Exploring the System Dynamics of Industrial Symbiosis (IS) with Machine Learning (ML) Techniques—A Framework for a Hybrid-Approach

Administrative Sciences

2020

Self Cite

Industrial Symbiosis (IS) is a systematic and collective (business) approach to optimizing the use of materials and energy among cross-sectoral industries in order to initiate and exhaust extended cascading systems; it is associated with (synergistic) environmental, technical, social, and economic benefits. For monitoring and controlling the development and progress of an IS system, an indicator system must be set up to standardize and assess the IS (sustainability) performance. This study aims to present a quantitative indicator system to enable the tracking of set sustainability targets of an IS system in Industrial Parks (IPs) for goal-directed IS management. The presented guiding framework encourages IP members in IS systems to set sustainability objectives and to evaluate and track their performance over time with a quantitative indicator system. In particular, established and (partly) internationally standardized methods—such as Material Flow Analysis (MFA), Material Flow Cost Accounting (MFCA), Social Network Analysis (SNA), and Life Cycle Assessment (LCA)—are used in order to place the indicator system on a solid and robust foundation and to adequately meet the multi-faceted sustainability perspectives in the form of a combinatorial application for deriving suitable quantitative indicators for all three (environmental, economic, social) dimensions of sustainability. The indicator system, once embedded in an Information Technology (IT)-supported IS tool, contributes crucially to the technology-enabled environment of IS systems, driving sustainability trajectories.

Section: Discussion and Future Researchmentioning

confidence: 99%

Tracking Sustainability Targets with Quantitative Indicator Systems for Performance Measurement of Industrial Symbiosis in Industrial Parks

Lütje

Administrative Sciences

2020

Self Cite

“…An agent-based simulation was incorporated to present the share allocation of the total economic benefits resulting from IS activities (Yazan and Fraccascia 2019). Lütje et al (2019b) proposed a first framework concept for a combinatorial approach of Agent-Based Modelling (ABM) and the Artificial Intelligence (AI) technique of Reinforcement Learning (which belongs to the sub-field of Machine Learning (ML)) for exploring the system dynamics of IS. So, the agents can take note of rewards received by specific actions and record it, in order to adapt the model and accelerate the learning phase for better solution finding.…”

Section: Introductionmentioning

confidence: 99%

“…So, the agents can take note of rewards received by specific actions and record it, in order to adapt the model and accelerate the learning phase for better solution finding. This hybrid-approach opens up the simulation of scenarios with optimally utilized IS systems in terms of system adaptability and resilience (Lütje et al 2019b).…”

Section: Introductionmentioning

confidence: 99%

“…Typical task classes for expert systems that are to be used in particular in the application context of the IT-supported IS tools are: The innovative part of the web application is the module of simulation and modelling (Table 2, TK3.0) which shall use methods of system dynamics and artificial intelligence (AI) algorithms such as Artificial Neural Networks (ANN) and/or hybrid-approaches of ABM and Reinforcement Learning (a machine learning technique, sub-field of AI) (Lütje et al 2019b). So, the web application should be able to simulate and model various transformation pathways with two anchor points: from the actual state to the desired future scenario which is aligned to the defined goals of, for example, "zero waste", "zero emission" or "CO2-neutral" park, or the IP pursues to align its business performance to the science based targets (SBT) (SBTi 2019) 4 .…”

mentioning

confidence: 99%

See 1 more Smart Citation

Requirements Engineering for an Industrial Symbiosis Tool for Industrial Parks Covering System Analysis, Transformation Simulation and Goal Setting

Lütje

Administrative Sciences

2020

Self Cite

Industrial Symbiosis (IS) is a collaborative cross-sectoral approach to connect the resource supply and demand of various industries in order to optimize the resource use through exchange of materials, energy, water and human resources across different companies, while generating ecological, technical, social and economic benefits. One of the main goals of IS is the set-up of advanced circular/cascading systems, in which the energy and material flows are prolonged for multiple utilization within industrial systems in order to increase resource productivity and efficiency, while reducing the environmental load. Many Information Communication Technology (ICT) tools have been developed to facilitate IS, but they predominantly focus on the as-is analysis of the IS system, and do not consider the development of a common desired target vision or corresponding possible future scenarios as well as conceivable transformation paths from the actual to the defined (sustainability) target state. This gap shall be addressed in this paper, presenting the software requirements engineering results for a holistic IT-supported IS tool covering system analysis, transformation simulation and goal-setting. This new approach goes beyond system analysis and includes the use of expert systems, system dynamics and Artificial Intelligence (AI) techniques, which turn the IT-supported IS tool to be developed into a comprehensive and holistic instrument with which future scenarios and transformation paths can be simulated.

“…As inter-company information exchange is crucial to identify and implement IS opportunities [9,10,[17][18][19], the proposed IT-supported IS tool shall provide companies with easy-to-understand, structured and specific information as well as specific recommendations for actions regarding potential IS activities to business users. Within this project an IS expert system, which is to be integrated into the IT-supported IS tool [16] and a hybrid-approach of agent-based modelling (ABM) and Reinforcement learning (Machine Learning technique) for simulating the dynamics of IS systems [20] are being developed. An expert system is a computer program that derives recommendations for action from a knowledge/data base like a human expert [21].…”

Section: Introductionmentioning

confidence: 99%

Recurring Patterns and Blueprints of Industrial Symbioses as Structural Units for an IT Tool

Rohde-Lütje

2020

Sustainability

Self Cite

Industrial Symbiosis (IS) deals with the set-up of advanced circular/cascading systems, in which the energy and material flows are prolonged for multiple material and energetic (re-)utilization within industrial systems. To facilitate the technology-enabling environment of IS systems, this work deals with the identification of recurring patterns in IS systems of specific IS case studies and deduction of elementary blueprints and structural units, setting an initial cornerstone to pool and synthesize existing IS knowledge and to deploy this knowledge base in an Information Technology (IT)-supported IS tool, which would remarkably advance the scope of action and development of IS systems. An explorative cross-case analysis was conducted by investigating 80 IS case studies in depth in order to illuminate recurring (key) patterns in IS systems by generalizing and abstracting IS main structures, compositions, resource exchange activities and measures. It has been shown that similar IS sectoral partnerships and resource exchanges have recurrently formed in different regions and hence, generalizable patterns can be deduced. This study identified common IS compositions, sector clusters and key/core/anchor entities and synthesized a content basis for a database of an IS resource exchange catalog based on existing/available IS information, which can be used in an IT-supported IS tool. It contains information of specific IS resource exchanges, broken down by industrial sectors, differentiating providing and receiving sectors and which respective exchanged waste flows were processed into which secondary material/product. Once this fundamental information/data base is incorporated and applied in an IT-supported IS tool, it enables the facilitated recommendation of potential IS partners and IS actions to optimize existing IS cases or to initiate IS development. Especially, first IS germ cells of (key) entities can be derived and connected to each other considering individual circumstances and (geographical) business environments.