Disassembly Graph Generation and Sequence Planning Based on 3D Models for the Disassembly of Electric Motors

Disassembly is an important first step for new and innovative end-of-life solutions for critical components and materials on the way to a circular economy. Generating possible disassembly sequences and finding optimal sequences for a given disassembly target is a major challenge in disassembly. Using electric traction motors as an example, this paper presents an approach to automatically generate possible disassembly sequences and calculate the optimal path to a given disassembly state. This is based on geometric data in the form of polygon meshes enriched with additional information about non-geometric joints (e.g. adhesive joints). Geometric joints are automatically detected by analysing the contacts between parts. All parts, joints and resulting constraints are modelled in an undirected graph. By analysing the joints and movement constraints, a directed graph is generated. This graph contains all possible disassembly sequences, which can then be searched for an optimal sequence.