Research On Optimization Of Multi-Manned Assembly Line Balancing Problem

Assembly line is a way of production organization which is widely used in industry.Assembly line balancing (ALB) helps to improve efficiency, save production costs andenhance the competitiveness of enterprises. With the changing economic environment inChina, labor cost rises dramatically these years, which forces many labor-intensiveenterprises to pay more attention to the control of their labor costs. As the need ofassembly line optimization gradually spreads from technology-intensive industries tolabor-intensive ones, multi-manned assembly line balancing problem emerges.This paper first briefly reviews the development of assembly line balancing problemand after analyzing existing line balancing theories points out that previous studies oftraditional assembly line balancing cannot meet the demand of labor-intensive industrieswhose major production cost comes from labor costs. As a result, the study ofmulti-manned assembly line balancing problem is of high value nowadays. In this paper,heuristic-mixed genetic algorithms (hGA) are proposed to solve two kinds ofmulti-manned line balancing problem. A heuristic rule is integrated during the decodingprocedure of genetic algorithm so that the optimization of line balancing and workerallocation is achieved simultaneously. Compared with the results obtained by solvingmathematic model directly, later numerical experiment results show the promisingperformance of hGA.Task time tends to be affected by random factors in manual production environment.To solve this problem, asynchronous transfer line (ATL) is introduced. ATL is modeledbased on queuing theory and its related characteristics are described accordingly. On thisbasis, design steps and multi-objective optimization algorithm of ATL are presented. Theeffectiveness of these proposed methods are proved by the following calculation andsimulation.Finally, a case study is carried out based on data from an interphone producer. Taskdistribution and worker allocation plan is obtained concurrently. Optimization result showsthat the algorithms proposed in this paper solve multi-manned assembly line problemeffectively and efficiently.