TY - JOUR
T1 - Data-Driven Approach for Decision-Making in Reactive Disassembly Planning to Enable Case-Based Reasoning
AU - Streibel, Lasse
AU - Jordan, Patrick
AU - Zaeh, Michael F.
N1 - Publisher Copyright:
© 2024 The Authors.
PY - 2024
Y1 - 2024
N2 - Environmental issues require sustainable manufacturing, which includes improving the circularity of products. The inefficiency of product disassembly is a crucial barrier to industrial-scale circularity. One cause of inefficiency is using deterministic disassembly plans despite the uncertainties of end-of-life products, such as their varying conditions. Instead, disassembly requires the online adaptation of these plans whenever new information deviates from prior assumptions. This is known as reactive disassembly planning. Reactive disassembly planning faces individual and dynamic planning problems that depend on the state of the product, the disassembly process, and the resources in the disassembly system. Decisions in reactive disassembly planning affect the behavior and output of the disassembly system, which influences the profitability and sustainability of the disassembly. Due to this complexity, decision-making in reactive disassembly planning should be supported methodically. This contribution presents a data-driven approach supporting decision-making in reactive disassembly planning. It allows for the automatic adaptation of disassembly plans in response to plan deviations while considering the impact on economic and environmental metrics. Methods for solution space management, design of experiments, and material flow simulation are used to identify and select solution alternatives for unknown deviations. Selected solution alternatives are linked with their corresponding deviations and stored as cases in a case base. The case base enables the application of case-based reasoning, which utilizes the case base without having to identify and select solution alternatives. Therefore, it enables learning from experience and accelerates reactions to recurring plan deviations. In this contribution, the modules and the procedure of the approach are presented.
AB - Environmental issues require sustainable manufacturing, which includes improving the circularity of products. The inefficiency of product disassembly is a crucial barrier to industrial-scale circularity. One cause of inefficiency is using deterministic disassembly plans despite the uncertainties of end-of-life products, such as their varying conditions. Instead, disassembly requires the online adaptation of these plans whenever new information deviates from prior assumptions. This is known as reactive disassembly planning. Reactive disassembly planning faces individual and dynamic planning problems that depend on the state of the product, the disassembly process, and the resources in the disassembly system. Decisions in reactive disassembly planning affect the behavior and output of the disassembly system, which influences the profitability and sustainability of the disassembly. Due to this complexity, decision-making in reactive disassembly planning should be supported methodically. This contribution presents a data-driven approach supporting decision-making in reactive disassembly planning. It allows for the automatic adaptation of disassembly plans in response to plan deviations while considering the impact on economic and environmental metrics. Methods for solution space management, design of experiments, and material flow simulation are used to identify and select solution alternatives for unknown deviations. Selected solution alternatives are linked with their corresponding deviations and stored as cases in a case base. The case base enables the application of case-based reasoning, which utilizes the case base without having to identify and select solution alternatives. Therefore, it enables learning from experience and accelerates reactions to recurring plan deviations. In this contribution, the modules and the procedure of the approach are presented.
KW - case-based reasoning
KW - decision-making
KW - disassembly planning
KW - simulation
UR - http://www.scopus.com/inward/record.url?scp=85213004010&partnerID=8YFLogxK
U2 - 10.1016/j.procir.2024.10.215
DO - 10.1016/j.procir.2024.10.215
M3 - Conference article
AN - SCOPUS:85213004010
SN - 2405-8971
VL - 58
SP - 1117
EP - 1123
JO - IFAC Proceedings Volumes (IFAC-PapersOnline)
JF - IFAC Proceedings Volumes (IFAC-PapersOnline)
IS - 27
T2 - 18th IFAC Workshop on Time Delay Systems, TDS 2024
Y2 - 2 October 2023 through 5 October 2023
ER -