Journal of Environmental Chemical Engineering 11 (2023) 111023 Contents lists available at ScienceDirect Journal of Environmental Chemical Engineering journal homepage: www.elsevier.com/locate/jece Measuring the symbiotic performance of single entities within networks using an LCA approach Fernando Arce Bastias a, Paula Daniela Rodríguez a, Alejandro Pablo Arena a, Laura Talens Peiró b, * a Grupo CLIOPE, Facultad Regional Mendoza, Universidad Tecnológica Nacional, Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), Mendoza 5500, Argentina b SosteniPrA Research Group (2021SGR00073), Institut de Ciència i Tecnologia Ambientals (ICTA-UAB), Unitat Excelencia Maria de Maeztu (MdM 2015–0552; CEX2019–000940-M), Universitat Autònoma de Barcelona, Cerdanyola del Vallès, Barcelona 08193, Spain A R T I C L E I N F O A B S T R A C T Editor: Teik Thye Lim This paper introduces the Symbiotic Performance Indicator, a novel indicator aiming to quantify the environ­ mental benefits generated by by-product exchanges in an industrial symbiosis network. Despite the significant Keywords: advancements in assessing industrial symbiosis, the lack of indicators for individual entities involved in by- Environmental performance product exchanges hinders a comprehensive understanding of its environmental benefits. This indicator ac­ By-product exchange counts for resource use and greenhouse gas emissions using a life cycle perspective. The resource use is measured Entity level using the Cumulative Exergy Extraction from the Natural Environment method, while greenhouse gas emissions Life cycle assessment Indicator are evaluated using the IPCC 2013 Global Warming Potential (100a) method. The use of this indicator is illus­ trated in a real case study where plastic waste is exchanged among three entities in Mendoza (Argentina). The overall results show that resource use and greenhouse gas emissions can be reduced by 19% and 15%, respec­ tively. Full and partial allocation methods are proposed within the formulation of the Symbiotic Performance Indicator. The indicator results show that the exchange of materials may seem less attractive when using full allocation methods, as one entity gets 100% of the benefits from the by-product exchange compared to the other. Partial allocations make the by-product exchange convenient for both entities which may encourage collabo­ ration. In conclusion, the proposed indicator helps account more precisely for individual environmental benefits behind by-product exchanges, and thus enables better decision-making to set up an industrial symbiosis network. 1. Introduction sustainable development by deploying a regenerative system [3]. Industrial Symbiosis (IS) presents a systems approach to promoting a In contrast to the current linear economic activity (“take-make- CE through the collaboration of various entities to exchange their by- dispose”), the Circular Economy (CE) proposes a broader and more products and reduce the use of primary resources overall [4]. The comprehensive approach to understanding systems, products and ser­ generally accepted definition of IS states that wastes or by-products of vices and their interaction with the environment, over the entire life one industry become inputs to other industry. Based on such coopera­ cycle of any process [1]. Although the basic principles for CE have been tion, these industries create an IS network that additionally generates developed for nearly 30 years, there is no commonly accepted definition economic, environmental and social benefits [5,6]. The most referenced [2]. However, the main objective of CE is to analyze systems using a and clear-cut case is the Kalundborg Symbiosis Park located in Denmark. systemic approach with the intention to optimize the use of resources In this park, water, energy and materials are exchanged in closed loops and reduce waste from a life cycle perspective. This is done to achieve between private and public entities. The Kalundborg Symbiosis began Abbreviation: CE, Circular Economy; CEENE, Cumulative Exergy Extraction from the Natural Environment; GHG, Greenhouse Gas; IS, Industrial Symbiosis; IPCC, Intergovernmental Panel on Climate Change; LCA, Life Cycle Assessment; MFA, Material Flow Analysis; MPM, Madera Plástica Mendoza; PE, polyethylene; PP, polypropylene; RBR, Recyclability Benefit Rate; RBROL, RBR for open-loop recycling; S-LCA, Social Life Cycle Assessment; SPI, Symbiotic Performance Indicator; UP, Urquiza Plásticos. * Corresponding author. E-mail address: laura.talens@uab.cat (L. Talens Peiró). https://doi.org/10.1016/j.jece.2023.111023 Received 29 May 2023; Received in revised form 16 August 2023; Accepted 12 September 2023 Available online 15 September 2023 2213-3437/© 2023 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by- nc-nd/4.0/). F. Arce Bastias et al. Journal of Environmental Chemical Engineering 11 (2023) 111023 with a couple of collaborative initiatives between a few firms 40 years 2. Literature review ago. The exchanges in the network were not originally intended to demonstrate the benefits of industrial symbiosis but were negotiated and 2.1. Assessment of the environmental benefits of IS practices established to pursue an economic benefit [7,8]. Since new firms have joined this network, the exchanges of waste flows and resources have Wadström et al. [23] examined the methods used to study IS and expanded gradually to result in an annual savings of 635,000 tonnes of found that interviews followed by LCA are the most frequently used CO2 and 87,000 tonnes of primary materials, including ethanol, fly ash, method to collect data for quantitative studies. Similarly, Harris et al. gypsum, sulfur and sand [9]. To date, few examples of IS developed in [24] showed that Material Flow Analysis (MFA) and LCA are the most emerging countries of America have been found. Brazil stands out as the common methods to evaluate resource savings and environmental per­ most advanced in the exchange of solid waste through IS [10]. For formance of IS networks. MFA is a systematic method to map the instance, Sellitto et al. [11] highlighted that a Brazilian network of changes in flows and stocks of any material-based system [25]. How­ manufacturing companies facilitates the exchange of nearly 300,000 ever, it does not provide information about the potential environmental tonnes of by-products per year. In Argentina, a study proposed a green impacts related to the system. LCA, on the other hand, is suitable to urban planning system that integrates agro-urban symbiosis with cir­ understand better the environmental performance and resource savings cular flows of energy and water [12]. of the system under analysis. As a result, most authors propose using Even though CE has acted as the main driver to set material and LCA to assess the environmental benefits of IS [26]. Besides, Corona energy exchanges, accounting for the potential environmental benefits et al. [27] stated that the advantage of LCA compared to MFA is that the of such exchanges is becoming gradually more relevant. When a life assessment framework could be expanded to an economic LCA –referred cycle perspective is applied, the environmental assessments show that to as Life Cycle Costing– and to a social LCA –referred to as S-LCA– to not all by-product exchanges yield an environmental benefit. For cover all three dimensions of sustainable development. The authors instance, Mohammed et al. [13] found that converting phosphogypsum concluded that, although LCA has a high potential to assess the envi­ –an industrial waste generated by phosphoric acid production– to so­ ronmental benefits of CE strategies, many essential issues remain un­ dium sulphate and ammonium sulphate leads to negative environmental resolved, e.g., allocating environmental costs/benefits in open-loop impacts (global warming potential, eutrophication potential and water recycling. Many authors have performed an environmental evaluation of use). Moreover, Rodríguez et al. [14] evaluated a symbiotic system IS networks with a life cycle perspective [28–32]. For instance, Kerdlap linking a fishmeal facility with a microalgae plant, where flue gas re­ et al. [33] developed a multi-level analysis where LCA results could be places pure CO2. The results showed that a net reduction of greenhouse disaggregated into three levels: network, individual companies and gas (GHG) emissions was achievable only when low-carbon electricity specific flows. The authors stated that this disaggregation allowed each was consumed (less than 9% of fossil fuels). Otherwise, the symbiotic stakeholder to decide whether they participate in the IS network based system entailed a rise in GHG emissions compared to the conventional on their environmental performance. scenario without exchange. Many scholars have used different metrics to evaluate the perfor­ To support the progress toward a CE, indicators have been the focus mance of CE strategies with a particular focus on material efficiency of numerous studies [15–17]. Most authors broadly agree that indicators [34–36]. Niero and Kalbar [37] highlighted the need to couple those can be essential tools for assessing CE at different levels. In the context of metrics with life cycle-based indicators to address CE trade-offs and IS, Dong et al. [18] and Chen et al. [19] have highlighted that methods rebound effects. For instance, Ardente and Mathieux [38] proposed the and indicators are expected to monitor, reflect and quantify the benefits Recyclability Benefit Rate (RBR), which uses LCA results. The RBR is of IS practices. Some of those indicators are referred to as eco-efficiency defined as the ratio of the environmental savings obtained from recy­ indicators, resource efficiency indicators and indicators that reflect the cling a product over the environmental burdens regarding production resilience of the IS network. Nevertheless, one existing limitation of from virgin material followed by disposal (landfill or incineration). most of them is the lack of a systemic perspective for measuring IS at the Huysman et al. [39] further developed the RBR for open-loop recycling entity level [20]. Thus, the aim of this work is to develop an LCA-based (RBROL). This indicator is applicable when the recycled materials are indicator to quantify the individual environmental benefits of only usable for other product applications, as in the case of IS exchanges. by-product exchanges. This indicator can be used as guidance for en­ Subsequently, Huysveld et al. [40] introduced several improvements to tities operating within the IS network. By performing an individual the RBROL indicator. One of the main modifications was the adjustment assessment of each by-product exchange, the indicator would aid en­ of the denominator to consider the two new terms: environmental tities in deciding whether to establish and further encourage synergies. impact of the disposal of the by-product and the environmental impacts Firstly, the paper examines the assessment of IS through methods and of the product made from virgin material (including virgin material indicators. An overview of the methodological considerations of the production, manufacturing, use and disposal). This variation allows a allocation method of environmental burdens in Life Cycle Assessment more accurate comparison between the primary raw materials-based (LCA) is done to in a later stage support the description of the possible and recycled materials-based products. Another notable example is the allocation method for the SPI. Secondly, the related credits in an IS Industrial Symbiosis Indicator proposed by Felicio et al. [41] which network are described, and then the proposed indicator is formulated. combines two methodologies: MFA and Environmental Impact Assess­ The indicator is later used in a case study to illustrate the potential ment. The derived IS indicator was created using the “Environmental environmental benefits of a real industrial network set to exchange Impact Momentum”, an index that considers the qualitative impact plastic by-products. The potential environmental benefits of such ex­ criteria and the quantity of by-products. Table 1 summarizes a list of the change are calculated through LCA and quantified in terms of environ­ diverse environmental assessments and their related indicators (if mental impacts. For simplification purposes, two single life cycle impact available) used to evaluate IS practices retrieved from the literature. assessment methods are used. The resource use is evaluated through the Although the indicators included in Table 1 use a systemic approach CEENE (Cumulative Exergy Extraction from the Natural Environment) to assess the environmental benefits, they are not capable of quantifying method, proposed by Dewulf et al. [21] and Alvarenga et al. [22]. The the individual benefits for each entity involved in the by-product ex­ potential GHG emission savings are calculated through the IPCC 2013 change. Fraccascia and Giannoccaro [20] already highlighted the need GWP 100a method. The paper concludes with a discussion on the po­ to develop indicators assessing the environmental performance at the tential use of the indicator to enhance further IS networks, including company or entity level. Such information is essential to motivate some aspects regarding the allocation method in IS exchanges. companies to engage in an IS practice and identify processes in the IS network to improve their environmental performance [33]. 2 F. Arce Bastias et al. Journal of Environmental Chemical Engineering 11 (2023) 111023 Table 1 Table 1 (continued ) Description of environmental performance methods and indicators used to Method Indicator Description Author assess Industrial Symbiosis practices in existing literature. NA means not available. Sweden using LCA. The assessment is Method Indicator Description Author performed for on-site Life Cycle NA The environmental Daddi et al. processes and for the Assessment (LCA) impacts of the [28] whole cycle, i.e., if current upstream and implementation of IS downstream in an Italian tannery processes are also cluster are assessed included. using LCA by Multi-level matrix- NA This study introduces Kerdlap et al. comparing them based modeling M3-IS-LCA, which is [33] with a non-symbiotic and analysis of constructed so that hypothetical the life cycle LCA results can be scenario. environmental provided at the levels LCA NA This study evaluated Hossain et al. impacts of IS of the network, the life cycle CO2 [29] networks (M3-IS- individual company, emissions of using LCA) and specific resource wood-derived fuel in flows. The case study the co-combustion focuses on a process of a cement potential food waste plant in Hong Kong. valorization IS LCA NA The article evaluates Martin [30] network in the environmental Singapore. performance of an IS LCA Recyclability The proposed Ardente and network comprising Benefit Rate for indicator uses LCA Mathieux firms in waste open-loop results to calculate [38], management, soil, recycling the environmental Huysman surfaces, paper, (RBROL) savings achieved et al. [39], lumber, and energy through recycling a Huysveld sectors. Using LCA, product compared to et al. [40] the study compares producing it from the current IS virgin material and network with a disposing of it. reference scenario Moreover, and a potential improvements have future development. been made to It emphasizes the account for open- importance of loop recycling, recognizing the specifically suited for benefits of IS at both IS exchanges the network level involving the use of and its potential recycling materials advantages for in other product individual firms and applications. products within the Material Flow Industrial The authors Felicio et al. network. Analysis (MFA) Symbiosis proposed an [41] LCA + socio- NA This study presents Martin and and Indicator indicator that economic an LCA and socio- Harris [31] Environmental accounts for assessment economic assessment Impact qualitative impact of an emerging IS Assessment criteria and the network in Sweden, quantity of by- which comprises products. A Brazilian existing firms eco-industrial park engaged in fish was evaluated, and processing three scenarios were industries, algae created: an industrial production, a system park without for recycling plastic symbiosis, with wastes, and a land- symbiosis, and with based salmon farm. perfect symbiosis. The research sheds light on the broader implications and 2.2. Allocation of environmental burdens in LCA benefits of the IS network for the In an IS network, a waste (hereinafter also referred to as by-product) involved firms and generated from one entity enters another entity to be reused or recycled, the overall regional sustainability and thus replaces virgin material input. As a result, the environmental LCA The ratio This study evaluates Røyne et al. burdens from the common process shared among the different entities between the environmental [32] must be partitioned. From a methodological standpoint, this partition­ investment cost impact reduction ing, also known as allocation in LCA, needs to be applied to IS ex­ and reduced potential of two global warming strategies for an changes. Although Mattila et al. [42] addressed the linkages between potential existing chemical LCA and IS networks in detail, only guidance to provide the environ­ industry cluster in mental performance of the system’s total output is given. The ISO 14044:2006 standard for LCA provides a conceptual framework and 3 F. Arce Bastias et al. Journal of Environmental Chemical Engineering 11 (2023) 111023 distinguishes between closed-loop recycling (material substitutes the production of Raw B. Furthermore, if an intermediate process may be virgin material and is used in the identical type of products as before) needed to enable the exchange, the impacts from this process should be and open-loop recycling (material is recycled from one product into equally shared between Entity A and Entity B. The partial allocation was another) [43,44]. This conceptual framework is intended to guide compared with two full allocation methods (100:0, 0:100) considered as practitioners in modelling end-of-life processes; however, it is general limit cases. In the 100:0 allocation, the impact credits produced when and open to various interpretations [45]. Martin et al. [46] have Entity B avoids using raw materials by using a by-product of Entity A are developed methodological considerations focusing on allocation pro­ assigned to Entity A. In the 0:100 allocation method, the impact credits cedures. The authors provided an approach to identify benefit-sharing are assigned to Entity B if it avoids using raw materials by using a between firms and distribute impacts and credits in the IS network. by-product of Entity A. Likewise, Kim et al. [47] applied three by-product impact allocation In this paper, the λ factor is introduced, which will be used to refer to methods (cut-off, avoidance and the 50:50 method) to distribute the the credits for avoiding Raw B. The value of λ factor is 0.5, 1, or 0 ac­ GHG impacts between two companies involved in a real IS exchange cording to the different allocation methods: 50:50, 100:0, or 0:100, developed in Ulsan (Korea). Furthermore, Ekvall et al. [48] highlighted respectively. Fig. 1 shows the meaning of each of these allocation an allocation problem in LCA when recycling systems are assessed since methods conceptually. In certain situations, alternative λ factors, such as two or more products use the same material stemming from another 0.25 or 0.33, might be appropriate to accommodate varying degrees of product’s life cycle. involvement by the relevant entities. However, any deviation from the standard values (0.5, 1, or 0) must always be well-justified with clear 3. Materials and methods motivations. In IS networks, companies are particularly interested in determining 3.2. Definition of the symbiotic performance indicator their individual environmental savings while also considering a holistic approach. When assessing the potential environmental benefits of by- For each entity involved in the by-product exchange, it is possible to product exchanges at the entity level, the allocation procedure for ac­ calculate the Symbiotic Performance Indicator (SPI). The SPI is defined counting for them needs to be well-defined. First, the possible methods as the individual environmental benefits that an entity gains by for allocating the environmental impacts and the allocation factor are participating in a by-product exchange over the environmental impact explained (Section 3.1). Once the allocation factor is defined, the pro­ of a reference scenario with no exchange of materials and energy. Then posed indicator is detailed (Section 3.2). The indicator is then tested in a the SPI of each entity is calculated for any impact category, as shown in real case study, where the entities involved in the IS network and the Eq. (1). possible scenarios are described (Section 3.3). An environmental benefits from by − product exchange assessment using LCA follows (Section 3.4), including the definition of SPI = (1) impact of the entity in the reference scenario the goal and scope, the inventory analysis and the impact assessment. Eq. (2) shows how to calculate the SPI is calculated for the source, 3.1. Allocation factor which is any process or plant that makes a resource available to the general system [49]. Calculating the benefits from the by-product ex­ As Martin et al. [46] stated, there is an unfair distribution of impacts change must consider the recycling rate (r), defined as the amount of and benefits from using by-products as raw materials when applying the recycled material produced over the amount of input waste material. system expansion method. Therefore, the authors recommended the Then, the benefit of the source for participating in the by-product ex­ 50:50 allocation method, which is used to equally distribute credits change is the impact from the disposal of by-product α0 (Dα0 ) minus the between the entities directly involved in the exchange. This approach for impact from intermediate processing (Iα0 ), and the latter is multiplied by crediting entities involves deducting the environmental impact of the the allocation factor (λ). The source is credited with the avoided impact virgin raw material production avoided through by-product exchange. of the production of virgin material α (Vα ), multiplied by the allocation In the 50:50 allocation method, credits are equally shared between both factor and the substitution percentage (m). The substitution percentage entities involved in the exchange. As seen in Fig. 1, By-product A sub­ is defined as the amount of virgin material substituted by the amount of stitutes Raw B; therefore, 50% of the impacts of Raw B are removed from recycled material (m = mv /mr ). Entity B and 50% from Entity A. Entity B must receive the impact for the [ ] virgin Raw B to avoid double-counting of credits created by the avoid­ r • (Dα0 − λ • Iα0 ) + m • λ • Vα ance. Given this, Entity B receives only 50% of the impact for the SPIsource = ⃒ ⃒r • Dα + IENTITY A ⃒ ⃒ (2) 0 Eq. (3) illustrates how to calculate the SPI for the sink, which is any process or plant that consumes a resource [49]. The environmental benefit of the by-product exchange for the sink involves summing the avoided impact of virgin material production with the impact from manufacturing (Mα ), multiplied by the substitution percentage. Then, the impact of the recycling (Rα0 ) and intermediate processing, multiplied by one minus the allocation factor (1 − λ), is subtracted from the previ­ ous result. The sink is also credited for avoiding the production of virgin material. However, the impacts from the production of virgin material are still summed to the sink to avoid double counting of credits created by the avoidance. Therefore, the avoided impact for the virgin produc­ tion, multiplied by the allocation factor and the substitution percentage, is subtracted. [ ( ) ] (m • (Mα + Vα ) ) − (1 − λ) • Iα0 + Rα0 − λ • m • Vα SPIsink = (3) |r • m • (Mα + Vα ) + IENTITY B | Fig. 1. Definition of λ factor for defining possible allocation methods of the potential environmental impacts and burdens (Modified from Martin The definition and design of the reference scenario are essential is­ et al. [46]). sues for the best quantification of the studied scenario’s environmental 4 F. Arce Bastias et al. Journal of Environmental Chemical Engineering 11 (2023) 111023 benefits. In this work, the definition of “reference scenario” proposed by (30% of the total input) –supplied by different informal sorting plants– Aissani et al. [50] was adopted. The reference scenario is a hypothetical and buying recycled plastic pellets (54% of the total input) –half of them non-symbiotic scenario built with no exchanges between the entities supplied by BARESI. A company producing virgin PP supplies the within the IS network. Therefore, the total environmental impacts of remaining fraction (16%). The processing of the mixed plastic waste each entity are evaluated assuming it is in isolation. The environmental consists of the following steps: shredding, separation in water and dry­ impact of the source in the reference scenario is the impact from the ing, and then extrusion. When processing mixed plastic waste, part of disposal of by-product α0 (Dα0 ) multiplied by the recycling rate (r). The the materials is lost during recycling. The recycling rate is assumed to be impacts from processing and other inputs (IENTITY A ) are summed with 90%, indicating that 0.9 kg of recycled plastic is produced per kg of the disposal impact. The environmental impact of the sink is the impact waste input. The recycled pellets and the mixed plastic waste processed of virgin material production, which is summed with the environmental in situ are blended in equal proportions to produce a recycled crate. The impact from manufacturing (Mα + Vα ). This result has to be multiplied plastic crates have an average weight between 0.4 kg and 2.1 kg, which by the recycling rate (r) and the substitution percentage (m). The im­ was estimated based on the data reported by the company. Plastic scrap pacts from processing and other inputs (IENTITY B ) are summed with the is generated during the extrusion process and transported to BARESI. UP latter. cannot reprocess it because its volume exceeds their shredding ma­ As the SPI accounts for net impacts, i.e., burdens minus avoided chine’s capacity. Based on Tua et al. [52], it was assumed that the plastic burdens or benefits, the indicator can have a positive or negative value. scrap generation is 100 g per crate. If the SPI results in a positive value, the larger the value, the more The symbiotic scenario implies that plastic by-products are supplied beneficial is the by-product exchange for the entity. If the SPI results in a to MPM and BARESI. Additionally, intermediate processing is required negative value, the by-product exchange is inconvenient for the entity to upgrade or change the properties of the by-products, such as the from an environmental perspective. The denominator is accounted for as shredding processes done at MPM and BARESI and the transportation of an absolute value to limit the change in the sign and, consequently, the plastic by-products. It was assumed that the same amount of virgin meaning of the indicator. material is substituted by recyclable materials (m=1). In this scenario, there are two material sources (UP and BARESI) and two material sinks 3.3. Case study (MPM and BARESI). It should be noted that BARESI is involved in two exchanges with two different roles. This is an interesting case to illus­ The applicability of the SPI was evaluated using a real case study on trate the use of SPI. plastic by-products (see Fig. 2). These plastic by-products are generated The reference scenario represents the hypothetical non-symbiotic during the processing phase and exchanged between three companies system with no exchange of materials between the three companies. located in Mendoza (Argentina). The dashed line in Fig. 2 represents the Moreover, it was assumed that MPM uses virgin PP to replace the by- system boundary of the study, i.e., establishes the processes considered products and that BARESI buys virgin PP to supply their customers. In in the assessment, and the red arrows depict the by-product exchanges. the reference scenario, plastic wastes were considered disposed of in a Company 1, called BARESI, produces two types of recycled plastic landfill because, as informed by the companies, it was a traditional pellets from plastic scraps and industrial packaging waste: polyethylene practice before the established exchanges. (PE) and polypropylene (PP). Table S1 of the Supporting Information 1 shows the inputs and outputs for BARESI in the IS network. The recycled 3.4. Environmental assessment using LCA plastic pellets are used to produce new products such as industrial packaging, strapping, pipes, hoses and crates. BARESI primarily deals As introduced in Section 3.2, the SPI indicator builds on the results of with post-industrial plastic waste, which is generated during the an LCA. For this reason, the first step in the assessment was to carry out manufacturing stage, as the impurities of this waste stream can be an LCA following the ISO 14040 [43] and 14044 [44] standards. The controlled and minimized. BARESI processes 1200 tonnes of industrial goal of the LCA was first to quantify the environmental impacts (burdens plastic scraps per year and generates 72 tonnes of mixed plastic waste and benefits) of the symbiotic and the reference scenarios, and second to per year, consisting of PP (50%) and PE (50%). provide the data to account for the SPI indicator. A cradle-to-gate Company 2, named Madera Plástica Mendoza (MPM), specializes in analysis was performed using the system boundary described in Fig. 2. producing recycled plastic posts used in vineyards. Table S2 of the The cut-off approach was applied to model recycling in LCA [45]. This Supporting Information 1 shows the inputs and outputs for MPM in the means that plastic waste was considered burden-free. The final handling IS network. MPM can use post-consumer plastic waste, which material of the products and by-products leaving the system was not included. consumers generate after its use (e.g., household waste), because their The environmental benefits of an IS network are accounted for by end-products have low technical requirements. At MPM, the recycling comparing the symbiotic scenario with a reference scenario. In this process consists of three main steps: shredding, further mixing of plastics work, the methodological recommendations for setting the functional and extruding the resulting mix of plastics into posts. This waste treat­ units proposed by Martin et al. [46] were adopted. The IS network ment system was described in an earlier study [51]. The amount of waste produces several main products and by-products as it is a multifunc­ generated is negligible; therefore, the recycling rate is 100% (r = 1). tional system. Therefore, the functional units were set to the annual This indicates that one kg of recycled plastic is produced per kg of waste output (in tonnes per year) of PP and PE pellets, plastic posts, and plastic input. Also, MPM uses an antioxidant* to improve the quality of recycled crates. As a result, both scenarios are designed to produce the same final plastics. The proportion to products of the additive intake is 0.1 wt%. functions and products. The values of the reference flows are listed in Company 3, Urquiza Plásticos (UP), is a producer of plastic crates Table 2. primarily supplied to the agricultural market, covering the needs for Regarding the inventory analysis, data were collected for the fore­ crates for harvest, export and transport trays. Table S3 of the Supporting ground and the background system. Data for the foreground system Information 1 shows the inputs and outputs for UP in the IS network. UP were collected in close collaboration with the companies involved in the specializes in the production of crates made of mainly three materials: plastic waste treatment system. In this study, the background system recycled PP (rPP), recycled PE (rPE) and virgin PP (vPP). The recycled included the data on the production and distribution of diesel, tap water, plastic crates are obtained by processing mixed plastic waste in situ electricity and virgin PP. It was modelled using the Ecoinvent 3.6 database [53] contained in the SimaPro® software (version 9.1.1.1). Infrastructure was not included due to unavailable data. Detailed in­ 1 The chemical compound of this additive is pentaerythritol tetrakis (3-(3,5- formation about the datasets from Ecoinvent is shown in Table S4 of the di-tert-butyl-4-hydroxyphenyl)propionate). Supporting Information 1. 5 F. Arce Bastias et al. Journal of Environmental Chemical Engineering 11 (2023) 111023 Fig. 2. Description of the IS case study in Mendoza (Argentina). All quantities are displayed in tonnes. Section 4.2. The complete results of the calculations can be found in Table 2 Supporting Information 2. Reference flows of entities in the industrial symbiosis network. Entity Flow Annual production Unit 4.1. Environmental assessment of the IS network BARESI Plastic pellets 1128 Tonnes MPM Plastic posts 366 Tonnes The resource use and the GHG emissions of the reference and the UP Plastic crates 566 Tonnes symbiotic scenarios are presented in Table 3. The implementation of exchanges between the industries generates significant benefits in the IS Two impact assessment methods were chosen to account for resource network. The production of virgin PP and electricity are the most use and GHG emissions. Resource use was calculated using the Cumu­ influential processes on the environmental burdens of both scenarios. lative Exergy Extraction from the Natural Environment (CEENE) v. 2013 The IS network can lead to an annual reduction in resource use of [21,22]. Although there is currently no consensus on assessing the im­ nearly 19%. This reduction is primarily attributed to replacing almost pacts of resource use in LCA, Gonzalez Hernandez & Cullen [54] high­ 60% of virgin plastics with recycled plastics. Electricity consumption lighted the main benefits of using exergy for resource accounting: exergy and transportation increase by 3% and 4%, respectively. However, their integrates material and energy into a single metric and evaluates the related impacts are offset by the substitution of virgin plastics. In the quality of resources. Therefore, the CEENE method was applied since it reference scenario, approximately 52% of the overall impacts of the uses exergy to quantify all resources extracted from the natural envi­ system stem from electricity use, followed by the production of virgin PP ronment over the life cycle of a product, and it is expressed in gigajoules with a contribution of 36%. In the symbiotic scenario, the electricity and of exergy (GJex). By definition, exergy is the maximum useful work that the production of virgin PP are responsible for about 66% and 18% of can theoretically be retrieved from a system or resource when it is the resource use, respectively. In both scenarios, the significant contri­ brought into thermodynamic equilibrium with its surroundings through bution of CEENE Fossil to CEENE Total results from the electricity mix, a reversible process [55]. The CEENE method covers the following with fossil sources contributing 73% [56]. categories: fossil energy, nuclear energy, metal ores, minerals, water Regarding GHG emissions, the IS network illustrates environmental resources, land use, abiotic renewable resources (including wind power, impact reductions of roughly 15%; once again, the largest impact re­ geothermal energy and hydropower) and atmospheric resources. More ductions stem from the decrease in the amount of virgin PP. For the detailed information on the CEENE method can be found in Supporting reference scenario, the most significant GHG emissions derive from the Information 1. direct electricity consumed, with a share of 52%. The second most sig­ The GHG emissions were estimated using the IPCC 2013 GWP 100a nificant GHG emissions are from the production of virgin PP, with a method, expressed in tonnes of CO2 equivalents (tCO2-eq). This was contribution of 27%. For the symbiotic scenario, the primary cause of done for two main reasons. First, a reduction of GHG emissions may this environmental impact is electricity consumption, with a contribu­ imply an increase in the use of resources. Therefore, it is relevant to tion of 64%, followed by transport of raw materials, products and waste develop a discussion around these statements. The second reason is that materials, with a contribution of 18%. the accounting method proposed by IPCC is well consolidated and agreed upon among a diverse range of stakeholders, including policy­ 4.2. SPI results for the by-product exchange between BARESI and MPM makers, LCA practitioners and industry. In this by-product exchange, BARESI is considered the “source” and 4. Results MPM defined as the “sink”. The SPI results in positive values for both entities with all three allocation methods. This suggests that both en­ In this work, the potential environmental impacts and savings of an tities benefit from the exchange regarding the two impact categories IS network are expressed in terms of resource use and GHG emissions assessed, as seen in Table 4. In the IS network, this synergy involves the –calculated through LCA– by comparing a reference scenario with a exchange of 72 tonnes of mixed plastic waste, whereas in the reference symbiotic scenario. In the reference scenario, plastic waste streams are scenario, MPM consumes 72 tonnes of virgin PP, and the mixed plastic disposed of in a landfill, and companies consume virgin PP to meet their waste is landfilled. annual production demand. In the symbiotic scenario, virgin plastics are In the case of the 0:100 allocation method (λ = 0), the SPI has its replaced by plastic by-products generated by two local entities. The lowest values for BARESI and the highest for MPM when assessing re­ results of the overall environmental impacts at the network level are sources and GHG emissions. The source results are because the numer­ presented in Section 4.1, while the results at the entity level are given in ator only considers the landfilling process. The credits and the 6 F. Arce Bastias et al. Journal of Environmental Chemical Engineering 11 (2023) 111023 Table 3 Results of the resource use (in Gigajoule of exergy) and the GHG emissions (in tonnes of CO2-eq) of the Reference and the Symbiotic scenarios. Scenario CEENE GHG emissions Metals Minerals Nuclear Land Renewables Water Fossil Total GJex % GJex % GJex % GJex % GJex % GJex % GJex % GJex % tonnes of CO2-eq Reference 0.25 0.0 22.5 0.0 2740 4.6 2690 4.5 5260 8.8 6530 11.0 42,400 71.1 59,600 100.0 2070 Symbiotic 0.15 0.0 11.5 0.0 2540 5.2 2720 5.6 5320 11.0 6170 12.8 31,700 65.4 48,500 100.0 1770 Table 4. The synergy implies the exchange of 76 tonnes of plastic scrap. Table 4 In the case of the 0:100 allocation method (λ = 0), the highest value Results of the SPI for the entities in the IS network using three different allo­ in the numerator of BARESI is achieved. The results are 130.7 tCO2-eq cation methods, in terms of resource use (CEENE Total) and GHG emissions. and 5316.1 GJex, whereas the numerator of UP has its lowest values Allocation Entity CEENE Total GHG emissions (12.8 tCO2-eq and 25.9 GJex). Clearly, BARESI has the most consider­ method BARESI → UP → BARESI → UP → able reductions in environmental impacts stemming from reducing the MPMa BARESIb MPMa BARESIb production of 76 tonnes of virgin PP. λ=0 BARESI 0.07% 20.11% 1.17% 13.30% In the case of the 50:50 allocation method, credits are equally (0:100) MPM 72.77% - 65.32% - distributed between BARESI and UP, resulting in both companies UP - 0.10% - 1.54% achieving benefits from the exchange. The numerator of the sink reaches λ = 0.5 BARESI 11.25% 8.10% 8.88% 4.77% (50:50) MPM 37.46% - 34.13% - 47.5 tCO2-eq and 2141.1 GJex. UP attains greater benefits from the UP - 12.91% - 11.56% exchange because the numerator results in 96.0 tCO2-eq and 3200.9 λ=1 BARESI 22.43% − 3.91% 16.58% − 3.58% GJex. (100:0) MPM 2.15% - 2.94% - In the case of the 100:0 allocation method, the numerator of the sink UP - 25.71% - 21.59% results in negative values (− 35.6 tCO2-eq and − 1033.8 GJex), and then a BARESI → MPM represents the exchange between BARESI (source) and MPM it suggests that the entity does not achieve a reduction in GHG emissions (sink) and resource use. The negative value is mainly influenced by the net b UP → BAR represents the exchange between UP (source) and BARESI (sink) impact of the recycling process of the waste stream. Nevertheless, the source illustrates the largest impact reductions since the numerator re­ environmental impacts of intermediate processing are not considered for sults in positive values (179.2 tCO2-eq and 6375.9 GJex). BARESI who acts as a source entity in the 0:100 allocation method. Another explanation for these results is that the denominator of the 5. Discussion source has a significant value because it is highly influenced by the electricity consumption of the company in the reference scenario. In the 5.1. Benefits and limitations of using the Symbiotic Performance Indicator SPI of the MPM who acts as a sink entity, the numerator accounts for the difference between the recycling process of 72 tonnes of mixed plastic In this paper, the environmental performance of the IS network is waste and the production of the same amount of virgin material. The assessed through LCA by comparing a symbiotic scenario with a refer­ numerator results in a positive value (160.7 tCO2-eq and 6090.3 GJex), ence scenario. Results were calculated at both network and entity levels, which indicates that the by-product exchange is environmentally which are crucial for assessing IS network performance. Policymakers feasible. would be interested in conventional network-level results when assess­ In the case of the 50:50 allocation method (λ = 0.5), the credits for ing the potential for environmental improvement in conventional in­ avoiding virgin PP production are distributed equally between the dustrial parks, municipalities, or regional areas where industrial activity source and the sink. The numerator of the source accounts for 50% of the concentrates. Conversely, the SPI indicator could be used for several credits and 50% of the environmental impacts of the intermediate pro­ ends by individual entities or companies operating in the IS network. cessing; therefore, the results are 88.4 tCO2-eq and 2972.7 GJex. The They would be interested in accounting for their own environmental numerator of the sink accounts for nearly the same values as the source benefits, as this would help them decide to set and further promote the (83.9 tCO2-eq and 3135.2 GJex). Although the numerators are virtually by-products exchanges. Indeed, this has been previously discussed by identical, the SPI results differ significantly; the difference in their de­ Fraccascia and Giannoccaro [20]. The SPI can help assess the environ­ nominators explains these results. The denominator of MPM is lower mental benefits attained by an entity according to its role (source or because, in contrast to BARESI, MPM accounts for 87% less electricity sink) in the exchange. For instance, BARESI’s assessment results in two consumption. SPI values as this entity is involved in two by-product exchanges. Since Finally, using the 100:0 (λ = 1) allocation method, the SPI results of one exchange is independent of the other, both results should be sepa­ BARESI are higher than the SPI of MPM. The numerator of the source rately evaluated. Therefore, the SPI helps to decide whether to partici­ results in 165.2 tCO2-eq and 5927.9 GJex, whereas for the sink, the pate in a specific exchange or not. Additionally, this indicator could results are 7.2 tCO2-eq and 180.1 GJex. Although the sink does not ac­ attract LCA practitioners aiming to assess the potential environmental count for the environmental impacts of the intermediate processing in impact of IS networks, as the SPI is fully aligned with LCA framework the numerator, the lower results in its SPI are mainly because the source and can be easily derived from conventional LCA results. From a receives the total credits for avoiding raw material production. methodological perspective, these findings are consistent with the multi-level approach developed by Kerdlap et al. [33]. The SPI is a scientifically sound indicator that is useful for guiding 4.3. SPI results for the by-product exchange between UP and BARESI companies towards reducing their overall carbon footprint by estab­ lishing agreements for the material exchange at the local or regional In this by-product exchange, UP is considered the “source” and level. As illustrated in the paper, this indicator has proven to be useful in BARESI defined as the “sink”. This by-product exchange benefits both measuring potential reductions at the entity level. These results help companies when applying the 0:100 and 50:50 allocation methods. emphasize the role that material reuse can play in enhancing resource Nevertheless, the results for BARESI show that the SPI is negative for the use and reducing GHG emissions at a broader geographical scale, two impact categories with the 100:0 allocation method, as shown in 7 F. Arce Bastias et al. Journal of Environmental Chemical Engineering 11 (2023) 111023 particularly in the context of conventional industrial parks. One of the In the 100:0 allocation method, the SPI of the source accounts for the benefits of using the SPI compared to other indicators is that it quantifies virgin material avoided through the by-product exchange. The greater the potential for environmental impact reduction feasible to be imple­ the environmental burdens of virgin material production (Vα1 ), the mented locally and considering the use of resource and the generation of larger the benefits attained. Therefore, this method incentivizes the emissions from a life cycle perspective. source to participate in the IS network even when waste disposal has Despite its importance and potential usefulness, the SPI has some lower environmental impacts than intermediate processing (Dα0 < Iα0 ). limitations. Further studies should be conducted to generalize the in­ The sink only benefits from the exchange when the product made from dicator’s usefulness. Aspects that could be explored further include virgin material has higher environmental burdens than the one made other processes, IS networks, typologies of industries and geographical from recycled material (m • Mα > Rα0 + Mα ). locations. Case studies involving energy and/or water exchanges should Regarding the 50:50 method, the source and the sink receive credits be conducted to demonstrate the applicability of the indicator and the for the by-product exchange; however, this method could be questioned. type of information managers can collect to support their decision- Kim et al. [47] concluded that the allocation method should be chosen to making. This will also provide a deeper understanding of the SPI and distribute credits achieved by IS exchanges according to their contri­ how it can be applied to other types of exchanges. Therefore, it would be bution to exchange development and maintenance. Therefore, the 50:50 beneficial to validate the indicator across other IS networks. allocation method is recommended from the perspective of equivalent responsibility and benefit-sharing. Different allocation factors, such as 5.2. Implications on entities engaged in an IS network 0.25 or 0.33, might be appropriate to reflect the different levels of commitment of the relevant entities. However, there is a need to set The environmental performance assessment of IS networks has im­ rules to assign value to the allocation factor for each analysis, as dis­ plications beyond quantifying overall impacts and benefits. A focus on cussed by Viganò et al. [57]. They point out that this decision will the entities of the symbiosis could help identify those who gain from it or fundamentally depend on the industries involved and the practitioner’s are mostly affected by the symbiotic activity [46]. Companies operating experience. in the IS network could use the SPI to identify if the by-product ex­ changes provide benefits. Such information could offer data that moti­ 6. Conclusions vates the initiation of new exchanges, especially when considering factors beyond just economic benefits. Additionally, the indicator can A review of previous studies concluded that there is a lack of in­ aid in promoting industrial symbiosis for business connections, process dicators for quantifying the benefits of IS at the entity level using a improvement and environmental performance communication [46]. holistic approach. This paper provides an LCA-based indicator, defined Conversely, the SPI could identify if the by-product exchange is incon­ as the environmental benefits that each entity (“source” and “sink”) venient for the entity from an environmental perspective. If benefits are obtains from a by-product exchange, compared to a non-symbiotic not attained, the use of conventional raw materials would be the reference scenario. The allocation of impacts and benefits was identi­ preferred option over exchanging materials and/or energy. This has fied as an important methodological aspect. Thus, an allocation factor been observed in the exchange where UP acts as a source and provides was introduced in the formulation to include the distribution of credits plastic waste to BARESI, and the allocation method only benefits the and impacts. To evaluate the influence of the allocation factor, three sourcing company (UP). In such cases, the results of the SPI were about allocation methods were considered in this study: full allocation (100:0 –4%. Since BARESI did not attain a reduction in environmental impacts, and 0:100) and partial allocation (50:50). they could be discouraged from participating in the by-product ex­ The SPI was used in an IS case study about plastic waste treatment to change. Alternatively, BARESI could potentially start a discussion illustrate its applicability. The IS network involves two exchanges of regarding the pricing of the by-products or perform a more detailed post-industrial plastic waste between three companies. At the network assessment of the potential reuse of other materials to compensate for level, the results showed annual reductions in resource use and GHG their environmental impact. emissions of 19% and 15%, respectively. These impacts were reduced due to a decrease in virgin plastic inputs, and also the related con­ 5.3. Allocation methods in SPI sumption of materials and energy from cradle to gate. At the entity level, each entity was assessed using the proposed indicator. The SPI for the As Kerdlap et al. [33] state, the LCA of IS networks should be flexible three entities resulted in a positive value except for BARESI, when in selecting different allocation methods according to the context. considering them as sink of the exchange and assigning the full envi­ Indeed, the allocation method chosen plays a significant role in the re­ ronmental benefits to UP who acts as the source entity (100:0 allocation sults of the SPI. In the 0:100 allocation method, each entity is assigned method). The SPI results indicated that the allocation factor significantly the environmental burdens of the processes in its life cycle, which as­ influences the individual environmental benefits. signs credits to the sink. When defining the boundary between the life Industry practitioners seeking to promote IS will find the practical cycle of the entities involved in the exchange, the boundary is set before implications of the SPI significant. The SPI can be used to make informed the waste material is collected for the exchange. This approach benefits decisions about which synergies to implement. The indicator can also the entity using recycled material (sink) as long as the product made help companies identify areas where they need to improve their business from virgin material has higher environmental burdens than those made model and quantify potential environmental savings. For example, by from recycled material. When this allocation method is applied to the identifying potential improvements in production processes, interme­ SPI of the source, the entity only benefits from the exchange if final diate processing, or waste management practices. Finally, the results of disposal has a negative net impact on the environment (r • Dα0 > 0). the SPI can serve as a way for companies to demonstrate and commu­ However, the entity cannot have an incentive to participate in the ex­ nicate their commitment to sustainability to consumers and other change when waste disposal has a positive net impact on the environ­ stakeholders, potentially gaining a competitive advantage in the market. ment (r • Dα0 < 0). For example, if final disposal is incineration with The SPI supports the environmental management system already energy recovery, avoided burdens may be larger than the induced bur­ implemented by companies, by providing a more realistic measure of the dens. If this net impact results in a negative value, the SPI also results in benefits achieved. However, the challenge for future research will be to a negative value, and then the exchange is not convenient for the source. include an economic and social analysis to complement the environ­ This approach discourages waste suppliers from participating in mental analysis of this study. by-product exchanges even when recycling has lower environmental burdens. 8 F. Arce Bastias et al. Journal of Environmental Chemical Engineering 11 (2023) 111023 Funding [11] M.A. Sellitto, F.K. Murakami, M.A. Butturi, S. Marinelli, N. Kadel Jr, B. Rimini, Barriers, drivers, and relationships in industrial symbiosis of a network of Brazilian manufacturing companies, Sustain. Prod. Consum. 26 (2021) 443–454, https://doi. This research was funded by Consejo Nacional de Investigaciones org/10.1016/j.spc.2020.09.016. Científicas y Técnicas-CONICET, Argentina. [12] N. Bonilla-Gámez, S. Toboso-Chavero, F. Parada, B. Civit, A.P. Arena, J. Rieradevall, X. Gabarrell Durany, Environmental impact assessment of agro- services symbiosis in semiarid urban frontier territories. Case study of Mendoza CRediT authorship contribution statement (Argentina), Sci. Total Environ. 774 (2021), 145682, https://doi.org/10.1016/j. scitotenv.2021.145682. Fernando Arce Bastias: Conceptualization, Data curation, Investi­ [13] F. Mohammed, W.K. Biswas, H. Yao, M. Tadé, Sustainability assessment of symbiotic processes for the reuse of phosphogypsum, J. Clean. Prod. 188 (2018) gation, Methodology, Validation, Visualization, Writing – original draft, 497–507, https://doi.org/10.1016/j.jclepro.2018.03.309. Writing – review & editing. Paula Daniela Rodríguez: Methodology, [14] P.D. Rodríguez, F. Arce Bastias, A.P. Arena, Modeling and environmental Investigation, Validation, Writing – review & editing. Alejandro Pablo evaluation of a system linking a fishmeal facility with a microalgae plant within a circular economy context, Sustain. Prod. Consum. 20 (2019) 356–364, https://doi. Arena: Investigation, Resources, Methodology, Validation, Writing – org/10.1016/j.spc.2019.08.007. review & editing, Funding acquisition. Laura Talens Peiró: Investiga­ [15] V. Elia, M.G. Gnoni, F. Tornese, Measuring circular economy strategies through tion, Methodology, Validation, Writing – review & editing, Supervision, index methods: a critical analysis, J. Clean. Prod. 142 (2017) 2741–2751, https:// doi.org/10.1016/j.jclepro.2016.10.196. Funding acquisition. [16] G. Moraga, S. Huysveld, F. Mathieux, G.A. Blengini, L. Alaerts, K. Van Acker, S. de Meester, J. Dewulf, Circular economy indicators: what do they measure? Resour. Conserv. Recycl. 146 (2019) 452–461, https://doi.org/10.1016/j. Declaration of Competing Interest resconrec.2019.03.045. [17] M. Saidani, B. Yannou, Y. Leroy, F. Cluzel, A. Kendall, A taxonomy of circular The authors declare that they have no known competing financial economy indicators, J. Clean. Prod. 207 (2019) 542–559, https://doi.org/ 10.1016/j.jclepro.2018.10.014. interests or personal relationships that could have appeared to influence [18] L. Dong, G.N. Taka, D. Lee, Y. Park, H.S. Park, Tracking industrial symbiosis the work reported in this paper. performance with ecological network approach integrating economic and environmental benefits analysis, Resour. Conserv. Recycl. 185 (2022), 106454, Data Availability https://doi.org/10.1016/j.resconrec.2022.106454. [19] X. Chen, M. Dong, L. Zhang, X. Luan, X. Cui, Z. Cui, Comprehensive evaluation of environmental and economic benefits of industrial symbiosis in industrial parks, Data can be found in the supporting materials. J. Clean. Prod. 354 (2022), 131635, https://doi.org/10.1016/j. jclepro.2022.131635. [20] L. Fraccascia, I. Giannoccaro, What, where, and how measuring industrial Acknowledgements symbiosis: a reasoned taxonomy of relevant indicators, Resour., Conserv. Recycl. 157 (2020), 104799, https://doi.org/10.1016/j.resconrec.2020.104799. The authors wish to thank Carlos Arce, Leonardo Cano y Pio De [21] J. Dewulf, M.E. Bösch, B.De Meester, G. Van Der Vorst, H.Van Langenhove, S. Hellweg, M.A.J. Huijbregts, Cumulative exergy extraction from the natural Amoriza from Madera Plástica Mendoza, Carlos Lara from BARESI, and environment (CEENE): a comprehensive life cycle impact assessment method for Javier Chiappa from Urquiza Plástico for their support and the provided resource accounting, Environ. Sci. Technol. 41 (2007) 8477–8483, https://doi.org/ data. This article contributes to ICTA-UAB ‘María de Maeztu Unit of 10.1021/es0711415. [22] R.A.F. Alvarenga, J. Dewulf, H. Van Langenhove, A new natural resource balance Excellence’ (MDM-2015–0552; MDM-2019–0940). and to the research indicator for terrestrial biomass production systems, Ecol. Indic. 32 (2013) group Sostenipra 2021SGR00734. 140–146, https://doi.org/10.1016/j.ecolind.2013.03.029. [23] C. Wadström, M. Johansson, M. Wallén, A framework for studying outcomes in industrial symbiosis, Renew. Sustain. Energy Rev. 151 (2021), https://doi.org/ Appendix A. Supporting information 10.1016/j.rser.2021.111526. [24] S. Harris, M. Martin, D. Diener, Circularity for circularity’s sake? Scoping review of Supplementary data associated with this article can be found in the assessment methods for environmental performance in the circular economy. Sustain. Prod. Consum. 26 (2021) 172–186, https://doi.org/10.1016/j. online version at doi:10.1016/j.jece.2023.111023. spc.2020.09.018. [25] P.H. Brunner, H. Rechberger, Handbook of Material Flow Analysis, CRC Press, Boca References Raton: Taylor & Francis, CRC Press, 2017, https://doi.org/10.1201/ 9781315313450. [26] A. Neves, R. Godina, S.G. Azevedo, J.C.O. Matias, A comprehensive review of [1] P. Ghisellini, C. Cialani, S. Ulgiati, A review on circular economy: the expected industrial symbiosis, J. Clean. Prod. 247 (2020), 119113, https://doi.org/10.1016/ transition to a balanced interplay of environmental and economic systems, j.jclepro.2019.119113. J. Clean. Prod. 114 (2016) 11–32, https://doi.org/10.1016/j.jclepro.2015.09.007. [27] B. Corona, L. Shen, D. Reike, J. Rosales Carreón, E. Worrell, Towards sustainable [2] R. Merli, M. Preziosi, A. Acampora, How do scholars approach the circular development through the circular economy—a review and critical assessment on economy? A systematic literature review, J. Clean. Prod. 178 (2018) 703–722, current circularity metrics, Resour. Conserv. Recycl. 151 (2019), 104498, https:// https://doi.org/10.1016/j.jclepro.2017.12.112. doi.org/10.1016/j.resconrec.2019.104498. [3] J. Kirchherr, D. Reike, M. Hekkert, Conceptualizing the circular economy: an [28] T. Daddi, B. Nucci, F. Iraldo, Using life cycle assessment (LCA) to measure the analysis of 114 definitions, Resour., Conserv. Recycl. 127 (2017) 221–232, https:// environmental benefits of industrial symbiosis in an industrial cluster of SMEs, doi.org/10.1016/j.resconrec.2017.09.005. J. Clean. Prod. 147 (2017) 157–164, https://doi.org/10.1016/j. [4] European Commission, Circular Economy Action Plan. For a cleaner and more jclepro.2017.01.090. competitive Europe, Brussels, Belgium, 2020. 〈https://eur-lex.europa.eu/leg [29] M.U. Hossain, C.S. Poon, M.Y. Kwong Wong, A. Khine, Techno-environmental al-content/EN/TXT/?uri=CELEX:52020DC0098〉. feasibility of wood waste derived fuel for cement production, J. Clean. Prod. 230 [5] M.R. Chertow, Industrial symbiosis: literature and taxonomy, Annu. Rev. Energy (2019) 663–671, https://doi.org/10.1016/j.jclepro.2019.05.132. Environ. 25 (2000) 313–337, https://doi.org/10.1146/annurev.energy.25.1.313. [30] M. Martin, Evaluating the environmental performance of producing soil and [6] Y. Zhang, H. Zheng, B. Chen, M. Su, G. Liu, A review of industrial symbiosis surfaces through industrial symbiosis, J. Ind. Ecol. 24 (2020) 626–638, https://doi. research: theory and methodology, Front. Earth Sci. 9 (2015) 91–104, https://doi. org/10.1111/jiec.12941. org/10.1007/s11707-014-0445-8. [31] M. Martin, S. Harris, Prospecting the sustainability implications of an emerging [7] J. Ehrenfeld, N. Gertler, Industrial ecology in practice: the evolution of industrial symbiosis network, Resour., Conserv. Recycl. 138 (2018) 246–256, interdependence at Kalundborg, J. Ind. Ecol. 1 (1997) 67–79, https://doi.org/ https://doi.org/10.1016/j.resconrec.2018.07.026. 10.1162/jiec.1997.1.1.67. [32] F. Røyne, R. Hackl, E. Ringström, J. Berlin, Environmental evaluation of industry [8] S.V. Valentine, Kalundborg symbiosis: fostering progressive innovation in cluster strategies with a life cycle perspective: Replacing fossil feedstock with environmental networks, J. Clean. Prod. 118 (2016) 65–77, https://doi.org/ forest-based feedstock and increasing thermal energy integration, J. Ind. Ecol. 22 10.1016/j.jclepro.2016.01.061. (2018) 694–705, https://doi.org/10.1111/jiec.12620. [9] S.E. Danielsson, P. Møller, L. Randers, Modelling CO2 savings and economic [33] P. Kerdlap, J.S.C. Low, D.Z.L. Tan, Z. Yeo, S. Ramakrishna, M3-IS-LCA: A benefits for the Kalundborg Symbiosis, (2018) 1–11. http://www.symbiosis.dk/ methodology for multi-level life cycle environmental performance evaluation of wp-content/uploads/2018/06/KalundborgSymbiosisPaperLCA2018–06.pdf industrial symbiosis networks, Resour., Conserv. Recycl. 161 (2020), 104963, (accessed April 28, 2021). https://doi.org/10.1016/j.resconrec.2020.104963. [10] E.A. Boom Cárcamo, R. Peñabaena-Niebles, Opportunities and challenges for the [34] M. Linder, S. Sarasini, P. van Loon, A metric for quantifying product-level waste management in emerging and frontier countries through industrial circularity, J. Ind. Ecol. 21 (2017) 545–558, https://doi.org/10.1111/jiec.12552. symbiosis, J. Clean. Prod. 363 (2022), 132607, https://doi.org/10.1016/j. jclepro.2022.132607. 9 F. Arce Bastias et al. Journal of Environmental Chemical Engineering 11 (2023) 111023 [35] M. Saidani, B. Yannou, Y. Leroy, F. Cluzel, How to assess product performance in [47] H.-W. Kim, S. Ohnishi, M. Fujii, T. Fujita, H.-S. Park, Evaluation and allocation of the circular economy? Proposed requirements for the design of a circularity greenhouse gas reductions in industrial symbiosis, J. Ind. Ecol. 22 (2018) 275–287, measurement framework, Recycling 2 (2017) 6, https://doi.org/10.3390/ https://doi.org/10.1111/jiec.12539. recycling2010006. [48] T. Ekvall, A. Björklund, G. Sandin, K. Jelse, J. Lagergren, M. Rydberg, Modeling [36] L. Talens Peiró, B. García Fernández, X. Gabarrell i Durany, Investigating a repair recycling in life cycle assessment, 2020. workshop: the reuse of washing machines in Barcelona, Sustain. Prod. Consum. 29 [49] C.A. Kastner, R. Lau, M. Kraft, Quantitative tools for cultivating symbiosis in (2022) 171–179, https://doi.org/10.1016/j.spc.2021.10.003. industrial parks; a literature review, Appl. Energy 155 (2015) 599–612, https:// [37] M. Niero, P.P. Kalbar, Coupling material circularity indicators and life cycle based doi.org/10.1016/j.apenergy.2015.05.037. indicators: a proposal to advance the assessment of circular economy strategies at [50] L. Aissani, A. Lacassagne, J. Bahers, S.L. Féon, Life cycle assessment of industrial the product level, Resour., Conserv. Recycl. 140 (2019) 305–312, https://doi.org/ symbiosis: a critical review of relevant reference scenarios, J. Ind. Ecol. 23 (2019) 10.1016/j.resconrec.2018.10.002. 972–985, https://doi.org/10.1111/jiec.12842. [38] F. Ardente, F. Mathieux, Identification and assessment of product’s measures to [51] F. Arce-Bastias, Beneficios ambientales del reciclaje de residuos plásticos improve resource efficiency: the case-study of an energy using product, J. Clean. posconsumo para la producción de postes en Mendoza, Argentina, Rev. U.D.C.A. Prod. 83 (2014) 126–141, https://doi.org/10.1016/j.jclepro.2014.07.058. Actual. Divulg. Científica. 25 (2022) 1–11, https://doi.org/10.31910/rudca.v25. [39] S. Huysman, S. Debaveye, T. Schaubroeck, S.D. Meester, F. Ardente, F. Mathieux, nSupl.1.2022.2145. J. Dewulf, The recyclability benefit rate of closed-loop and open-loop systems: a [52] C. Tua, L. Biganzoli, M. Grosso, L. Rigamonti, Life cycle assessment of reusable case study on plastic recycling in Flanders, Resour., Conserv. Recycl. 101 (2015) plastic crates (RPCs), Resources 8 (2019) 110, https://doi.org/10.3390/ 53–60, https://doi.org/10.1016/j.resconrec.2015.05.014. resources8020110. [40] S. Huysveld, S. Hubo, K. Ragaert, J. Dewulf, Advancing circular economy benefit [53] G. Wernet, C. Bauer, B. Steubing, J. Reinhard, E. Moreno-Ruiz, B. Weidema, The indicators and application on open-loop recycling of mixed and contaminated ecoinvent database version 3 (part I): overview and methodology, Int. J. Life Cycle plastic waste fractions, J. Clean. Prod. 211 (2019) 1–13, https://doi.org/10.1016/ Assess. 21 (2016) 1218–1230, https://doi.org/10.1007/s11367-016-1087-8. j.jclepro.2018.11.110. [54] A.Gonzalez Hernandez, J.M. Cullen, Exergy: a universal metric for measuring [41] M. Felicio, D. Amaral, K. Esposto, X. Gabarrell Durany, Industrial symbiosis resource efficiency to address industrial decarbonisation, Sustain. Prod. Consum. indicators to manage eco-industrial parks as dynamic systems, J. Clean. Prod. 118 20 (2019) 151–164, https://doi.org/10.1016/j.spc.2019.05.006. (2016) 54–64, https://doi.org/10.1016/j.jclepro.2016.01.031. [55] R.U. Ayres, L. Talens Peiró, G. Villalba Méndez, Exergy efficiency in industry: [42] T. Mattila, S. Lehtoranta, L. Sokka, M. Melanen, A. Nissinen, Methodological where do we stand? Environ. Sci. Technol. 45 (2011) 10634–10641, https://doi. aspects of applying life cycle assessment to industrial symbioses, J. Ind. Ecol. 16 org/10.1021/es202193u. (2012) 51–60, https://doi.org/10.1111/j.1530-9290.2011.00443.x. [56] International Energy Agency, Monthly electricity production and trade data for all [43] International Organization for Standardization, ISO 14040:2006 Environmental OECD member countries and electricity production data for a selection of other management - Life cycle assessment - Principles and framework, (2006). economies, (2021). https://www.iea.org/data-and-statistics/data-tools/monthly- [44] International Organization for Standardization, ISO 14044:2006 Environmental electricity-statistics. management - Life cycle assessment - Requirements and guidelines, (2006). [57] E. Viganò, C. Brondi, S. Cornago, A. Caretta, L. Bua, L. Carnelli, G. Dotelli, [45] K. Allacker, F. Mathieux, S. Manfredi, N. Pelletier, C. De Camillis, F. Ardente, M. Martin, A. Ballarino, The LCA modelling of chemical companies in the industrial R. Pant, Allocation solutions for secondary material production and end of life symbiosis perspective: allocation approaches and regulatory framework. Life Cycle recovery: proposals for product policy initiatives, Resour. Conserv. Recycl. 88 Assessment in the Chemical Product Chain, Springer International Publishing, (2014) 1–12, https://doi.org/10.1016/j.resconrec.2014.03.016. Cham, 2020, pp. 75–98, https://doi.org/10.1007/978-3-030-34424-5_4. [46] M. Martin, N. Svensson, M. Eklund, Who gets the benefits? An approach for assessing the environmental performance of industrial symbiosis, J. Clean. Prod. (2015), https://doi.org/10.1016/j.jclepro.2013.06.024. 10