Abstract: In order to improve the efficiency of object selective disassembly sequence planning(OSDSP) for complex mechanical products, an approach based on bottom-up ideas is proposed, which combines the disassembly hybrid graph model(DHGM) with particle swarm optimization (PSO) algorithm. The DHGM is constructed to describe the constraint and disassembly priority relationships among constituting components of product. Furthermore, the disassemblability constraint expression is deduced from it. Based on PSO, both the mathematic description of OSDSP and the particle fitness (i.e., objective function) are given. Depth first search(DFS) is carried out to locate the selective component, then a new object-driven recursive reasoning method is designed to generate possible disassembly sequences which are used to initialize the particles. PSO is used to obtain the optimum disassembly sequence effectively. A full-automatic washer case study proves the validity and feasibility of the proposed method .

Key words: Disassembly hybrid graph, Disassembly sequence planning, Selective disassembly