.png){kind=icon}
.png){kind=icon}
Return
A failure-driven and knowledge-enhanced self-planning method for disassembly process reconstruction
.png){kind=icon}
PRE
AI
DOI：10.1080/00207543.2025.2575842.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Disassembly plays a crucial role in the recycling and remanufacturing of retired electromechanical products. However, the various structural damages of subassemblies sometimes results in disassembly failures, such as fracture, wear, and corrosion. Uncertain disassembly failures lead to complex process reconstruction, which involves updating information, reconfiguring elements, and replanning sequences. Thus, this study proposes a failure-driven and knowledge-enhanced self-planning method for disassembly process reconstruction. First, a failure-based disassembly knowledge graph is constructed, which integrates various types of disassembly failure knowledge and supports disassembly information updates. Then, the multi-dimensional disassembly elements are reconfigured through rule-based reasoning and logical reasoning, on the basis of which three reconstruction strategies are proposed, and the disassembly sequences under disassembly failures are rapidly replanned by a reconstruction strategy selection-based genetic algorithm. Finally, a hybrid Li-ion battery pack of Audi A3 Sportback e-tron is selected as the case study and applied to test the proposed self-planning method. Experimental results demonstrate the method's effectiveness in reconstructing disassembly processes under various failure types and degrees, limiting profit reduction to below 7%. The hybrid strategy significantly outperforms single strategies in managing multiple failures, and shows superior performance over PSO and ABC algorithms in solving complex disassembly planning problems.
Keywords:
Retired electromechanical products
disassembly failure
knowledge-enhanced
disassembly process reconstruction
rule-based reasoning
Journal
IF:
7.3
Papers:
1.1W
Citations:
3.7W
View more organizations