Research On Modeling For Two-sided Disassembly Line Balancing Problem Under Tool Constraints And Pareto Cat Swarm Optimization Algorithm

With the exhaustion of natural resources and the grabing for source increasingly,the shortage of resources has become the focus of attention in the world today.Recycling and remanufacturing have received extensive attention for their good economic and environmental benefits,the remanufacture of waste products can't be separated from disassembly,and disassembly line was the main way of disassembly in a large scale,but disassembly line balancing problem restrictived the development of disassembly line.This article studied the disassembly line balancing problem for remanufacture.In view of the insufficient that two-sided disassembly line balancing problem has not been reported,but two-sided disassembly line has been applied to industrial production widely,this paper analyzed the characteristics of two-sided disassembly line and the influence mechanism of various factorson it.Aiming at the shortage that the tool has been ignored,the influence of the tool factors on the disassembly operation was analyzed and the multi-objective evaluation model of two-sided disassembly line balancing problem under tool constraints included the number of workstations,dismantling efficiency,balance index and tool replacement index was bulit from the perspective of operating time,disassembly efficiency and the line balancing.In this paper,the decoding was divided into on-line decoding and in-station decoding to optimize the results of decoding and simplify decoding process.In the on-line decoding,the minimum number of stations and the sub-prioritized minimum station idle time were used to determine the specific workstation assigned to them.In the in-station decoding,the task disassembly time axis was analyzed to find out the feasible starting disassembly time of the current task under the priority relationship and disassembly situation.An improved cat swarm optimization algorithm was proposed to solve the multi-objective disassembly line balancing model.According to the characteristics of the disassembly line balancing problem,a searching model based on random number and random perturbation was proposed.The cross operation and the mutation operation of the genetic algorithm were introduced in the tracking model of the cat swarm optimization algorithm to guide the proposed algorithm to approach the global optimum and to avoid the prematurity of the traditional cat swarm optimization algorithm effectively.Combining with the multi-objective characteristics of the disassembly line balancing problem,the Pareto was introduced into the proposed algorithm to avoid the interference of human factors to all targets.The elite strategy was adoptd by replacing the population with the same number of individuals in the external file to converge faster.The operating efficiency of the proposed algorithm was improved by the adoption of crowded distance to external file.The effectiveness and high efficiency of the proposed algorithm were verified by testing existing examples and comparing with other algorithms in domination,average value and C index.The result verifies the effectiveness of the genetic manipulation by comparing with the cat swarm optimization algorithm.Final,the proposed model and algorithm were adopted to design the disassembly line,and the result provided good reference for the policymakers.