Abstract: In order to address the problem that the variation of constraint leads to variation of assembly sequence, a method based on the grey theory associated with genetic algorithm (GA) is presented. The variation of constraint on assembly site of complex products includes absence of parts, fixtures, resource and so on. A seven-cell coding format is built based on complex genes. The fitness function is built taking the constraint relation into account, and synthesizing the parallelism, stabilization, repeated directional and aggregation of the sub-assembly. The alternative set of assembly sequence is generated from the variation node after implementing GA operation. The grey associated model is built considering the alternative set of assembly sequence as the grey decision set. The reference sequence is generated from original assembly sequence. By analyzing the grey associated value, an optimal or near-optimal assembly sequence is created. This method is tested on a practical case and the result shows its feasibility and superiority.

Key words: Assembly sequence, Genetic algorithms, Grey theory, Process planning