Heat Exchanger Networks: Applications for Industrial Integrations

Abstract

Heat integration is a crucial concept in process engineering and energy management. It refers to using heat exchangers and process modifications to maximize energy efficiency, lowering cost and/or carbon emissions within industrial processes through minimizing the external heating and cooling requirements (utility savings). There are two key aspects of heat integration. “Heat Exchanger Network” is an approach to designing efficient connections among the heat exchangers to transfer heat between several hot and cold streams. “Pinch Analysis” is a systematic methodology that determines the optimal energy recovery by identifying the “pinch point” to maximize heat recovery. The paper aims to review the actual status of research in the field of application of heat exchanger networks for industrial integrations and highlight the perspectives.

1. Introduction

Global warming, driven by human activities, poses a serious threat to sustainable development and life on our planet. The increasing global average temperature appears to be causing extreme weather events in every corner of the Earth, bringing substantial economic and social challenges. The combustion of fossil fuels is widely regarded as a primary cause of this phenomenon by intensifying the atmospheric greenhouse effect.

In 2022, fossil fuels dominated the energy mix in the European Union (EU), accounting for approximately 71% of the available energy sources. Industry and buildings are about 25% and 40%, respectively, among the top energy consumers in the EU, with a significant proportion of their energy coming from fossil fuels. While a part of the energy consumption is used for production and construction in these sectors, a significant share of the energy is consumed for providing heat. Heat integration (HI), the most established area of process integration, has gained increasing importance in recent decades. Eco-industrial parks (EIPs) and industrial–urban symbiosis initiatives have embraced HI techniques to resemble the closed circuits of natural materials and energy flows to increase overall efficiency. The idea is to reduce heating and cooling demands using available heat sources and sinks.

The design of a heat exchanger network (HEN) is the methodological platform through which HI is usually implemented. Therefore, significant literature efforts have focused on formulating optimization problems to determine optimal solutions concerning energy efficiency or cost effectiveness.

The focus of this work is the topic of HI. HI aims to recover and reuse thermal energy within and beyond industrial systems, thereby reducing the overall energy demand and costs. HENs are the practical implementation of HI principles. The synthesis and optimization of HENs are crucial for an effective HI. Therefore, “heat exchanger network” has been selected as the main keyword for the literature review.

2. Bibliometric Study

Modern methods of search machines and keyword clustering have been applied by the authors to this set of information. Figure 1a shows the number of publications worldwide on the topic of HENs found in the Scopus database for the time period between 1982 and 2024. Of the 3238 Scopus documents, 2893 are in English. A significant rise in the number of documents can be observed from the early 2000s for approximately 15 years, reaching a sharp peak in 2018. Figure 1b shows the top international journals in this field.

The distribution of publications among subject areas is shown in Figure 2. Expectedly, chemical engineering holds the highest share since the topic of HENs was born to increase the overall efficiency of chemical processes and plants. The other fields with high reference to HENs are engineering in general, energy, and chemistry. An interesting observation is the relatively high share of computer science and mathematics, which shows the growing use of mathematical programming and optimization methods to solve problems related to this field. However, during the 2018–2024 period, the number of chemical engineering-related applications dropped. Figure 1b confirms this observation. The “Chemical Engineering Transaction” journal (with 30+ publications in the year 2018) had no more publications in the field of HEN for several years. In contrast, during the last few years, the subject areas of energy, engineering, and environment have grown.

After two-level filtering, 1890 Scopus documents were further evaluated to discover the keyword clustering using VOSViewer software v. 1.6.16 (Figure 3).

The first level of keywords (used for the filtering) includes heat integration/indirect heat integration, synthesis, operation and control/control structure, and optimization. It also concerns the synthesis of intermediate heat recovery circuits as a special form of HENs. The term more commonly found in the literature for recovering thermal energy is “Total Site Heat Integration” (TSHI). For evaluation purposes, the thermodynamic-based methods (minimizing the entropy generation or exergy analysis) within the framework of the HEN are used for maximum energy recovery or mathematical optimization methods to minimize the total annual costs.

The second level of keywords includes modern application aspects: eco-industrial parks, total site heat integration, and flexibility. In the second group of keywords, the subject of operation and control of HENs is the focus. Control structures are evaluated regarding their effect on the thermodynamic performance of HENs, again by exergy-based formulation of the optimization problem, when designing the control structure.

The bibliometric network shown in Figure 3 highlights the missing or weak co-occurrence of the selected keywords. The limited number and weak connections of keywords are evident, suggesting a lack of comprehensive studies or insufficient integration of the keywords used as two-level filters. Eco-industrial parks are a separate cluster that is a long distance from the center. However, this topic is precisely mentioned within modern publications as the most energetically, economically, and environmentally promising for applying the HENs and hybridizations of HENs with additional options.

A further narrative literature review discusses a comprehensive overview of the history and development of eco-industrial parks (EIPs). The most attended topics of research in EIPs appear to be the following:

• Reports on the practical experiences as well as guidelines in several countries;
• Industrial symbiosis tools and optimization;
• The topic of HI and, more recently, the simultaneous integration of heat and work;
• Integration and management of water/wastewater and material.

3. Industrial Symbiosis and Eco-Industrial Parks

3.1. Industrial Symbiosis

The idea of a special economic zone was first proposed in the 1950s in Shannon, Ireland, to rescue the local airport from shutdown, and the concept has since been applied in thousands of special zones worldwide (Zeng, 2015 [1]). One of the various types of special economic zones is industrial parks. Industrial parks, also addressed by many other names, such as industrial zones, industrial estates, industrial investment regions, etc., are networks of industries, often heavy industry and manufacturing, located collectively in a geographically defined and usually secured area having access to the same infrastructure. This clustering of businesses essentially pursues the goal of fulfilling specific environmental, economic, and social objectives.

The inspiring publication by Frosch & Gallopoulos (1989) [2] drew attention to the benefits of more integrated industrial activities, in which energy and material use are optimized and waste generation is minimized. The model was named industrial ecology due to its analogy to biological ecosystems. The discipline of industrial ecology (IE) attempts to resemble the closed-circuit cycling of materials and energy flows in natural ecosystems, in which the by-products of some processes are used as resources for others (Ehrenfeld & Gertler, 1997 [3]).

In the first textbook in the field of IE (Graedel & Allenby, 1995 [4]), the ultimate goal of IE is stated to be an evolution of manufacturing in a manner that all wastes are recycled. In this sense, IE can be seen as an approach for analyzing the environmental impacts of goods and services. It demands innovative technologies and processes to reduce environmental damage while also reducing production costs. Approaches such as material flow analysis, environmental design, energy technology assessment, and EIPs are within the scope of industrial ecology.

As a result of more strict environmental laws, public awareness of environmental issues, and technological progress in increased efficiency and recycling at the end of the 20th century, new concepts such as industrial symbiosis were introduced.

Industrial symbiosis is considered one aspect of industrial ecology, which combines environmental improvement with economic progress and promotes local regeneration through the development of EIPs (Gibbs, 2008 [5]). The definition of industrial symbiosis, most widely used in the academic literature, is provided by Chertow (2000) [6] as follows: “The part of industrial ecology known as industrial symbiosis engages traditionally separate industries in a collective approach to competitive advantage involving physical exchange of materials, energy, water and/or by-products. The keys to industrial symbiosis are collaboration and the synergistic possibilities offered by geographic proximity”.

The phrases of this definition are redefined based on practical experience in the field to provide a clear communication basis with industrial symbiosis practitioners, stakeholders, and policymakers to enhance its innovative and transformative potential toward a more sustainable economy (Lombardi & Laybourn, 2012 [7]) as follows: “Industrial symbiosis engages diverse organizations in a network to foster eco-innovation and long-term culture change. Creating and sharing knowledge through the network yields mutually profitable transactions for novel sourcing of required inputs, value-added destinations for non-product outputs, and improved business and technical processes”. They argue that using the term “organization”, the role of research and government can be captured alongside industry. The term “network” emphasizes the mutual learning and information sharing of the members. The “physical exchange of resources” is replaced by the term “eco-innovation” as the result of industrial symbiosis, affiliated with actions towards environmental benefits accompanied by competitiveness and economic growth. The most critical part of the redefinition opposes the focus on physical exchanges and geographical proximity, reasoning that this focus is counter-productive to the advancement of industrial symbiosis and undermines its ability to promote innovation, economic growth, and regeneration.

EIPs are the actual realization of the concept of industrial symbiosis. The definition that is most broadly cited for EIPs by many international organizations is based on the initial definition created by the Indigo development team in 1992 and later expanded and refined as follows: “An eco-industrial park or estate is a community of manufacturing and service businesses located together on a common property. Member businesses seek enhanced environmental, economic, and social performance through collaboration in managing environmental and resource issues. By working together, the community of businesses seeks a collective benefit that is greater than the sum of individual benefits each company would realize by only optimizing its individual performance” (Lowe et al., 1996 [8]). “The goal of an EIP is to improve the economic performance of the participating companies while minimizing their environmental impacts. Components of this approach include green design of park infrastructure and plants (new or retrofitted); cleaner production, pollution prevention; energy efficiency; and inter-company partnering. An EIP also seeks benefits for neighboring communities to assure that the net impact of its development is positive” (Lowe, 2001 [9]).

The EIP opportunity can turn into a disadvantage; while the geographic concentration of industries and businesses can provide the potential for better collective environmental performance due to inter-firm cooperation, when not executed successfully, environmental damage and pollution can exceed the local ecological capacity. Tudor et al. (2007) [10] and Heeres et al. (2004) [11] discussed the drivers and barriers to the development of EIPs.

The most famous examples of EIPs worldwide are Kalundborg (Denmark), Mantford Boys’ Town, Suva (Fiji), Intervale Food Center, Burlington (Vermont), Eco-industrial Park of Devens (Devens, MA, USA), and Burnside Park (Oatlands, NSW, Canada).

3.2. Eco-Industrial Parks

The first successful implementation of an eco-industrial park was in Kalundborg in Denmark, and its success has since been regarded as an inspiration for other industrial parks. The development of Kalundborg EIP evolved over the course of two decades, leaving EIP researchers with the question of whether it is possible to apply the lessons learned to the planning phase of EIPs. Another good example of an industrial recycling network can be found in the Austrian province of Styria, through which tons of material are locally recycled every year in addition to supplying district heating.

The EIP examples in the USA and Europe are collected by Gibbs (2008) [5]. The main goal is to investigate to what extent these developments comply with the key features of IE, i.e., firms in geographical proximity that interchange material and energy flow through networks. In this listing, there are 26 EIPs in Europe, out of which 10 are in the UK. The investigation concludes that progress in the social science dimensions of the EIP development leads to successful EIP operation.

The strategy for planning and developing EIPs in the USA was initiated by the President’s Council of Sustainable Development of the federal government in 1994, defining and promoting four demonstration projects (Heeres et al., 2004 [11]; Côté et al., 1998 [12]) as follows:

• Fairfield (Baltimore, Maryland), which mainly contains petroleum and organic chemical companies. The Baltimore Development Corporation developed the EIP concept with the assistance of Cornell University.
• The Brownsville Regional Industrial Symbiosis Project (Brownsville, TX, USA) is pursuing the idea of a “virtual EIP” that links industries that are not necessarily located in the same area.
• Cape Charles Sustainable Technologies Industrial Park (Cape Charles, VA, USA).
• Chattanooga (TN, USA) has several defined goals, such as developing multi-purpose facilities, affordable housing, and a clean public transport system based on electric buses.

Six selected EIP projects, three of which are in the Netherlands and the other three in the USA, are compared by Heeres et al. (2004) [11]. At the time of publishing, all studied projects were in an early development phase without real implementation; thus, no solid conclusion could be drawn based on facts and numbers about their success. However, the paper proposed six aspects for the systematic scoring of the project to assess their success. Furthermore, important processes and physical factors found in the literature were compared to reveal the most significant differences between the projects in the two countries as follows:

• Project objectives (for the Netherlands, the economic and environmental aspects are valued equally, while in the USA, the prime objective is creating local jobs, and economic aspects are weighted more than environmental aspects).
• Initiators and finance (in the Netherlands, the initiators are local entrepreneur/employer associations in close cooperation with the government who also participate financially in the development process, whereas in the USA, the local industry is more passive, and the local government is the main initiator of such projects).
• Public participation (local non-governmental organizations and local communities are actively involved in developing the USA EIP projects. In the Netherlands, the companies and direct stakeholders carry out the EIP development process).

The EIP approach caught the attention of Chinese environmental policymakers first in 1997 through the publication of Industry and Environment in Chinese by the United Nations Environment Program (Shi et al., 2010 [13]). As a result, two industrial park-related demonstration programs were initiated in 1999 and 2000, respectively, and until 2010, 60 industrial parks were approved as National Trial EIPs (Zhang et al., 2010 [14]). Due to the large number of industrial parks in China and their diversity in character and region, the State Environmental Protection Administration (SEPA) developed comprehensive guidelines for describing and evaluating EIP projects, which shaped the first national standard to guide EIPs in the world. The work by Geng et al. (2009) [15] introduced and examined this standard, described its calculation methods, and analyzed the benefits and challenges of this regulation in the framework of developing EIPs in China.

South Korea, with over 1000 industrial complexes, initiated a three-stage, 15-year national EIP program in 2005. Park et al. (2016) [16] introduced the Korean approach to developing EIPs based on the SWOT method (strengths, weaknesses, opportunities, and threats) in Korean industrial complexes and analyzed the success and barriers of this approach based on the experience of the first phase of the plan at five pilot sites. The results show economic benefits 10 to 100 times the government funding together with a 0.1% reduction in greenhouse gas emissions from industrial activities in Korea in 2011 and 0.48% in 2004 (Ban et al., 2016 [17]). Gao et al. (2008) [18] applied the principle of energy cascading to an existing combined cooling, heating, and power system in the Sintai Optical Industrial Park of China by reconstructing the waste utilization and adding elements such as liquid dehumidification and biogas generation. The comparison was made based on both the first and second laws of thermodynamics by comparing the energy utilization factor and exergy efficiency of the systems with 20 and 4.55% increases, respectively, in addition to a 48% reduction in CO₂ emissions.

The same principles have also been applied to the Yeosu petrochemical complex in South Korea. Kim et al. (2010) [19] presented detailed formulations for several industrial site components together with useful variables and parameters used in the mathematical formulation of the optimization problem. The MILP optimization problem finds the optimal utility network, considering four different seasonal utility requirements. The base case and the optimized utility network were analyzed and compared thoroughly in this study from economic and environmental viewpoints. Moreover, sensitivity analyses were carried out to investigate the influence of variations in fuel, water, and electricity prices.

A mathematical programming approach based on a superstructure to find the optimum intra- and inter-plant HI scenario has been developed by Hipólito-Valencia et al. (2014) [20]. The paper considers the use of low-grade waste heat as the source for ORC to reduce external electricity demand. The problem formulation allows configurations with ORCs in each plant and/or a shared ORC for the total site. The objective is to find a solution with minimum annual costs. The methodology is applied to two examples with two and three plants, respectively. Different scenarios are defined for the optimization problem to include/exclude inter-plant energy integration and ORCs, imposing different constraints in the formulations. The results are compared in terms of total annual costs per year for different scenarios, mentioning that the influence of configurations, construction complexity, and associated costs have been neglected.

Industrial symbiosis is put into practice by material and energy networks. IE investigates the material and energy flows in industrial activities and tries to find solutions to reduce the undesirable environmental effects while considering the economic, social, organizational, and political aspects. In this regard, the state of research is reviewed in the next section with a focus on material, water, and energy exchange, specifically in industrial parks.

4. Integration Options Within Eco-Industrial Parks

4.1. Heat Integration in Eco-Industrial Parks

HI increases the overall efficiency of chemical processes and industrial plants and contributes significantly to their economic operation. It is considered the primary development in the field of process integration in an attempt to make industrial plants and processes more efficient. It can be deduced from the literature that increasing the overall energy efficiency of chemical and industrial plants by recovering heat between their parts and reusing waste heat is a common practice and is being successfully executed. Industrial parks as clusters of (usually large) chemical, petrochemical, and industrial units offer significant energy-saving potential for TSHI that is yet to be exploited.

TSHI has received growing interest since its inception in the 1990s (Dhole & Linnhoff, 1993 [21]; Klemeš et al., 1997 [22]). Matsuda et al. (2009) [23] applied the combination of the so-called R-curve analysis and the total site sink/source profile approach to one of the largest chemical complexes in Japan to identify theoretical site-wide energy-saving potential. Based on the theoretical results, the study makes practical improvement suggestions, some of which have been partly implemented. The R-curve is the illustration of maximum available efficiency for different power-to-heat ratios in the site, which takes different shapes for different power and heat production technologies.

In the study by Chae et al. (2010) [24], a framework was proposed for optimizing industrial waste heat utilization, including the waste heat network within and beyond an EIP. The methodology was applied to the Yeosu National Petrochemical complex in South Korea, which was considered together with the heat demand of the neighboring city. The heat demand was calculated roughly based on the average temperature and the population of the city. The optimization problem was formulated as an MILP and was solved with three different network objectives, i.e., to minimize (a) total cost, (b) extra fuel consumption, and (c) future expansion.

For industrial sites with low pinch temperatures, including non- and semi-continuous processes, using heat recovery loops for indirect heat transfer between processes is a good solution for total site integration. This finding was obtained by Atkins et al. (2011) [25]. Incorporating thermal storage in the HI system is investigated to handle the variabilities in operation. Heat integration within the industrial parks has been reported by Ji et al. (2022) [26]. The authors considered centralized and distributed waste heat recovery systems.

The TSHI problem has been considered a multi-agent problem by Cheng et al. (2014) [27], and they proposed the application of the mathematical tools of game theory to synthesize a cost-optimal HEN, which satisfies the participating agents, i.e., the involved parties.

A methodology to find the optimal combination of direct and indirect heat exchange to improve the HI in total sites has been developed by (Wang et al., 2015 [28]). A step-by-step method was proposed using a superstructure model combined with heuristic rules.

The research by Zhang et al. (2016a, 2016b) [29,30] focused on the low-grade heat in industrial parks, which is estimated to account for 20% to 50% of the energy used by the process industry according to Luong (2013) [31]. The authors suggest the utilization of low-grade waste heat as an energy source for absorption chillers to provide a part of the required cooling in the industrial park. The suggested method decomposes the large-scale modeling problem into three parts: the absorption refrigeration system, the refrigeration stations, and the pipe networks in the form of an MINLP. The formulation is described thoroughly, showing the simplifications made in mathematical modeling. Zhang et al. (2016) [30] argued that multi-objective optimization of EIPs has not yet been given enough attention in the literature, nor has the configuration and the heat loss of the waste heat transportation network. The presented case study of the industrial park in Jurong Island, Singapore, shows the significant influence of intermittent (as opposed to continuous) waste heat profiles on the optimization results. Temperature drops through the transportation network are considered in mathematical modeling to eliminate infeasible connections. The paper describes two common methods for solving a multi-objective optimization problem: scalarization and the Pareto approach. Chew et al. (2013) [32] classified and declared the challenges to a successful TSHI methodology, such as design, operation, reliability and availability, policy, and economic-related issues. Two examples in the study highlight the impact of these issues on the development of TSHI designs that are realistic and practically implementable.

The suggested methodology aims to create a stepwise roadmap to successful HI in industrial clusters, starting with simple heat exchange between a few participants, which can be expanded to more complex site-wide systems. However, one essential element of the methodology relies on the expert’s knowledge and experience: discovering the main issues, categorizing the heat exchangers, and prioritizing the measures. The uncertainty in the price of steam and investment costs has been investigated and illustrated through sensitivity analyses by variations of ±30%.

Liew et al. (2014) [33] built on the time slice (TS) methodology based on the work of Varbanov & Klemeš (2011) [34] to address and handle the fluctuating nature of renewable energy sources and the urban energy demand in the TSHI problem. The table-based algorithm develops a Total Site Heat Storage Cascade to calculate the recoverable heat to be stored or let down in different time slices of the daily plant start-up and operation.

Hiete et al. (2012) [35] approached the subject of inter-company HI from two perspectives. In the first part of this paper, the conventional pinch method is expanded, considering the cost of piping regarding the relative distances between the plant companies. Furthermore, the necessary backup utilities and their associated costs are considered to account for demand/supply fluctuations and operation interruptions. These extensions change the results compared to the conventional pinch analysis and allow for more realistic and implementable solutions. In the second part of the article, the critical issue of allocating costs and savings for establishing long-term cooperation between participants in the industrial symbiosis is taken into consideration. Different methods of cooperative game theory were briefly introduced and applied to a case study. It was then discussed that these methods regard partners of the exchanges as equally important and do not consider the different prices of energy streams. The cost/benefit allocation calculations were repeated considering the actual energy prices and were compared to those of the game theory approach.

A review of the methodologies on energy consumption targeting in the context of TSHI can be found in Liew et al. (2014) [33]. This paper provides insights for future research in the framework of integrating industrial, urban, and renewable energies in locally integrated energy systems.

4.2. Heat and Work Integration in Eco-Industrial Parks

The simultaneous integration of work and heat can increase energy efficiency and reduce the environmental impact of industrial symbiotic systems. Therefore, Work Exchange Networks (WENs) or Work and Heat Exchange Networks (WHENs) are new fields of research and practice. These networks involve, in addition to the exchange of heat at different temperature levels, the integration of work through compressors, turbines, and valves at different pressure levels.

A methodology that combines pinch and exergy analysis, specifically applied to sub-ambient processes, has been developed by Aspelund et al. (2007) [36]. The main idea is to expand the pinch analysis, which only deals with the temperatures, to include pressure as a design variable. This methodology involves not only heat exchangers but also expanders and compressors and, therefore, is more complex than the conventional pinch analysis. Consequently, solving the problem relies on a set of heuristics and the know-how of the designer. The same research was published by Gundersen et al. (2009) [37], who introduced the idea of the appropriate placement of compressors and expanders regarding the pinch in a heat recovery system.

An MINLP model that combines pinch analysis and exergy analysis as a design tool to minimize the irreversibility through varying pressure levels of the process streams has been formulated by Wechsung et al. (2011) [38]. Their model involves a pinch operator to calculate the minimum required utilities and to locate the pinch, a pressure operator that involves state equations to calculate properties for streams undergoing isentropic changes at different pressure levels, and an exergy operator to calculate the exergy of streams and the exergy efficiencies. The objective function builds on these modules to find a solution that minimizes the irreversibility or the costs. The research by Razib et al. (2012) [39] reports an MINLP model that facilitates preliminary decisions on the feasibility of a WEN, including only single-shaft turbine–compressor combinations. This work has been extended by Onishi et al. (2014) [40], who proposed a superstructure MINLP formulation for the integration of a “heat exchange network between work”, i.e., the interstage cooling (for compressors) and/or interstage superheating (for turbines).

The optimal integration of compressors and expanders in above-ambient or sub-ambient HENs or cryogenic applications has been investigated by Fu & Gundersen (2015, 2016) [41,42].

The latest advancements in HENs for industrial parks, associated with the implementation of the Organic Rankine cycle and absorption refrigeration cycle, can efficiently convert waste heat to other forms of energy (heat, cooling, and electricity) in industrial parks.

4.3. Water and Material Stream Integration in Eco-Industrial Parks

Water integration networks were designed for individual companies to conserve water and reduce costs. The early approaches for this purpose were based on graphical methodologies (also insight-based techniques) as an extension of pinch analysis for HI (Wang & Smith, 1994 [43]; Dhole et al., 1996 [44]; Hallale, 2002 [45]; El-Halwagi et al., 2003 [46]; Manan et al., 2004 [47]; Prakash & Shenoy, 2005 [48]). In order to overcome some of the limitations of the graphical techniques, mathematical programming and optimization approaches were developed. Takama et al. (1980) [49] used mathematical programming to build a superstructure of all operations using water in a refinery as a multi-contaminant system and carried out optimization to remove the technically and economically undesirable connections. Huang et al. (1999) [50] and Alva-Argáez et al. (1999) [51] also proposed alternative mathematical programming models to solve multi-contaminant Water/Wastewater Allocation Planning (WAP) problems. Bagajewicz & Valtinson (2014) [52] presented LP/MILP formulations for designing water networks in process plants, considering forbidden and compulsory water flows between processes involving a single contaminant.

In the framework of EIPs, the minimization of fresh water and wastewater by reusing/recycling water through inter-plant water networks has been a research area during the last few decades. The EIP networking demonstrates additional complications compared to the developed tools and methods for the synthesis of water integration networks mentioned above. A prerequisite for the success of an EIP is to achieve collective benefits greater than those of stand-alone units. Therefore, new formulations are required for the optimal design and operation of water integration networks in EIPs. Lovelady & El-Halwagi, 2009 [53] developed an optimization-based approach based on recycling, reuse, and separation to manage wastewater between several plants.

5. Challenges in Designing HENs for Industrial Parks

Challenges in designing HENs for industrial parks can be divided into two large groups: (a) supporting the flexibility in the operation of individual plants and (b) increasing the length of pipes with associated pumping capacities that lead to an increase in capital cost and capital and operating expenses.

The game theory approach as a decision-making tool to address the inter-plant water reuse/recycle problem has been applied by Chew et al. (2009) [54]. The main concern here has been the consideration of maximum economic profitability from the point of view of the participating companies to compare the water consumption for different inter-plant water integration (IPWI) schemes. A systematic design procedure based on game theory has been developed to address the challenge of distributing cost savings fairly among involved parties in industrial parks, as reported by Tian & Li (2023) [55].

Aviso et al. (2010) [56] formulated the problem of water exchange in EIPs as a bi-level fuzzy optimization problem. This approach acknowledges the conflicting interests of the EIP leaders and the participating companies. Therefore, the optimization problem was formulated as an iterative process in which the EIP authority was defined as an upper-level decision-maker who specifies a tolerance range for its objectives and variables for minimizing freshwater consumption. The participating companies, defined as lower-level decision-makers, can optimize their objective function, which is commonly minimizing their annual levelized costs within the defined tolerance range. Boix et al. (2012) [57] formulated the mono-contaminant water network design problem in MILP form and solved a multi-objective optimization problem to gain a Pareto front instead of a single optimal solution. The defined objective functions to be minimized are the entering freshwater flow rate to the plant and the water flow rate at the inlet of regeneration units while imposing a limit on the number of connections. A further decision-making strategy based on the estimated associated costs allows for choosing the best solution among the Pareto fronts. As an alternative to the multi-objective optimization method, Ramos et al., 2016 [58] extended the approach by Lou et al. (2004) [59] and Aviso et al. (2010) [56] by formulating the problem as a multi-leader–follower game optimization to achieve a Nash equilibrium in which the participating companies make simultaneous optimal decisions given the strategies chosen by other participants. Rubio-Castro et al. (2012) [60] presented the application of an MINLP programming model to two examples for the investigation of optimal retrofit of existing in-plant and possible inter-plant water networks for the integration of the existing water sources and sinks in industrial zones. Alnouri et al. (2014) [61] focused on the pipeline merging options to improve the performance of the inter-plant water network and decrease the design complexity. Two pipeline branching schemes were applied to a case study and compared in terms of network costs. The choice of branching schemes was discussed according to the position of fresh water and wastewater in the plant.

Unlike the water (and energy) networks, material and waste networks are by nature more individual or specific and, therefore, less generalizable. Grant et al. (2010) [62] and Chertow (2000) [6] reviewed the tools for by-product exchange in industrial networks. The latter classified these tools as (a) input–output matching, (b) stakeholder processes, and (c) material budgeting. Raabe et al. (2017) [63] argued that a major barrier to the success of industrial symbiosis is the lack of information on available by-product exchange. To address this, the paper introduces a collaboration platform for planning industrial symbiosis, illustrated with a case study on food waste in Singapore. The proposed tool enables different entities to analyze the economic viability of potential symbiotic exchanges.

Various studies worldwide have also addressed the issue of industrial solid waste (ISW) management. Lü et al. (2012) [64] proposed a concept for the flow analysis of ISW and introduced a methodology for sustainable waste management. Two case studies, considered representative of many EIPs in China and Asia, illustrate the application of the proposed methodology. The correlation between ISW and industrial output is discussed, and conclusions are drawn for ISW management at enterprise, EIP, and regional scales. Maillé & Frayret (2016) [65] proposed a multicriteria optimization model that allows decision-makers to conduct sensitivity analyses on factors like waste selling prices and equipment processing capacities. Using data from the Kalundborg EIP, the model demonstrated potential savings and the optimal timing for initiating synergies. The findings revealed that companies typically prioritize economic considerations, with a payback period of around three years. The authors point out that minor financial adjustments could enhance resource preservation efforts. Taskhiri et al. (2015) [66] introduced a method based on fuzzy mixed-integer linear programming to optimize waste-to-energy networks (waste incineration). The maximum energy recovery and minimum payback period were integrated into one objective function, which the authors consider as representative of the satisfaction level of the stakeholders.

6. Conclusions

The review has been conducted to help provide context, identify the gaps in existing research, and emphasize the new developments and perspectives. Chosen keywords have been used to gain statistical insight into the existing research. Moreover, network maps are used to visualize the research gap by mapping out the relationships and frequencies between key concepts and terms. The shown network maps reveal areas where connections are sparse or non-existent. This tool helps effectively visualize the underexplored topics addressed and emphasizes the contributions of this work. A classical literature review has been performed, which offers a detailed overview of the conducted research within the field, identifying the key topics that have been extensively studied and those that have remained less attended. By summarizing the content of various studies, this review provides a comprehensive understanding of the existing research landscape.

In conclusion, advancements in heat exchanger technology and optimization-based frameworks have contributed to improved energy efficiency (by enabling the recovery and utilization of surplus heat generated by different manufacturing and processing plants) and reduced overall costs (by applying an optimization-based framework for simultaneous integration of heat exchanger networks, utility, and waste heat recovery cycle systems) in industrial park applications. The environmental impacts of heat exchanger networks in industrial parks are significant, offering energy-saving opportunities and contributing to the reduction in waste and energy consumption among a cluster of industries.

Future activities in developing efficient HENs for industrial parks should focus on investigating design, part load, and dynamic analysis uncertainties.