Research on Numerical Module Identification Approaches

Point to product module identification, a new numerical optimization is proposed in this paper. This approach combines the improved particle swarm optimization (IPSO) algorithm with the synthesis design structure matrix (DSM), in which a series of property correlations facing the product lifecycle are divided into functional, geometrical, physical and auxiliary ones to obtain the synthesis DSM. An optimization function for module identification, based on the axiomatic design theory, is established. The numerical optimization process is given by particular analysis. The optimization approach using IPSO is tested several times with an example and compared with other classical algorithms, such as PSO, simulated annealing algorithm (SAA) and genetic algorithm (GA). The result demonstrates the feasibility of the proposed approaches.