlled. 7.3.4. Overall assessment of Landskrona as a niche Clearly, the Landskrona project emphasises more than the other two cases, the role of knowledge creation and use processes in the development and maintenance of competitive industrial ecosystems. In terms of participation, it includes most major social groups and related organisational entities (universities, industry representative, national and local government, etc.). More importantly, the role of technological and organisational innovation is embedded in its development and experimentation processes. So far protection is offered by government funding. Nevertheless, the Landskrona project is still in its early phases and a realistic assessment is very difficult to be made. Its aims and designs, however, are promising, and after these plans are implemented, it can easily contribute to the diffusion of the IEI industrial production system. 8. Discussion and conclusions Undoubtedly, the vision of industrial ecology, as it was described in the early writings of its pioneers, is highly influenced by implemented projects and cases such as the Kalundborg one. These cases resulted in a renewed and more scholastic interest in connecting industrial processes (an old well-documented practice), shaped the theory of industrial ecology, and initiated new efforts towards more planned and ‘‘designed’’ industrial ecosystems. However, so far, industrial ecology has failed to deliver and gain the acceptance it deserves as an innovative production model that attracts independent firms and other types of organisations [8]. In searching the reasons for its slow rise away from linear lists of barriers of different flavour and intensity, we have considered industrial ecology and industrial ecosystems from the technology studies’ systemic perspective. Through this lens, the formation of industrial ecosystems constitutes system innovation and defines a novel state of the industrial production socio-technical system (or

E.D. Adamides, Y. Mouzakitis / Journal of Cleaner Production 17 (2009) 172–180

a novel system, depending on how one defines transient behaviours). For policy makers, the objective in relation to the diffusion and adoption of this system as the current dominant industrial production regime is to identify and develop in cooperation with developers, producers and users, the necessary technologies and technological artefacts and the appropriate selection mechanisms within the use system, which will eventually select, in a co-evolutionary manner, the appropriate development and production activities that eventually will shift the current industrial production regime and will substitute it by an IEI one by eliminating the barriers depicted in Table 1. Above, it was mentioned how strategic niche management works as a policy instrument for shifting a technological regime in an organised and systematic way. It should be noted that, so far, most analyses using SNM have been done ex post, despite its promotion as a real-time monitoring tool [46]. In this line, after sketching a conceptual model for industrial ecosystem development and maintenance, we have examined three cases of industrial ecosystems from the niche creation and management perspective to discover whether they play, or have the potential to play, this role. Overall, the suitability and effectiveness of an existing industrial ecosystem as a technology niche can be assessed along two perspectives. On the one hand, the magnitude and diversity of the participating firms, governmental and non-governmental authorities and other organisations that form the network of social groups determine the niche’s potential to act as a micro-instantiation of the overall IEI production system. On the other, the processes accomplished for shifting the barriers of Table 1 and the associated diffusion mechanisms determine the effectiveness of the industrial ecosystem as an example of good practice worth copying. The three cases discussed in the previous section represent three different but complementary approaches to industrial ecosystems. Kalundborg is a rather unplanned system of symbiosis formed to overcome the shortage of natural resources and to survive in costdriven mature industrial sectors. The gains for sustainability followed the initial economic ones. The involvement and protection offered by government have been minimal, as it has been the involvement of knowledge producing institutions. As a result, the level of technological innovations produced in Kalundborg is rather insignificant. The lessons that Kalundborg spreads as a niche are more oriented towards the processes necessary for shifting cultural and psychological barriers to build embeddedness, especially when having in mind the local community and its well being. The use selection environment intuitively realised the value of cooperation for economic and environmental benefits, but the structure of the overall (socio-technical) system did not promote inherently the role of technological innovation and its associated processes. The stability of the sectors on which the particular companies belonged, eliminated this need. On the other hand, the intentions of Heidelberg and Landskrona have been more technology-oriented. The tradition of these industrial districts and the involvement of university and research institutions contributed to this orientation. Both projects have been planned efforts, the Landskrona one having a more structured development process. In addition, both projects spread along greater geographical areas than a limited industrial estate. In Landskrona, symbiotic operations were partially supported by governmental funds, while in the Heidelberg region the support has been more endogenous. Nevertheless, with the exception of some indirect process technology, the knowledge that both projects can offer so far to similar or potential efforts is confined to the planning of the departing stages. Communications of the experiences regarding the projects’ subsequent phases have to follow. None of the three industrial ecosystems presented above, neither any other, to our current knowledge, identifies itself as technological niche. Most projects seem eager to produce results in

179

terms of waste absorption and short-term economic benefits to justify investments, rather than to experiment with innovative technologies as an inherent need for survival and expansion. Nevertheless, complementary characteristics of niches can be found in different industrial ecosystems. The role of more organised niches and SNM is to shift the barriers imposed by the current production technology regime(s) and to cultivate the conditions for the diffusion of the IEI production system. Having this in mind, industrial ecosystem projects may act as niches if: a. They are part of a national or regional technology policy, in addition, or complimentary to regional development one. b. They include as much as possible diverse organisations whose activities span along the full spectrum of the three sociotechnical subsystems: knowledge producers and technology developers, technology producers, and users of technologies. c. Companies belong to different sectors that are open to competition. Appropriate resources and processes are gradually built into the system to stimulate innovation and manage technology-dependent embeddedness when the industrial ecosystem responds to changes in the external business environment. d. Users of technologies (hard and soft, direct and indirect) are encouraged to try novel technological solutions provided by other participants, or developed jointly. e. Incentives and protection are provided to developers, producers and users of innovative technologies. These incentives should be gradually substituted by supporting mechanisms for the diffusion of technologies. f. Inter-project communication and coordination mechanisms are developed to exchange knowledge concerning different aspects of the system. An industrial ecosystem/niche-to-be like Kalundborg will benefit from the experience of a niche involving more technology developers (universities and research institutes) that are present in, for instance, Landskrona. g. The niche perspective is present even if the initial objective of a project is regional development or revitalisation of a region. In any case such projects represent opportunities for development of, and experimentation with innovative technologies, as well as for testing novel organisation and management practices.

References [1] Tibbs BC. Industrial ecology: an environmental agenda for industry. Whole Earth Review 1992;(Winter)::4–19. [2] Lowe E, Evans L. Industrial ecology and industrial ecosystems. Journal of Cleaner Production 1995;3:47–53. [3] Boons F. Connecting levels: a systems view on stakeholder dialogue for sustainability. Progress in Industrial Ecology 2004;1(4):385–96. [4] Rip A, Kemp R. Technological change. In: Rayner S, Malone EL, editors. Human choice and climate change, vol. 2. Columbus, OH: Battelle Press; 1998. [5] Tukker A, Butter M. Governance of sustainable transitions: about the 4(0) ways to change the world. Journal of Cleaner Production 2007;15:94–103. [6] Lazonick W. Business organization and the myth of the market economy. Cambridge: Cambridge University Press; 1991. [7] Heeres RR, Vermeulen WJV, de Walle FB. Eco-industrial park initiatives in the USA and the Netherlands: first lessons. Journal of Cleaner Production 2004; 12(8–10):985–95. [8] Vermeulen WJV. The social dimension of industrial ecology: on the implications of the inherent nature of social phenomena. Progress in Industrial Ecology 2006;3(6):574–98. [9] Berkhout F. Technological regimes. Path dependency and the environment. Global Environmental Change 2002;12:1–14. [10] Kemp R, Rotmans J. Managing the transition to sustainable mobility. In: Elzen B, Geels F, Green K, editors. System innovation and the transition to sustainability: theory, evidence and policy. Cheltenham, UK: Edgar Elgar Publishing; 2004. [11] Geels FW. Technological transitions as evolutionary reconfiguration processes: a multi-level perspective and a case-study. Research Policy 2002;31:1257–74. [12] Kemp R, Schot J, Hoogma R. Regime shifts to sustainability through processes of niche formation: the approach of strategic niche management. Technology Analysis and Strategic Management 1998;10:175–95.

180

E.D. Adamides, Y. Mouzakitis / Journal of Cleaner Production 17 (2009) 172–180

[13] Piore M, Sabel C. The second industrial divide: possibilities for prosperity. New York: Macmillan; 1984. [14] Womack JP, Jones DT. Lean thinking. 2nd ed. New York: The Free Press; 2003. [15] Geels FW. From sectoral systems of innovation to socio-technical systems. Insights about dynamics and change from sociology and institutional theory. Research Policy 2004;33:897–920. [16] Sabel CF, Zeitlin J. Stories, strategies, structures: rethinking historical alternatives to mass production. In: Sabel CF, Zeitlin J, editors. World of possibilities. Flexibility and mass production in western industrialization. New York: Cambridge University Press; 1997. [17] Hamlin D. Flexible specialization and the German toy industry, 1870–1914. Social History 2004;29(1):30–40. [18] Kurz HD. Goods and bads: sundry observations on joint production, waste disposal, and renewable and exhaustible resources. Progress in Industrial Ecology 2006;3(4):280–301. [19] Von Hippel E. Democratizing innovation. Cambridge, MA: The MIT Press; 2006. [20] Slack N, Lewis M. Operations strategy. Harlow, UK: FT Prentice Hall; 2002. [21] Sterr T, Ott T. The industrial region as a promising unit for eco-industrial development – reflections, practical experience and establishment of innovative instruments to support industrial ecology. Journal of Cleaner Production 2004;12:947–65. [22] Nelson RR, Winter SG. An evolutionary theory of economic change. Cambridge, MA: Bellknap Press; 1982. [23] Tsoutsos TD, Stamboulis YA. The sustainable diffusion of renewable energy technologies as an example of an innovation-focused policy. Technovation 2005;25:753–61. [24] Malcolm R, Clift R. Barriers to industrial ecology: the strange case of ‘‘The Tombesi bypass’’. Journal of Industrial Ecology 2002;6(1):4–7. [25] Frosch RA. Industrial ecology: adapting technology for a sustainable world. Environment 1995;37(10):16–37. [26] Elzen B, Wieczorek A. Transitions towards sustainability through system innovations. Technological Forecasting and Social Change 2005;72:651–61. [27] Opoku HN. Policy implications of industrial ecology conceptions. Business Strategy and the Environment 2004;13:320–33. [28] Korhonen J, von Malmborg F, Strachan PA, Ehrenfeld JR. Management and policy aspects of industrial ecology: an emerging research agenda. Business Strategy and the Environment 2004;13:289–305. [29] Canie¨ls MCJ, Romijn HA. Strategic niche management as an operational tool for sustainable innovation: guidelines for practice. Paper for Schumpeter Conference, 2006. Also available from ; 2008. [30] Korhonen J. Two paths to industrial ecology: applying the product-based and geographical approaches. Journal of Environmental Planning and Management 2002;45(1):39–57.

[31] Lambert AJD, Boons FA. Eco-industrial parks: stimulating sustainable development in mixed industrial parks. Technovation 2002;22:471–84. [32] Mirata M, Emtairah T. Industrial symbiosis networks and the contribution to environmental innovation: the case of the Landskrona industrial symbiosis programme. Journal of Cleaner Production 2005;13:993–1002. [33] Chertow MR. Industrial symbiosis: literature and taxonomy. Annual Reviews of Energy and the Environment 2000;25:313–37. [34] Gibbs D, Deutz P, Proctor A. Sustainability and the local economy: the role of EIPs. Paper presented to the conference Ecosites and Eco-Centres in Europe, Brussels; 2002. [35] Mouzakitis Y, Adamides E, Goutsos S. Sustainability and industrial estates: the emergence of eco-industrial parks. Environmental Research, Engineering and Management 2003;26:85–91. [36] Eilering JAM, Vermeulen WJV. Eco-industrial parks: toward industrial symbiosis and utility sharing in practice. Progress in Industrial Ecology 2004;1: 245–70. [37] Uzzi B. Social structure and competition in interfirm networks: the paradox of embeddedness. Administration Science Quarterly 1997;42:35–67. [38] Zucchella A. Local cluster dynamics: trajectories of mature industrial districts between decline and multiple embeddedness. Journal of Institutional Economics 2006;2:21–44. [39] Coˆte´ R, Cohen-Rosenthal E. Designing eco-industrial parks: a synthesis of some experiences. Journal of Cleaner Production 1998;6:181–8. [40] Walter AI, Scholz RW. Sustainable innovation networks: an empirical study on interorganisational networks in industrial ecology. Progress in Industrial Ecology 2006;3(5):431–50. [41] Jacobsen NB. The industrial symbiosis in Kalundborg, Denmark: an approach to cleaner industrial production. In: Cohen-Rosenthal E, editor. Eco-industrial strategies. Sheffield, UK: Greenlief Publishing; 2003. [42] von Malmborg FM. Stimulating learning and innovation in networks for regional sustainable development: the role of local authorities. Journal of Cleaner Production 2007;15(17):1730–41. [43] Nahapiet J, Ghosal S. Social capital, intellectual capital, and the organizational advantage. Academy of Management Review 1998;23(2):242–66. [44] Van Berkel R. Regional resource synergies for sustainable development in heavy industrial areas: an overview of opportunities and experiences, Report for Australian Research Council. Perth, Australia: Curtin University of Technology; 2006. [45] Maltin M. Industrial symbiosis and its alignment with regional sustainability: exploring the possibilities in Landskrona, Sweden. In: IIIEE Reports 2004, vol. 17. Lund University; 2004. [46] Hommels A, Peters P, Bijker WE. Techno therapy or nurtured niches? Technology studies and the evaluation of radical innovations. Research Policy 2007;36:1088–99.