The Study On Soft Computing Methods For Assembly Sequence Planning

The soft computing method for assembly sequence planning (ASP) problems has the characteristics of low cost, simple operation and high intelligence, thus it draws many researchers and technologists'attention in recent years. The author carries on a thorough investigation in soft computing methods for ASP and establishes more effective ASP methods on the basis of analyzing and summarizing the relevant researches on ASP, especially the application of soft computing methods in ASP problems. The work done in this paper includes:(1) How to establish the interference matrix and analyze the geometrical feasibility of assembly sequence and how to deduce the set of feasible assembly directions of parts in the assembly process are explained. The relevant theory analysis of the assembled part stability is expatiated, too. Finally, the author discusses the stability of assembly sequence from the viewpoint of assembly and disassembly, respectively.(2)The principles, processes and basic problems of ant colony algorithm are anatomized. Furthermore, the author analyzes the merits and weakness of ant colony algorithm applied in ASP and an idea of adding part stability analysis to ant colony algorithm is put forward aiming at amending its weakness. The improved ant colony algorithm for assembly sequence planning problems was tested by illustrative cases, and the results show good performance.(3)The advantages and disadvantages of genetic algorithm for assembly sequence planning are analyzed. To develop an efficient method for ASP, the idea of combining genetic algorithm and ant colony algorithm to a hybrid algorthm is put forward, which aims at improving the efficiency of problem solving. The experimental results show that the hybrid algorithm reveals good performance.(4)The principles, processes and basic problems of immune algorithm are expatiated and an immune algorithm imitating the biological immune system is proposed to solve ASP problem. Tested by many engineering applicable examples, the immune algorithm outperforms other algorithms in handling assembly sequence planning problems.