Scholarly article on topic 'Towards a broader perspective on the forms of eco-industrial networks'

Towards a broader perspective on the forms of eco-industrial networks Academic research paper on "Social and economic geography"

CC BY-NC-ND
0
0
Share paper
Academic journal
Journal of Cleaner Production
OECD Field of science
Keywords
{"Industrial networks" / "Industrial symbiosis" / "Environmental sustainability" / "Systematic literature review"}

Abstract of research paper on Social and economic geography, author of scientific article — Samuli Patala, Sari Hämäläinen, Anne Jalkala, Hanna-Leena Pesonen

Abstract This paper explores various forms of eco-industrial networks in advancing environmental sustainability. Prior research on environmental sustainability primarily identifies industrial actors as autonomous entities or considers the role of networks in advancing environmental sustainability from a rather narrow perspective. However, the networks of ties in which industrial firms are embedded profoundly impacts not only their own performance, but also the natural environment in which they operate. Based on a systematic literature review, we identify four forms of eco-industrial networks that have the potential to advance environmental sustainability: 1) symbiosis networks, 2) sustainable supply networks, 3) environmental issue networks and 4) environmental solution networks. The paper presents important insights on the operational logic for each of these network forms and the dimensions of their network architecture. The main implications of this comparison are that policy-makers and practitioners need to become aware of the various mechanisms through which inter-organisational networks can reduce environmental load. Furthermore, there is a need to build broad coalitions of organisations that are mobilised to address environmental issues. We suggest that network architecture which maximises its members' capability to self-organise while also including a coordinating organisation can be highly suitable for eco-industrial networks. We also suggest some fruitful avenues for future research on eco-industrial networks. In integrating research on eco-industrial networks with existing research on inter-organisational alliances and networks, the paper provides more understanding on the multifaceted role played by various forms of eco-industrial networks in advancing sustainability.

Academic research paper on topic "Towards a broader perspective on the forms of eco-industrial networks"

Contents lists available at ScienceDirect

Journal of Cleaner Production

journal homepage: www.elsevier.com/locate/jclepro

Towards a broader perspective on the forms of eco-industrial networks

CrossMark

Samuli Patala a' ■1, Sari Hämäläinen b, Anne Jalkala a, Hanna-Leena Pesonen

a School of Industrial Engineering and Management, Lappeenranta University of Technology, P.O Box 20, Lappeenranta FI-53851, Finland b Jyväskyla School of Business and Economics, University of Jyvaskyla, P.O Box 35, FI-40014, Finland

ARTICLE INFO

Article history: Received 23 May 2013 Received in revised form 6 March 2014 Accepted 18 June 2014 Available online 26 June 2014

Keywords: Industrial networks Industrial symbiosis Environmental sustainability Systematic literature review

ABSTRACT

This paper explores various forms of eco-industrial networks in advancing environmental sustainability. Prior research on environmental sustainability primarily identifies industrial actors as autonomous entities or considers the role of networks in advancing environmental sustainability from a rather narrow perspective. However, the networks of ties in which industrial firms are embedded profoundly impacts not only their own performance, but also the natural environment in which they operate. Based on a systematic literature review, we identify four forms of eco-industrial networks that have the potential to advance environmental sustainability: 1) symbiosis networks, 2) sustainable supply networks, 3) environmental issue networks and 4) environmental solution networks.

The paper presents important insights on the operational logic for each of these network forms and the dimensions of their network architecture. The main implications of this comparison are that policymakers and practitioners need to become aware of the various mechanisms through which inter-organisational networks can reduce environmental load. Furthermore, there is a need to build broad coalitions of organisations that are mobilised to address environmental issues. We suggest that network architecture which maximises its members' capability to self-organise while also including a coordinating organisation can be highly suitable for eco-industrial networks. We also suggest some fruitful avenues for future research on eco-industrial networks. In integrating research on eco-industrial networks with existing research on inter-organisational alliances and networks, the paper provides more understanding on the multifaceted role played by various forms of eco-industrial networks in advancing sustainability.

© 2014 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND

license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

1. Introduction

Firms and industries do not accomplish the goals of environmental sustainability in isolation. Organisations are embedded in networks of social, professional and exchange relationships with other organisational actors (Ahuja et al., 2012; Gulati et al., 2000). In addition, problems relating to environmental sustainability are embedded in a complex web of actors comprising businesses, consumers, NGOs and governmental agencies which are involved in collaboration and contestation concerning the problems and their solutions (Wittneben et al., 2012). However, in the extant literature on organisations and the environment, the unit of

* Corresponding author.

E-mail addresses: samuli.patala@lut.fi (S. Patala), sari.m.hamalainen@jyu.fi

(S. Hamaläinen), anne.jalkala@lut.fi (A. Jalkala), hanna-leena.pesonen@jyu.fi (H.-L. Pesonen).

1 Tel.: +358 40 6853840.

analysis primarily lies at the level of individual actors instead of a network of actors. As the locus of change primarily lies in the network of interactions between organisations and individuals, we need a better understanding on the linkages between environmental issues and organisational networks. This paper advances the shift of focus in the unit of analysis by examining eco-industrial networks, defined here as industrial networks that advance environmental sustainability through inter-organisational collaboration.

However, research on eco-industrial networks is surprisingly limited and has focused primarily on industrial symbiosis (IS), which refers to inter-firm activities that focus on the re-utilisation of waste and by-products and the exchange of resources (e.g. Chertow, 2000; Doménech and Davies, 2011). While the IS literature focuses on the potential of energy and material linkages between firms, it provides a limited perspective on the forms of eco-industrial networks in advancing sustainability. Thus, we suggest that a broader understanding is needed on the mechanisms

http://dx.doi.org/10.1016/jjclepro.2014.06.059

0959-6526/© 2014 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

through which industrial networking can advance environmental sustainability. Analysis of the operational logic and network architecture of eco-industrial network forms could shed light on the processes advancing environmental sustainability. Operational logics of networks describe the focus of environmental action and the mechanisms through which a network aims to achieve environmental benefits. Network architecture represents the form of networks in response to their function (Fjeldstad et al., 2012). A better understanding on the underlying structural elements of these networks is needed to gain a deeper understanding on how these networks operate. Our main objective in this research is to make salient the potential mechanisms that have been addressed in the extant literature and establish a starting point for further empirical work. Specifically, we address two research questions: What forms of eco-industrial networks with the potential to advance environmental sustainability can be identified in the literature? What are the principal operational logic and architecture of these network forms? We answer these questions through a systematic literature review by analysing four different literature streams.

The paper begins with a general discussion on networks and sustainability, followed by an introduction to four forms of eco-industrial networks that have been identified through a systematic literature review. Subsequently, the identified forms of eco-industrial networks are compared and their similarities and differences evaluated through dimensions adopted from the literature on inter-organisational alliances and networks. The paper ends with conclusions and implications for practice and research.

2. Networks and sustainability

There is an extensive body of knowledge on how networks can contribute to the creation of various types of outcome. Networks enable organisations to access resources that might otherwise be difficult to develop or acquire (Ahuja et al., 2012; Gulati et al., 2000). They transfer information that gives rise to attitude similarity, imitation and generation of innovations (Ahuja, 2000; Brass et al., 2004). Thus, networks are powerful carriers of new norms, values and practices. In addition, they serve as governance mechanisms that can constrain opportunism and enhance trust (Ring and Van De Ven, 1992).Conversely, networks also have inertial properties that can constrain change (Kim et al., 2006).

Networks are potentially powerful tools with which to influence the context and shape the practices of involved actors. However, we have limited understanding on the transformative power of networks in addressing the problem of environmental sustainability. Prior network research primarily focuses on the outcomes of networks for individual firms and industries, rather than on the functioning of the natural environment. From the perspective of an individual firm, network-based collaboration has been shown to reduce risk, speed products to market and decrease the cost of process improvement and product development (for a review, see Fjeldstad et al., 2012).

Environmental sustainability, however, is profoundly affected by complex networks of actors that comprise industries, NGOs and governmental agencies. These networks are involved in collaboration and contestation over the urgency of environmental problems and the role of government and markets in addressing these problems (Wittneben et al., 2012). In addition, specific forms of collaborative networking can contribute to reducing the environmental load of industrial operations. For example, Ostrom (2009) regards environmental degradation as a problem of collective action and proposes a polycentric approach that relies on small-scale regional institutions and governance mechanisms which take advantage of local incentives for cooperation.

Collaborative inter-organisational networks are often characterised by novel ways of organising that suit the collective purpose of the network (Fjeldstad et al., 2012). It is thus vital to understand these networks' architecture, which can be defined as the synthesis of their form in response to their function. These collaborative arrangements can be conceptualised through an actor-oriented network architecture, which is focused on enabling the set of involved organisations to dynamically form collaborative relationships. Actor-oriented network architecture comprises three major elements: actors, commons and also protocols, processes and infrastructures (Fjeldstad et al., 2012). Furthermore, a study by Gulati et al. (2012) identified two other elements through which a network's architecture can be characterised. The first element is the permeability of boundaries; in other words, the extent to which a network's boundaries are open or closed. Also, the internal stratification of decision making affects a network's design (Gulati et al., 2012).

Research on business and sustainability appears to draw surprisingly little from the substantial literature on inter-organisational alliances and networks. We propose that research on eco-industrial networks will benefit from better integration with traditional network research. In addition, more understanding on the potential and characteristics of network cooperation in addressing environmental problems is needed.

3. Methodology

As the existing research on industrial networks and sustain-ability is divided into mostly unrelated research streams, we adopted a multidisciplinary systematic literature review to address our research questions. A systematic literature review can provide a more comprehensive analysis of the literature in comparison to traditional, more subjective literature reviews (Denyer and Tranfield, 2006). It includes a clear statement on the purpose of the review, a comprehensive search protocol to obtain the relevant literature and explicit criteria for qualifying the relevant literature. Its strength lies in reproducibility due to careful documentation of the search protocol, thus enabling other researchers to generate similar findings by following the same protocol (Denyer and Tranfield, 2006).

First, the search was conducted by two researchers on two different academic databases to cover a diverse range of publications. The two chosen databases were Scopus and Web of Science. The date range chosen for analysis was 1990-2012, and the selected keywords were as follows:

("Industrial" OR "Business" AND "Network" AND "Sustainability" OR "Environmental") OR "eco-industrial network" OR "industrial symbiosis" OR "eco-industrial park" OR "eco-cluster" OR("indus-trial ecology" AND "network")

Both databases were searched individually with the chosen keywords. The search covered the title, keywords and abstracts of the articles and found a total of 808 articles. Based on a review of their abstracts, 160 articles were qualified and their full texts scanned. This resulted in 36 articles being chosen for further analysis. An additional three articles were identified through a snowball method of scanning the references of the 36 selected articles, bringing the total number of articles to 39. After this process, a need for more recent articles on some eco-industrial network forms became apparent. For this reason, selected journals from industrial ecology, supply chain management and marketing, which are named in the next chapter, were scanned for the years 2012 and 2013. Along this process, two more articles matching the employed criteria were found. Thus, in total, 41

articles are included in the final review. The details of the inclusion and exclusion criteria employed to qualify the papers for analysis are listed in Table 1.

The final set comprised peer reviewed journal articles in English for which full texts were available. We focused on conceptual, normative and descriptive articles that put forward a direct contribution to the theoretical and/or operational understanding on industrial networks. As the purpose of this literature review is to advance understanding on the different forms of industrial network in advancing sustainability, we chose to include articles that would provide as comprehensive a perspective on these network forms as possible, thereby enabling us to compare the characteristic dimensions of the identified forms. The systematic review process is shown in Fig. 1.

The full set of 41 articles was analysed in detail by two researchers, both of whom gathered data on the articles on a predefined data collection form. The gathered data comprised study-type methodological factors (i.e. research method; data set; data analysis), objectives of the study, theoretical background, country of study, and the main concepts and primary findings. In the next section, we put forward a detailed description of the articles chosen for the final analysis.

4. Description of the literature

The majority of the articles that met the inclusion criteria (78%, n = 32) were published between 2007 and 2012, which demonstrates the considerable academic interest on eco-industrial networks in recent years. The three most common journals represented in our article set were the Journal of Cleaner Production (34%, n = 14), the Journal of Industrial Ecology (24%, n = 10) and Business Strategy and the Environment (12%, n = 5). In addition to two other journals in the area of environmental management (i.e. Ecological Economics; Annual Review of Energy and the Environment), the journals represent the fields of industrial marketing (Industrial Marketing Management; Journal of Business and Industrial Marketing), geographical sciences (Geoforum; Local Economy), supply chain and operations management (Journal of Supply Chain Management; Journal of Operations Management; International Journal of Production Research) and knowledge management (Industrial Management and Data Systems).

After careful analysis of the main concepts and theoretical backgrounds of the selected articles, we were able to identify four distinct literature streams with a common focus on industrial networks that aim to decrease environmental impacts through cooperative action. However, there are key differences in the operational logics and network structures among the networks described by these four streams. In essence, the four streams represent various forms of eco-industrial networks. The first, an industrial symbiosis network, has received the greatest attention over recent years in the fields of industrial ecology and environmental management. According to the review, it seems that industrial symbiosis has developed through exchange of material based resources (e.g. waste and by-products) to exchange of nonmaterial based resources (e.g. knowledge), and social embedded-ness has been highlighted in the discussion only during recent

Table 1

Inclusion and exclusion criteria.

Inclusion criteria Exclusion criteria

Journal articles; Conference proceedings;

Full text available; Reviews, commentaries, editorials;

Conceptual, normative, and descriptive Specific technical models of networks;

studies that advance understanding No abstract available;

on industrial networks. Non-English articles.

Fig. 1. Summary of the systematic review.

years. The second form, a sustainable supply network, is grounded on the supply chain management discipline. The third and fourth network forms, an environmental issue network and an environmental solution network, have received less attention and have been studied mainly in the field of industrial marketing, in which industrial and business networks have a long research tradition. We will next discuss each of these models in more detail.

4.1. Industrial symbiosis networks

Industrial symbiosis (IS) focuses on the cooperative management of resource flows through firms' networks. IS has been defined as "traditionally separate industries in a collective approach to competitive advantage involving physical exchange of materials, energy, water, and by-products" (Chertow, 2000). In addition, exchange of non-material resources such as knowledge and expertise is central in IS (Lombardi and Laybourn, 2012).Thus, key issues that relate to IS are the exchange of resources (i.e. material and nonmaterial), geographic proximity of actors and collaboration between industries. Geographic proximity has been regarded as central to IS due to facilitation of material exchanges, transportation, trust and collaboration (Lombardi and Laybourn, 2012; Taddeo et al., 2012), and also the sharing of information and norms within social networks (Ashton and Bain, 2012). Repeated interactions lead to the creation of shared norms that influence actors' behaviour and patterns of relationships (Ashton and Bain, 2012).

Symbiotic collaboration between firms and industries yields financial and environmental benefits through the exchange of complementary resources (e.g. Behera et al., 2012; Dimitrova et al., 2007; Tudor et al., 2007). Companies can also share utilities such as energy, water and wastewater treatment, and also services such as transportation, landscaping and waste collection (Ashton, 2008). Furthermore, IS aims at sustainable strategies in industrial development (Baas and Boons, 2004). Creating and sharing knowledge through IS networks facilitates implementation of modern technologies, eco-innovations and cultural change in organisations (Behera et al., 2012; Lombardi and Laybourn, 2012; Mirata and Emtairah, 2005). For instance, as a comprehensive state policy, the Chinese government has implemented a so-called circular

economy that relies on cleaner production, development of eco-industrial parks and ecological modernisation; for example, an eco-city (Yuan et al., 2006).

IS is a subset of industrial ecology that examines the sustain-ability of material and energy flows and cycles through industrial systems. It encompasses studies on eco-industrial parks (see e.g. Gibbs and Deutz, 2005; Sakr et al., 2011; Tudor et al., 2007) and industrial ecosystems (see e.g. Ashton, 2008; Liu et al., 2012). Some authors have employed related terms such as eco-industrial network (Ashton and Bain, 2012; Van Ha et al., 2009), eco-cluster (Dimitrova et al., 2007) and green industry park (Tudor et al., 2007).

4.2. Sustainable supply networks

A sustainable supply network is a form of eco-industrial network that focuses on taking full advantage of by-products and reusable materials while minimising waste in the supply chain (Zhu and Cote, 2004). Sustainable supply networks, also termed green supply chain management (GSCM), have their origin in the supply chain management literature. Sustainable supply chains aim to balance environmental concerns with performance across the supply chain, leading to better operational efficiency and less wasted resources, with the ultimate goal of sustainable competitive advantage (Tudor et al., 2007). Activities extend over the entire life-cycles of products, and comprise cooperation with suppliers for waste reduction, improved efficiency in manufacturing and transport and the development of reverse supply chains for recycling and reusing used products (Zhu and Cote, 2004).

Bansal and McKnight (2009) compared GSCM and IS as two distinct strategies for the reuse of industrial waste and by-products. They identified a key difference in the strategic logic between IS and GSCM actors. In conventional supply chains, firms manufacture and design products to meet customer needs and identify strongly with the end products that they sell, and seek to minimise waste and emissions along the supply chain. In contrast, firms in IS networks have a more entrepreneurial mindset, seeking new opportunities to extract value from waste and by-products, which might be completely independent from the firms' main identities. The reuse of waste and by-products in IS networks is focused on the system-level reduction of environmental impacts, while, in GSCM, it is more focused on single firms and value chains (Bansal and McKnight, 2009).

Zhu and Cote (2004) identified GSCM as being important for the development of localised eco-industrial parks, provided that the integration of GSCM begins to form a web-type structure instead of concentrating on the supply chains of single products. They found that an integrated approach to GSCM bears many similarities to the formation of eco-industrial networks (Zhu and Cote, 2004). Seuring (2004) also referred to an integrated approach to supply chain management and found that it requires wider cooperation and networklevel environmental goals among actors along the value chain compared to traditional supply chain management. However, there remain focal firms which are largely responsible for the coordination of activities in the supply chain. Vachon and Klassen (2007) studied how supply chain integration relates to focal companies' selection of environmental technologies in which to invest. They found that integration with downstream suppliers is positively linked to a larger share of investments in pollution prevention technologies (e.g. product and process innovation that decreases the environmental footprints of products), whereas integration with upstream suppliers is positively linked to a larger share of investment in pollution control technologies (e.g. end-of-pipe solutions to decrease environmental impacts). Kocabasoglu et al. (2007) found that developing reverse supply chain activities, which relate to decreasing environmental impacts, should not be conducted in isolation; instead, they should be

integrated with the development of forward supply chain activities such as demand and order management.

4.3. Environmental issue networks

The third form of eco-industrial network identified from the literature was an environmental issue network. Environmental issue networks refer to relatively loose collaborative coalitions formed around specific issues such as environmental problems or policies. They aim for institutional change through collective action and focus on developing policies, norms and values among the network actors. They are often temporary in nature, varying in length depending on the life-cycle of the focal issue. They also involve a diverse set of actors with asymmetrical power and resources, comprising private sector firms, governmental authorities, NGOs and even powerful individuals (Ritvala and Salmi, 2010, 2011).

Ritvala and Salmi (2010) studied how environmental issue networks are formed through mobilisation. They describe three separate initiatives that aim to improve the state of the Baltic Sea, a contemporary environmental concern that requires the mobilisation of actors from the region's various countries to address the issue. Ritvala and Salmi (2010) conceptualise environmental issue networks to be triggered by actors with strong values for taking responsibility to solve the focal issue. These actors act as mobilisers which challenge other actors to join the network through high media visibility and public framing. This mobilisation is enabled by social capital and shared values among the actors and, while their cooperation aims at environmental benefits, participating firms can also gain business benefits and improved public image (Ritvala and Salmi, 2010). In another study by Ritvala and Salmi (2011), motives of target firms to participate in issue networks are explored. It seems that, here, individual and organisational values and also relationships play important roles. Furthermore, network benefits such as potential new business opportunities and partners are good motivators (Ritvala and Salmi, 2011).

Another example of environmental issue networks is presented in Andersson and Sweet's (2002) paper in which the role of a mobiliser in the development process of a recycling system was studied in a food retail chain in Sweden. They highlight the mobiliser's role as a bridging point between different actors. Furthermore, there are several relationship changes between participating organisations during the development process, which lead to changes in organisational structures, exchanges of information and knowledge, and administrative routines. Thus, due to the activity of a dominant change agent, recycling ideas spread further into the food retail system and to other business networks (Andersson and Sweet, 2002).

Veal and Mouzas (2010) examine environmental issue networks on a global scale, focusing on the global network that addresses climate change. This network comprises various actors including governments, private sector businesses, research institutions, NGOs, international governance bodies and business associations. They found that environmental issue networks can be hindered by various cognitive biases that act as barriers to collaboration. Three processes were identified through which collaboration between actors develops: 1) the development of cognitive frames, 2) negotiation with each other and 3) attempting wise trades that lead to joint gains among the actors. The barriers to each of these processes need to be overcome for successful collaboration.

4.4. Environmental solution networks

The fourth identified network form was an environmental solution network, which refers to a collaborative approach that involves combining knowledge, technologies and other resources across organisational boundaries to create an eco-efficient solution. These

Table 2

Comparison of the operational logic and network architecture of eco-industrial network forms.

Industrial symbiosis networks

Sustainable supply networks

Environmental issue networks

Environmental solution networks

Operational logic Focus of environmental action

Drivers

Improving eco- efficiency of Decreasing environmental

production through by-product and footprints of products through life-

waste reuse, utility and service sharing, information exchanges.

Economic and environmental benefits; legislation; personal values.

Network architecture

Actors

Commons

Protocols, processes, infrastructures

Complementary industrial firms from various industries; authorities; NGOs.

Shared knowledge, and tangible resources and/or energy.

Formal exchange of resources, shared norms, social embeddedness.

Stratification of Ranges from hierarchical (i.e. decision making planned) to heterarchical (i.e. self-organising)

Permeability of Regionally constrained. boundaries

cycle thinking.

Consumer demands; stakeholder pressures; legislation; personal values.

Industrial firms, typically from the same supply chain; vertical connections.

Shared GSCM information and product life-cycle knowledge.

Formal mechanisms for flows of information and materials; standard operating practices; cooperative green actions.

Mostly hierarchical with formal supply chain relations.

Supply chain constrained.

Pollution control/prevention Developing eco-innovative

measures and increased awareness solutions that integrate the

on environmental challenges resources and capabilities of

through collaborative projects. multiple network actors.

Existing environmental challenges; Acquisition of knowledge on new

personal and organisational values; environmental solutions (i.e.

network benefits. technologies and/or services);

personal values.

Industrial firms; municipalities; authorities; NGOs; research organisations.

Shared knowledge and potential other intangible assets (e.g. legitimacy and brand).

Mobilisation of actors; negotiations; formal agreements.

Typically heterarchical, but can include hierarchical aspects.

Horizontally and vertically connected actors with complementary resources, mostly industrial firms, their customers, and research institutes.

Shared knowledge; potentially shared intellectual property rights and technology.

Mobilisation of network resources; co-development; technology adaptation.

Typically, hierarchical management by hub firms.

Geographical constraint depending Constrained by core technology/ on scale of the issue (i.e. local or solution. global); temporal limitations.

solutions often require a combination of products and services from several suppliers that are combined, usually through a central actor which acts an integrator, to create a new marketable solution. The benefits of technological innovation and integration can lead to an offering that has lower environmental impacts in comparison to its alternatives (Baraldi et al., 2011).

Baraldi et al. (2011) examined the Leaf House project and its accompanying network, which is an example of a technological network that promotes eco-sustainability. The Leaf House project focused on developing the first zero-emission house in Italy, which was a large step forward in terms of energy efficiency relative to existing alternatives. The project was coordinated by the Loccioni group, which acted as the technological integrator. The project comprised 80 involved partners, ranging from suppliers of standardised products to full subsystem suppliers. Some of the involved partners were Whirlpool (i.e. energy efficient appliances), IKEA (i.e. sustainably produced furniture) and Cisco Systems (i.e. ICT systems for smart energy usage) (Baraldi et al., 2011).

The project, while serving the main function of an integrated solution that provided revenue for the involved network actors, also served other purposes. Several of the more prominent network members could utilise the Leaf House in its development phase as a testing facility for new environmentally friendly technologies. In addition, many of the network members utilised the Leaf House project as a PR tool to boost their corporate sustainability image. The study also highlighted the concept of embedding technology for the successful operation of the network. To successfully develop an integrated solution, the involved network needed to evolve, through actions such as co-development and mutual technology adaptation, to diffuse the new technology (Baraldi et al., 2011).

5. Comparison of network forms

In this section, we compare the four identified forms of eco-industrial networks to identify their differences and similarities.

A detailed comparison is shown in Table 2. Based on the analysed literature, we found key differences and similarities in the operational logic and also in the architecture of the network forms. The proposed categorisation shows the perspective, (and the potential bias), of the extant literature regarding the forms of eco-industrial networks. As it is based on an emerging field of research, the categorisation is suggestive rather than exhaustive. The typology should be perceived as an initial step towards shedding light on the mechanisms through which industrial networking can advance sustainability. It makes salient the potential mechanisms that have been addressed in the extant literature and establishes a basis for further empirical work.

The identified network forms are empirically overlapping although analytically distinct as they differ fundamentally in terms of the primary logic through which they seek to advance environmental sustainability. While the network forms might facilitate advancement of similar outcomes, each form advances environmental objectives through primarily different operational logic, or the mechanisms through which it pursues environmental benefits. For example, all four network forms have the potential to facilitate waste reduction. In industrial symbiosis networks, this objective is achieved primarily by turning waste and by-products into material inputs. In sustainable supply networks, the same objective can be achieved by implementing waste minimisation standards across the supply chain. In environmental issue networks, the object can be addressed by creating awareness on waste management and, in environmental solution networks, by facilitating the co-development of new technologies and solutions for waste reduction. A combination of operational logics can also be present in some networks. For example, recent work on industrial symbiosis networks suggests that simultaneous issue networks can form as the network evolves and industrial symbiosis becomes a collective network-level goal. However, as suggested by the identified articles on issue networks, these networks can take various forms that are not always focused on resource exchanges and, thus, we classify them as a separate network form.

The architecture of the network forms are compared through the three dimensions that comprise an actor-oriented architecture: actors, commons and also protocols, processes and infrastructures (Fjeldstad et al., 2012). Actors are the members of the network who have the capabilities and values to self-organise. Actors can be organisations such as private firms or key individuals. Commons are the tangible or intangible resources that are collectively owned and available to the actors. One example of a common resource is an information system that facilitates network activities. Intangible common resources can include standards for network activities or a brand associated with the network. Lastly, a network requires protocols, processes and infrastructures that enable collaboration. Protocols are the codes of conduct employed by the actors in their collaborative activities, while infrastructures are the systems that connect actors. In addition, various network-level processes can be utilised to coordinate the network's activities (Fjeldstad et al., 2012). In addition to these three dimensions, two additional ones are employed to describe network architecture: the permeability of boundaries and decision-making stratification (Gulati et al., 2012). The extent to which a network's boundaries are open or closed can vary greatly in collaborative arrangements. A network can be open to all relevant actors or membership can be restricted by some criteria. A collaborative supply network is an example of a restricted network in which membership is typically limited to industrial firms involved in the network's economic activities. Open network models include, for example, open-source software projects such as Wikipedia or business ecosystems formed around mobile operating systems such as Android or iOS. Lastly, the internal stratification of decision-making affects a network's design. Many networks have some structural hierarchy. This is typical, for example, of networks in which an integrator is in charge of aggregating the innovative efforts of multiple members. Stratification helps reduce the complexity of coordination by dividing the collective into smaller subgroups that can specialise in specific activities. Conversely, heterarchical networks with low decision-making stratification give all members similar or overlapping rights or responsibilities, resulting in a "community of equals". This can increase members' sense of ownership and commitment to network activities, but can also discourage task specialisation and make coordination of the network more difficult (Gulati et al., 2012). In the following, we compare the operational logic and architecture of the four network forms in detail.

5.1. Operational logic

The analysed forms of eco-industrial networks have primarily different operational logics. These range from material reuse and reduction of environmental footprints through life-cycle thinking to taking collective action that addresses problems and generates eco-innovations.

Industrial symbiosis (IS) networks concentrate mainly on the exchange of material resources, although participating companies can also share utilities and services (Ashton, 2008), and also information (Chertow, 2007; Chertow and Ehrenfeld, 2012; von Malmborg, 2004; Van Ha et al., 2009). Collaboration in symbiosis networks can yield financial and environmental benefits (e.g. Behera et al., 2012; Dimitrova et al., 2007; Tudor et al., 2007). In fact, economic advantages are essential for IS as, without them, companies are not motivated to participate and develop symbiotic relationships. Opportunities to improve competitiveness by means of IS networks include the reduction of operating costs and/or increasing revenues through more efficient use of materials and energy (e.g. by-product sale; waste minimisation), and innovative product and process changes (Dimitrova et al., 2007; Lombardi and Laybourn, 2012; Taddeo et al., 2012; Tudor et al., 2007). Environmental benefits such as increasing efficiency of resource use, reducing emissions and

eliminating waste are also important drivers of IS (Ashton and Bain, 2012; Chertow, 2007). This helps participating firms to gain legitimacy and comply with environmental legislation (Ashton, 2011; Dimitrova et al., 2007). Governmental policy that supports and enables waste exchange is important for the development of IS networks (Chertow, 2007; Domenech and Davies, 2011; Gibbs and Deutz, 2007).

Sustainable supply networks are mainly driven by stakeholder pressures and regulations for manufacturers to improve the eco-efficiency of their operations and develop reverse supply chains to manage used products (Bansal and McKnight, 2009). Stakeholder demands from governmental agencies, NGOs, employees and environmentally conscious customer segments bring pressure to bear on companies to introduce measures aimed at reducing environmental impacts across the supply chain (Vachon and Klassen, 2007). The main focus of sustainable supply networks is to decrease the environmental footprints of products through life-cycle thinking. While symbiosis networks focus on waste reduction across entire systems of firms, the focus in sustainable supply networks is usually on firms' waste reduction through product and process innovations. Thus, activities of sustainable supply networks, such as pollution prevention within a single firm, can be counterproductive for IS activities as a potentially valuable resource can be reduced beyond reusable levels (Bansal and McKnight, 2009). Similarly, governmental regulation mechanisms designed from a sustainable supply networks' perspective, such as restrictions for the handling and transportation of waste, can be a hindrance to IS activities (Bansal and McKnight, 2009).

In the analysed literature, it is possible to find two distinct operational logics of environmental issue networks. The Baltic Sea initiatives described by Ritvala and Salmi (2010, 2011) were mostly formed through the mobilisation of actors which chose to voluntarily participate in the projects, driven by existing environmental challenges and also personal and organisational values. In contrast, the global climate change network (Veal and Mouzas, 2010) comprises actors which are not necessarily involved to promote collective environmentally sustainable action but rather to promote their self-interests, such as keeping the restriction of CO2 emissions as small as possible to minimise short term costs in respect of pollution prevention. Common to both of these network examples is their focus on pollution prevention and control measures concerning a specific environmental issue. Furthermore, collaboration with a diverse set of actors can also result in business benefits, such as new business opportunities, to participating firms (Ritvala and Salmi, 2011). It also facilitates the alteration of actors' mindsets (Andersson and Sweet, 2002).

The focus of environmental solution networks is to develop integrated solutions that decrease environmental impacts in comparison to existing alternatives. The activities in the network include co-development, knowledge sharing, joint adaptation and also standard market transactions to achieve technology embedding and integration. Integration can increase environmental sustainability due to a higher focus on the service element instead of physical goods, as highlighted by the literature on product-service systems (e.g. Morelli, 2006). Collective creation of product, process or service innovations is the primary operational logic of environmental solution networks, in which involved actors can benefit from the combination of complementary firms' resources, improved fit of solutions and customer preferences, and enhanced user adoption and public image (Baraldi et al., 2011; Chesbrough, 2003).

5.2. Network architecture

5.2.1. Actors

The key actors in IS networks are participating industrial firms, often from different industries. Heterogeneity between participating

companies facilitates the development of IS by creating more opportunities to find suitable partners (Dimitrova et al., 2007; Taddeo et al., 2012). Furthermore, a dense web of relationships is typical of IS (Bansal and McKnight, 2009). In addition, governmental organisations and NGOs might be involved as knowledge providers or coordinators.

Two key factors differentiate the types of actor involved in sustainable supply networks and IS. Supply networks are usually regarded with a vertical orientation of the flows of physical goods downstream in the chain, and used products and components upstream in reverse logistics, funds upstream and information throughout the chain (Bansal and McKnight, 2009). Although supply chains involve multiple customers and suppliers in a larger supply network, industrial firms generally have a smaller subset of strategically important suppliers and customers and, thus, supply networks are usually conceptualised in a linear manner (Bansal and McKnight, 2009).

Environmental issue networks differ from the other identified network types by encompassing a broader variety of involved actors. In addition to private firms from different sectors, environmental issue networks typically comprise public authorities, NGOs, interest associations and research organisations (Veal and Mouzas, 2010). This results in a complex network structure with potentially conflicting interests (Veal and Mouzas, 2010).

Environmental solution networks are often headed by a hub firm which acts as an integrator for the technological solution and brings together the products, services and subsystems provided by various suppliers and partners (Baraldi et al., 2011). The hub firm can integrate actors vertically across the supply chain and also horizontally to obtain complementary products and services for the integrated offering. In addition to industrial firms, other actors such as research organisations and customers might be involved in technological development.

5.2.2. Commons

Typically, IS networks have shared knowledge bases that enable actors to find new opportunities and potential new partners for exchange (Chertowand Ehrenfeld, 2012). Commons can also include tangible resources through the pooled use and management of shared resources such as energy and water (Chertow, 2007).

Sustainable supply networks include the sharing of information that relates to supply chain management activities and also deeper product life-cycle knowledge to find opportunities for pollution control and prevention in manufacturing processes (Seuring, 2004).

Environmental issue networks aim to diffuse knowledge on the state of specific environmental issues and also to related possible pollution prevention and control (Ritvala and Salmi, 2010). This creates a shared situational awareness on challenges and opportunities in the environment and available resources with which they might be addressed (Fjeldstad et al., 2012). Issue networks can also provide other intangible assets to the network members such as shared legitimacy that facilitates further opportunities for resource mobilisation by, for example, access to political actors (Ritvala and Salmi, 2010).

Environmental solution networks are also characterised by knowledge sharing as a common resource to achieve integrated solutions. Existing research on innovation networks suggests that they can potentially include further commons such as shared intellectual property rights and technologies to strengthen relationships between network actors (Baraldi et al., 2011).

5.2.3. Protocols, processes and infrastructures

There are three different methods by which an IS network evolves: 1) self-organising, 2) facilitation by organisations or

individuals and 3) planning (see e.g. Chertow and Ehrenfeld, 2012; Paquin and Howard-Grenville, 2012). Self-organised networks develop by themselves without a coordinator, whereas an executive coordinator develops planned networks. Thus, in planned networks a coordinator (e.g. a governmental organisation) influences a network's structure and norms. Between self-organised and planned networks lie facilitated networks, in which a coordinator's main task is to find potential partners among participating companies (Paquin and Howard-Grenville, 2012). For example, the National Industrial Symbiosis Program (NISP) in the UK facilitates the formation of IS relationships by sharing information and analysing potential exchanges between participating companies (Doménech and Davies, 2011). To operate, regardless of how a network has evolved, an IS network requires partnerships based on trust, shared norms, long mutual dependence and personal relationships for knowledge transfer (Baas, 2011). Tangible resource exchanges between actors are usually governed by formal market mechanisms.

Coordination in a sustainable supply network is usually accomplished through formal information exchanges, such as orders, forecasts and inventory information and also market-based transactions for the flow of goods between actors (Bansal and McKnight, 2009). Supply networks are also characterised by highly standardised operating practices between actors to achieve efficiency (Bansal and McKnight, 2009). Standards can also be employed to coordinate green actions; for example, by imposing environmental criteria on suppliers (Seuring, 2004). However, studies have shown that "greening" actions aimed at decreasing environmental impacts within the supply network also require cooperation and integration among actors in the network (Seuring, 2004).

An Environmental issue network can display considerable variance in protocols, processes and infrastructure to achieve collaboration. Collaborative action in voluntary issue networks can be achieved through processes such as network mobilisation and framing and communication of environmental challenges (Ritvala and Salmi, 2010), whereas more coercive forms of issue network require negotiations and wise trades to achieve consensus, which can be formalised through binding agreements such as, in the case of climate change, the Kyoto protocol (Veal and Mouzas, 2010).

An environmental solution network requires coordination processes by hub firms to mobilise the required network resources for the development of a solution and also to manage the network's evolution. Coordination processes in the network also include co-development, knowledge sharing and joint action to achieve the embedding and integration of technology (Baraldi et al., 2011).

5.2.4. Hierarchy of decision making

As previously noted, IS networks evolve in different ways and these evolutionary paths differ in the extent of decision-making stratification within a network. Planned IS networks are the most hierarchical form; network coordinators can even design the network by top-down planning and recruiting suitable companies to join the network (Gibbs and Deutz, 2005). This is especially typical in eco-industrial parks. Facilitated networks have some formal coordination mechanisms for managing the network's evolution, diffusing knowledge and finding new exchange opportunities; however, they are also partly driven by serendipitous relations that arise between network actors. Self-organised IS networks have the most heterarchical decision-making processes that emphasise community, shared values and embedded relations between actors, which enables the actors to find exchange opportunities (Chertow and Ehrenfeld, 2012).

Sustainable supply networks are often dominated by large and powerful global firms, leading to more hierarchical control and coordination mechanisms (Fjeldstad et al., 2012; Gulati et al., 2012).

Relationships between actors are often formal in nature and actors might employ competitive tactics to achieve more power in the network. They often lack the informality of social relations that have been found to be important for the identification of new exchange opportunities in self-organising IS networks (Bansal and McKnight, 2009).

Environmental issue networks also have varied forms of decisionmaking, depending on the scale and importance of the issue. At one end of the scale, they can be characterised by heterarchical self-organisation and voluntary cooperation among actors with shared values (Ritvala and Salmi, 2010), while, at the other end, highly important global issues such as climate change also require hierarchical decision-making, such as enforced restrictions on CO2 emission limits (Veal and Mouzas, 2010).

Environmental solution networks are characterised by hub firms which act as integrators for the network. While network members are autonomous actors, hub firms have a largely hierarchical role in managing the integration. Hub firms need to coordinate various involved partners and suppliers to manage possible conflicts of interest, controlled expansion of the network and various contexts of embedding technology: development activities such as R&D, production operations and usage activities on the customer's side (Baraldi et al., 2011). However, some innovation networks, such as open source communities in the IT sector, can also be highly heterarchical in nature (Gulati et al., 2012). This suggests that, depending on the extent of control exercised by hub firms, environmental solution networks can also be heterarchically formed.

5.2.5. Permeability of boundaries

The degree of openness in a network was the final examined dimension of the identified eco-industrial network forms' architecture. It is difficult to classify them as distinctly open networks (e.g. open source communities) or those that are closed (e.g. strategic alliances). However, all network forms have some constraining factor that limits expansion. IS networks are distinguished by a regional focus, as best demonstrated by eco-industrial parks. Long transportation distances for waste and by-products are not desirable due to added transportation costs and environmental impacts. The proximity of involved actors also plays a major role in enabling the development of embedded relations that foster a network's development. Conversely, sustainable supply networks are rarely confined by geography; rather, global relationships between suppliers and customers are common and driven by cost concerns (Bansal and McKnight, 2009). Thus, geographic proximity between actors plays a smaller role than in IS networks. However, unlike IS networks in which cross-industry relationships are more common, supply networks are commonly constrained to actors along a single supply chain (Liu et al., 2012).

Environmental issue networks are the most open form of the identified networks, especially in the case of voluntary participation. However, depending on the scale of the environmental issue, they can still be limited by geography. Regionally limited issues typically mobilise actors which are closely affected by that particular issue (e.g. contamination of a ground or water area), while global challenges call for global cooperation. It is also noteworthy that some environmental issue networks are temporary and might disband as the issue loses significance. However, many environmental issues are difficult to solve and, thus, require long-term collaboration (Ritvala and Salmi, 2010). The global climate change network is a good example of one that is very likely to have a long lifespan (Veal and Mouzas, 2010).

Environmental solution networks are typically constrained by the same membership criteria as many innovation networks (Gulati et al., 2012). Members are chosen depending on the technology, resources and capabilities that they possess, so that they

complement the central technological solution (Baraldi et al., 2011). Some degree of redundancy in network resources might be desirable to provide flexibility in a changing environment, but it might also restrict cooperation by creating a competitive situation between actors (Gulati et al., 2012). Technological fit is ultimately the most important membership criteria in solution networks (Baraldi et al., 2011).

6. Discussion

According to our findings, we believe that this study advances the interaction of industrial symbiosis (IS), sustainable supply, environmental issue and solution networks and also the identification and analysis of alternative forms of eco-industrial networks. In addition, our results contribute to calls for shifting the focus of sustainability research from individual actors to a more systemic and network-based perspective (Manring, 2007; Wittneben et al., 2012). The study's findings yield the following implications for practice, policymakers and future research on networks and sustainability.

6.1. Implications for policy-making and practice

Our analysis of the operational logics of eco-industrial networks offers three key implications for policy-making and practice. First, policymakers and practitioners need to become aware of the different mechanisms through which industrial networking can reduce environmental load. By acknowledging the variety of different operational logics and mechanisms, policymakers can form a wider portfolio of platforms that serve the alternative networking forms and enable firms to identify potential collaborators. To facilitate the development of various types of eco-industrial network, policymakers need to build systems that take into account the differences and similarities in alternative network forms that enable actors to connect with each other and access relevant information in a meaningful and purposeful manner. For example, the advancement of symbiosis networks requires comprehensive regional databases on the inputs and outputs of production plants, whereas the advancement of environmental solution networks requires information on ongoing and planned technology development projects and platforms to pair firms with synergistic R&D opportunities. While sustainable supply networks can be promoted by raising consumer awareness, the emergence of issue networks can be facilitated by arranging networking events and opportunities for firms that share an environmental concern to take collective action.

To date, legislation is often a trigger for eco-industrial networking. Good examples of this are Kalundborg in Denmark, the National Industrial Symbiosis Program (NISP) in the UK and the municipality of Chamusca in Portugal (Costa and Ferrao, 2010; Doménech and Davies, 2011), which were all encouraged to form IS networks by regional regulations. Furthermore, government support and incentives are essential for advancing environmental sustainability, as was the case with, for example, the Ulsan eco-industrial park in South Korea and the NISP in the UK (Behera et al., 2012; Doménech and Davies, 2011). However, regulation can be an obstacle to turning waste into raw materials, as with the IS case in the Gulf of Bothnia in Finland and Sweden (Salmi et al., 2011) and in Sarnia in Canada (Sakr et al., 2011). Although these are examples of IS, similar situations concern, at least, sustainable supply chains and environmental solution networks. Similarly, policies promoting transparency in the use of resources by industrial firms can facilitate especially sustainable supply networks, but can also promote other eco-industrial network forms. However, there is a need for all mentioned forms of governmental support because significant opportunities for reducing environmental load

might be missed by designing programs that facilitate only one particular form of eco-industrial networking.

Second, environmental issue networks highlight the role of issue-based mobilisation of actors for solving environmental problems. While these networks do not always have a concrete agenda for providing business benefits to participants, this form of organising can be utilised for the formation of other types of eco-industrial networks. For example, recent research on IS has highlighted the role of social embeddedness for the diffusion of environmental knowledge and values among actors (Ashton and Bain, 2012; Chertow and Ehrenfeld, 2012). We suggest that issue-based mobilisation can be a key activity in promoting social relations and initiating collaboration between actors that can eventually lead to eco-industrial networks. The extant literature on the concept of imprinting in organisational research has shown that organisations adopt elements of their founding environment that can persist well beyond the founding stage (Marquis and Tilcsik, 2013). Networks that emerge through issue-based mobilisation can thus be imprinted with the environmental values that served as the catalyst for their foundation (Ritvala and Salmi, 2010).

Third, industrial firms should consider a broader variety of exchange options available for decreasing environmental impacts through inter-firm relations. One example of this is the option between GSCM and IS; decreasing the output of waste or byproduct through GSCM activities beyond a particular limit might prevent it from being re-utilised in IS. As such, the supply chain's limited perspective can be a hindrance. Environmental solution networks provide another set of options for collaboration: those focused on horizontal relations between actors. New technological solutions generated by horizontal collaboration can enable the development of other eco-industrial networks; for example, by generating novel technologies to reprocess by-products in IS networks or pollution prevention solutions for GSCM activities. For example, IS might be a feasible solution for an industrial firm only after it acquires sufficient technological resources for reprocessing waste.

The analysis of the identified eco-industrial network forms' architectures can also offer some interesting insights for policy-making and practice. First, collaboration among a wide variety of organisations is needed for eco-industrial networks. Collaboration can span vertical relations in supply chains and also horizontal collaboration with competitors or actors possessing complementary offerings. IS typically requires cross-industrial relations to find novel usages for waste and by-products. In addition, eco-industrial networks should also span different sectors, bringing together private sector firms with governmental organisations and NGOs, which can bring new knowledge on opportunities for environmental protection. We suggest that strategic coalitions should be formed around key environmental issues, which maximise knowledge on opportunities by bringing together a wide variety of organisations from different sectors that are stakeholders in the issue. A polycentric governance approach for environmental issues suggested by Ostrom (2009), which maximises local-scale opportunities and incentives for cooperation, could facilitate these types of coalition.

The dilemma between central coordination and self-organisation is another key issue that has received much attention, especially with regard to IS. Much of the literature makes a sharp distinction between these two evolutionary forms of IS (e.g. Baas, 2011). Our review suggests that this issue is more complex. The facilitated model of IS in the UK has proved to be highly successful for developing relations. Green supply networks and environmental solution networks also favour some degree of hierarchy in the coordination of the networks' activities. Often, this might be because smaller firms in the networks lack the resources or

knowledge to be able to undertake purposeful initiatives for developing environmental activities. This suggests that an eco-industrial network needs one or several key organisations actively furthering the collective goal of the network. Simultaneously, the network architecture should also facilitate opportunities for serendipitous relations, as an overly hierarchical network can also hinder progress (Paquin and Howard-Grenville, 2012).

With this in mind, we suggest that a modified form of the actor-oriented network architecture, which has proven successful in areas such as innovation, would be highly suitable for the development of eco-industrial networks. The key elements of this architecture, common norms and protocols for network activities and also common resources and infrastructure that enable members to serendipitously self-organise, are well suited for promoting environmental awareness in industrial firms and facilitating the flexible formation of environmentally friendly business relations. However, a coordinating organisation can provide a tremendous boost to a network's activities. It can take a broader perspective on developing elements of the network's architecture and also gain new resources for the network through issue-based mobilisation. For example, a coordinating organisation for IS can communicate the network's activities to the wider society, actively facilitate the formation of new network relations and also influence political decision-making to develop a legal environment that supports IS. However, as our review suggests, eco-industrial networks can take different forms through the operational logic by which they aim to provide environmental benefits and, thus, the most suitable form of network architecture is likely to be context dependent.

6.2. Implications for further research

Our research uncovered several fruitful avenues for academic research. First, given the potential transformative power of networks in advancing environmental sustainability, it is surprising that studies on networks and sustainability are relatively scarce and dispersed into separate streams of the literature. For example, IS has had only a little influence in the supply chain literature; however, there is a considerable opportunity to combine these perspectives (Bansal and McKnight, 2009). Moreover, insights from the extensive research on alliances, networks and inter-organisational collaboration (Ahuja et al., 2012; Brass et al., 2004; Fjeldstad et al., 2012; Gulati et al., 2000) are not integrated with studies on eco-industrial networks. Based on our findings, we propose that researchers need to acknowledge the multifaceted potential of industrial networks in advancing environmental sustainability and focus their attention on the design, functioning and processes of multi-actor collaborative networks. One way to accomplish this might be to explore hybrid eco-industrial networks that simultaneously enact multiple logics to reduce the environmental load. Concurrent materials reuse, collective action, value-chain optimisation and co-innovation among network partners might result in highly influential networks that play a substantial role in advancing environmental sustainability. Research methodologies from network research could offer new insights for the study of eco-industrial networks. Social network analysis can uncover the underlying structural elements and dynamics of business networks (Ahuja et al., 2012) and is already gaining interest among eco-industrial networks, especially with regard to IS research (Ashton, 2008). Qualitative methodologies focusing on network processes and focal events can also help to gain a rich understanding on the dynamics of eco-industrial networks (Halinen et al., 2013).

Second, further research on the role of specific network forms in addressing particular environmental issues is also needed. For example, IS networks and environmental solution networks are

critical for the development of renewable energy systems, an important area for further research on eco-industrial networks.

Third, more research is needed on the interplay between heter-archical processes of self-organisation and the more hierarchical processes of central coordination in the development of eco-industrial networks. For example, decision-making in IS networks is described as inherently mostly heterarchical, with self-organisation among actors. However, some examples (Paquin and Howard-Grenville, 2012, 2013) suggest that coordination of an IS network can be very effective when orchestrated by a third party organisation. As mentioned in the previous section, our research found a variety of network architectures that can support the development of eco-industrial networks. Future research could study how contextual factors, such as the operational logic of the network or the institutional environment in which the network operates, affect the success of different architectural arrangements.

Finally, our research focused on inter-organisational networks aiming at beneficial environmental action and not on the potential negative consequences that networking can have in this context. For example, networks can also facilitate the diffusion of environmentally hazardous practices among industrial firms. Industry associations also commonly attempt to influence political decision-making to create a more favourable operating environment for firms in their sector, which can include lobbying against tighter environmental laws. Further research is needed to determine how these potential negative network effects impact eco-industrial networks.

7. Conclusions

Network-based collaboration is critical to the solution of complex problems such as the environmental load of production and consumption. A substantial body of the literature documents how collaborative networks operate and yield several benefits for involved actors. However, research on networks and sustainability is at a relatively early stage and dispersed among different streams of the literature. By analysing various forms of eco-industrial networks, their operational logics and architecture, this study's objective is to broaden the relatively limited understanding on ways by which networks can advance environmental sustainability.

Prior research perceives the potential of eco-industrial networks primarily from the perspective of industrial symbiosis. However, a multidisciplinary literature review suggests that the potential role of networks in advancing environmental sustainability is much broader. Based on a systematic literature review, we identified four

Appendix

forms of eco-industrial networks: 1) symbiosis networks, 2) sustainable supply networks, 3) environmental issue networks and 4) environmental solution networks. There are major differences in the logics through which these network forms advance environmental sustainability. Analysis of these network forms suggests that collaborative eco-industrial networks have the ability to rationalise and improve resource utilisation through industrial symbiosis and sustainable supply networks, accelerate the diffusion of more sustainable practices and foster environmental awareness and collective action through environmental issue networks, and also generate novel technologies and solutions for reducing environmental impacts through environmental solution networks. Firms can achieve cost reductions resulting from improved resource efficiency and increased revenues through selling reusable waste and by-products, and also through eco-innovative solutions. In addition, all identified network forms have the potential to improve participating firms' corporate images, which is important for the management of key stakeholders' expectations and to achieve legitimacy. However, orchestrating the rather complex network architecture and the potential conflicts of interests and goals between network participants appears to be a key challenge for all forms of eco-industrial networks.

It seems that there is a need to build broad coalitions of organisations that are mobilised to address environmental issues. Thus, network architecture that maximises the members' capability to self-organise while also including a coordinating organisation can be highly suitable for eco-industrial networks. By providing some fruitful avenues for future research on eco-industrial networks, we hope that this study advances the integration of IS, sustainable supply, environmental issues and solution networks, and also the identification and analysis of alternative forms of eco-industrial network. We suggest that this broader perspective will benefit future research by providing more understanding on the multifaceted role of eco-industrial networks in advancing sustainability.

Acknowledgements

We are grateful to the Finnish Funding Agency for Technology and Innovation (TEKES) for its financial support of the study, which was conducted as part of TEKES' Green Growth programme. The authors also acknowledge cooperation and support from the Green Growth programme and project management. We also want to thank two anonymous reviewers for providing helpful feedback for improving the paper.

Author

Year Name

Journal

Major concepts

Andersson and Sweet 2002

Ashton 2008

Ashton 2009

Ashton 2011

Ashton and Bain 2012

Baas 2011

Bansal and McKnight 2009

Towards a framework for ecological strategic change in business networks. Understanding the organization of industrial ecosystems. A social network approach.

The Structure, Function, and Evolution of a Regional Industrial Ecosystem. Managing performance expectations of industrial symbiosis. Assessing the "Short Mental Distance" in eco-industrial networks. Planning and uncovering industrial symbiosis: comparing the Rotterdam and Ostergótland regions.

Looking forward, pushing back and peering sideways: analysis the sustainability of industrial symbiosis.

Journal of Cleaner Production Journal of Industrial Ecology

Journal of Industrial Ecology Business Strategy and the Environment Journal of Industrial Ecology Business Strategy and the Environment

Journal of Supply Chain Management

Network, strategic change.

Industrial symbiosis, social network theory/analysis, industrial ecosystem.

Industrial ecosystem.

Industrial symbiosis.

Social embeddedness, social capital, industrial symbiosis. Industrial symbiosis, social embeddedness.

Industrial symbiosis, supply chains, sustainability.

(continued on next page)

(continued )

Author

Year Name

Journal

Major concepts

Baraldi et al. 2011

Behera et al. 2012

Boons and Spekkink 2012

Boons et al. 2011

Chertow 2000

Chertow 2007

Chertow and Ehrenfeld 2012

Chertow and Miyata 2011

Costa and Ferrao 2010 Côté and Cohen-Rosenthal 1998

Desrochers 2004

Dimitrova et al. 2007

Doménech and Davies 2011

Gibbs 2003

Gibbs and Deutz 2005

Gibbs and Deutz 2007

Kocabagsoglu et al. 2007

Liu et al. 2012

Lombardi and Laybourn 2012

Paquin and 2012

Howard-Grenville

Ritvala and Salmi 2010

Ritvala and Salmi 2011

Sakr et al. 2011

Salmi and Toppinen 2007

Salmi et al. 2012

Seuring 2004

Sterr and Ott 2004

Network evolution and the embedding of complex technical solutions: The case of the Leaf House network. Evolution of "designed" industrial symbiosis networks in the Ulsan eco-industrial park: "research and development into business" as the enabling framework. Levels of institutional capacity and actor expectations about industrial symbiosis. Evidence from the Dutch stimulation program 1999-2004. The dynamics of industrial symbiosis: a proposal for a conceptual framework based upon a comprehensive literature review. Industrial symbiosis: Literature and Taxonomy.

"Uncovering" industrial symbiosis.

Organizing self-organizing systems: towards a theory of industrial symbiosis. Assessing collective firm behavior: comparing industrial symbiosis with possible alternatives for individual companies in Oahu, HI. A case study of industrial symbiosis development using a middle-out approach. Designing eco-industrial parks: A synthesis of some experiences. Industrial symbiosis: the case for market coordination.

Managerial factors for evaluating eco-clustering approach. The role of embeddedness in industrial symbiosis networks: phases in the evolution of industrial symbiosis networks. Trust and networking in inter-firm relations: the case of eco-industrial development.

Implementing industrial ecology? Planning for eco-industrial parks in the USA. Reflections on implementing industrial ecology through eco-industrial park development.

Linking forward and reverse supply chain investments-the role of business uncertainty.

Going beyond the sectoral boundary: a key

stage in the development of a regional

industrial ecosystem.

Redefining industrial symbiosis: crossing

academic-practitioner boundaries.

The evolution of facilitated industrial

symbiosis.

Value-based network mobilization: A case study of modern environmental networkers.

Network mobilizers and target firms: the case of saving the Baltic Sea. Critical success and limiting factors for eco-industrial parks: global trends and Egyptian context.

Embedding science in politics - "Complex utilization" and industrial ecology as models of natural resource use. Governing the interplay between industrial ecosystems and environmental regulation: heavy industries in the Gulf of Bothnia in Finland and Sweden.

Integrated chain management and supply chain management-comparative analysis and illustrative cases.

The industrial region as a promising unit for eco-industrial development - reflections, practical experience and establishment of innovative instruments to support industrial ecology.

Industrial Marketing Management Journal of Cleaner Production

Journal of Industrial Ecology

Journal of Cleaner Production

Annual Review of Energy and the Environment

Journal of Industrial Ecology Journal of Industrial Ecology Business Strategy and the Environment

Journal of Cleaner Production

Journal of Cleaner Production

Journal of Cleaner Production

Industrial Management and Data Systems

Business Strategy and the Environment

Local Economy

Geoforum

Journal of Cleaner Production

Journal of Operations Management

Journal of Cleaner Production

Journal of Industrial Ecology Journal of Industrial Ecology Industrial Marketing Management

Industrial Marketing Management Journal of Cleaner Production

Journal of Industrial Ecology

Journal of Industrial Ecology

Journal of Cleaner Production Journal of Cleaner Production

Industrial networks, technology embedding, sustainability.

Industrial symbiosis, eco-industrial park.

Industrial symbiosis, institutional capacity.

Industrial symbiosis, institutional capacity, institutional theory.

Industrial symbiosis, eco-industrial park.

Industrial symbiosis, eco-industrial park.

Industrial symbiosis, self-organising systems, complexity. Industrial symbiosis.

Industrial symbiosis.

Eco-industrial park.

Industrial symbiosis, Austrian economics, central planning. Eco-clustering.

Industrial symbiosis, social embeddedness.

Eco-industrial park.

Industrial ecology, eco-industrial parks.

Industrial ecology, eco-industrial parks; economic development policy.

Reverse supply chains.

Industrial ecosystem.

Industrial symbiosis.

Industrial symbiosis, serendipitous and goal-directed processes. Value-based networks, mobilisation, common issues.

Mobilisation, issue networks.

Industrial symbiosis, eco-industrial park.

Industrial ecology, complex utilisation, political embeddedness.

Industrial symbiosis, environmental regulation, common pooled resources.

Integrated supply chain management.

Eco-industrial networks, regional industrial ecosystems, material flow management.

( continued )

Year Name

Author

Taddeo et al. 2012

Tudor et al. 2007

Vachon and Klasssen 2007

Van Ha et al. 2009

Veal and Mouzas 2010

Von Malmborg 2004

Zhu and Cote 2004

Implementing eco-industrial parks in existing clusters. Findings from a historical Italian chemical site.

Drivers and limitations for the successful development and functioning of EIPs (eco-industrial parks): A literature review. Supply chain management and environmental technologies: the role of integration.

Techno policy aspects and socio-economic impacts of eco-industrial networking in the fishery sector: experiences from An Ging Province, Vietnam.

Learning to collaborate: a study of business networks.

Networking for knowledge transfer: towards an understanding of local authority roles in regional industrial ecosystem management.

Integrating green supply chain management into an embryonic eco-industrial development: a case study of the Guitang Group.

Journal

Journal of Cleaner Production

Ecological Economics

International Journal of Production Research

Journal of Cleaner Production

Journal of Business & Industrial Marketing

Business Strategy and the Environment Journal of Cleaner Production

Major concepts Eco-industrial park.

Eco-industrial park.

Integrated supply chain management.

Eco-industrial network

Networks, collaboration.

Industrial ecology, local authority, public-private partnerships, knowledge bank, knowledge broker.

Integrated supply chain management, green supply chain management, eco-industrial network.

References

Ahuja, G., 2000. Collaboration networks, structural holes, and innovation: a longitudinal study. Adm. Sci. Q. 45 (3), 425-455.

Ahuja, G., Soda, G., Zaheer, A., 2012. The genesis and dynamics of organizational networks. Organ. Sci. 23 (2), 434-448.

Andersson, P., Sweet, S., 2002. Towards a framework for ecological strategic change in business networks. J. Clean. Prod. 10, 465-478.

Ashton, W., 2008. Understanding the organization of industrial ecosystems. A social network approach. J. Ind. Ecol. 12 (1), 34-51.

Ashton, W.S., 2011. Managing performance expectations of industrial symbiosis. Bus. Strat. Env. 20 (5), 297-309.

Ashton, W.S., Bain, A.C., 2012. Assessing the "short mental distance" in eco-industrial networks. J. Ind. Ecol. 16 (1), 70-82.

Baas, L., 2011. Planning and uncovering industrial symbiosis: comparing the Rotterdam and Ostergotland regions. Bus. Strat. Env. 20 (7), 428-440.

Baas, L.W., Boons, F.A., 2004. An industrial ecology project in practice: exploring the boundaries of decision-making levels in regional industrial systems. J. Clean. Prod. 12 (8-10), 1073-1085.

Bansal, P., McKnight, B., 2009. Looking forward, pushing back and peering sideways: analysis the sustainability of industrial symbiosis. J. Supply Chain Manag. 45 (4), 26-37.

Baraldi, E., Gregori, G.L., Perna, A., 2011. Network evolution and the embedding of complex technical solutions: the case of the Leaf House network. Ind. Mark. Manag. 40 (6), 838-852.

Behera, S.K., Kim, J.-H., Lee, S.-Y., Suh, S., Park, H.-S., 2012. Evolution of "designed" industrial symbiosis networks in the Ulsan eco-industrial park: "research and development into business" as the enabling framework. J. Clean. Prod. 29-30, 103-112.

Brass, D.J., Galaskiewicz, J., Greve, H.R., Tsai, W., 2004. Taking stock of networks and organizations: a multilevel perspective. Acad. Manag. J. 47 (6), 795-817.

Chertow, M.R., 2000. Industrial symbiosis: literature and taxonomy. Annu. Rev. Energy Environ. 25, 313-337.

Chertow, M.R., 2007. "Uncovering" industrial symbiosis. J. Ind. Ecol. 11 (1), 11-30.

Chertow, M., Ehrenfeld, J., 2012. Organizing self-organizing systems: towards a theory of industrial symbiosis. J. Ind. Ecol. 16 (1), 13-27.

Chesbrough, H.W., 2003. Open Innovation: The New Imperative for Creating and Profiting from Technology. Harvard Business School Press, Boston.

Costa, I., Ferrao, P., 2010. A case study of industrial symbiosis development using a middle-out approach. J. Clean. Prod. 18, 984-992.

Denyer, D., Tranfield, D., 2006. Using qualitative research synthesis to build an actionable knowledge base. Manag. Decis. 44 (2), 213-227.

Dimitrova, V., Lagioia, G., Gallucci, T., 2007. Managerial factors for evaluating eco-clustering approach. Ind. Manag. Data Syst. 107 (9), 1335-1348.

Doménech, T., Davies, M., 2011. The role of embeddedness in industrial symbiosis networks: phases in the evolution of industrial symbiosis networks. Bus. Strat. Env. 20 (5), 281-296.

Fjeldstad, 0.,D., Snow, C.C., Miles, R.E., Lettl, C., 2012. The architecture of collaboration. Strateg. Manag. J. 33, 734-750.

Gibbs, D., Deutz, P., 2005. Implementing industrial ecology? Planning for eco-industrial parks in the USA. Geoforum 36, 452-464.

Gibbs, D., Deutz, P., 2007. Reflections on implementing industrial ecology through eco-industrial park development. J. Clean. Prod. 15 (17), 1683-1695.

Gulati, R., Nohria, N., Zaheer, A., 2000. Strategic networks. Strateg. Manag. J. 21, 203-213.

Gulati, R., Puranam, P., Tushman, M., 2012. Meta-organization design: rethinking design in interorganizational and community contexts. Strateg. Manag. J. 33, 571-586.

Halinen, A., Toornroos, J., Elo, M., 2013. Network process analysis: an event-based approach to study business network dynamics. Ind. Mark. Manag. 42 (8), 1213-1222.

Kim, T.Y., Oh, H., Swaminathan, A., 2006. Framing interorganizational network change: a network inertia perspective. Acad. Manag. Rev. 31 (3), 704-720.

Kocabasoglu, C., Prahinski, C., Klassen, R.D., 2007. Linking forward and reverse supply chain investments: the role of business uncertainty. J. Oper. Manag. 25 (6), 1141-1160.

Liu, C., Ma, C., Zhang, K., 2012. Going beyond the sectoral boundary: a key stage in the development of a regional industrial ecosystem. J. Clean. Prod. 22 (1), 42-49.

Lombardi, D.R., Laybourn, P., 2012. Redefining industrial symbiosis. Crossing academic-practitioner boundaries. J. Ind. Ecol. 16 (1), 28-37.

Manring, S.L., 2007. Creating and managing interorganizational learning networks to achieve sustainable ecosystem management. Organ. Environ. 20 (3), 325-346.

Marquis, C., Tilcsik, A., 2013. Imprinting: toward a multilevel theory. Acad. Manag. Ann. 7 (1), 192-243.

Mirata, M., Emtairah, T., 2005. Industrial symbiosis networks and the contribution to environmental innovation: the case of the Lanskrona industrial symbiosis programme. J. Clean. Prod. 13 (10-11), 993-1002.

Morelli, N., 2006. Developing new product service systems (PSS): methodologies and operational tools. J. Clean. Prod. 14 (17), 1495-1501.

Ostrom, E., 2009. A Polycentric Approach for Coping with Climate Change. World Bank, Washington, DC.

Paquin, R.L., Howard-Grenville, J., 2012. The evolution of facilitated industrial symbiosis. J. Ind. Ecol. 16 (1), 83-93.

Paquin, R.L., Howard-Grenville, J., 2013. Blind marriages blind dates and arranged marriages: longitudinal processes of network orchestration. Org. Stud. 34 (11), 1623-1653.

Ring, P.S., Van De Ven, A.H., 1992. Structuring cooperative relationships between organizations. Strateg. Manag. J. 13 (7), 483-498.

Ritvala, T., Salmi, A., 2010. Value-based network mobilization: a case study of modern environmental networkers. Ind. Mark. Manag. 39 (6), 898-907.

Ritvala, T., Salmi, A., 2011. Network mobilizers and target firms: the case of saving the Baltic Sea. Ind. Mark. Manag. 40, 887-898.

Sakr, D., Baas, L., El-Haggar, S., Huising, D., 2011. Critical success and limiting factors for eco-industrial parks: global trends and Egyptian context. J. Clean. Prod. 19 (11), 1158-1169.

Salmi, O., Hukkinen, J., Heino, J., Pajunen, N., Wierink, M., 2011. Governing the interplay between industrial ecosystems and environmental regulation. Heavy industries in the Gulf of Bothnia in Finland and Sweden. J. Ind. Ecol. 16 (1), 119-128.

Seuring, S., 2004. Integrated chain management and supply chain management comparative analysis and illustrative cases. J. Clean. Prod. 12 (8-10), 1059-1071.

Taddeo, R., Simboli, A., Morgante, A., 2012. Implementing eco-industrial parks in existing clusters. Findings from a historical Italian chemical site. J. Clean. Prod. 33, 22-29.

Tudor, T., Adam, E., Bates, M., 2007. Drivers and limitations for the successful development and functioning of EIPs (eco-industrial parks): a literature review. Ecol. Econ. 61 (2-3), 199-207.

Vachon, S., Klassen, R.D., 2007. Supply chain management and environmental technologies: the role of integration. Int. J. Prod. Res. 45 (2), 401-423.

Van Ha, N.T., Ananth, A.P., Visvanathan, C., Anbumozhi, V., 2009. Techno policy aspects and socio-economic impacts of eco-industrial networking in the fishery sector: experiences from An Giang Province, Vietnam. J. Clean. Prod. 17 (14), 1272-1280.

Veal, G., Mouzas, S., 2010. Learning to collaborate: a study of business networks. J. Bus. Ind. Mark. 25 (6), 420-434.

von Malmborg, F., 2004. Networking for knowledge transfer: towards an understanding of local authority roles in regional industrial ecosystem management. Bus. Strat. Env. 13 (5), 334-346. Wittneben, B.F., Okereke, C., Banerjee, S.B., Levy, D.L., 2012. Climate change and the emergence of new organizational landscapes. Organ. Stud. 33 (11), 1431-1450. Yuan, Z., Bi, J., Moriguichi, Y., 2006. The circular economy. A new development

strategy in China. J. Ind. Ecol. 10 (1-2), 4-8. Zhu, Q., Cote, R.P., 2004. Integrating green supply chain management into an embryonic eco-industrial development: a case study of the Guitang Group. J. Clean. Prod. 12 (8-10), 1025-1035.