Research On Seru Production Joint Optimization Problem Based On Learning Effect Of Multi-Skilled Workers

With the changes in the international market environment,the manufacturing industry needs to have the ability to respond quickly to provide a variety of products,personalized,traditional assembly line production is difficult to adapt to the complex and changeable market demand.Seru Production was first born in Japan’s manufacturing industry,which has overcome the problem of insufficient flexibility in assembly line production,suitable for multi-batch,small-batch production needs.Seru production not only continues the high efficiency of assembly line production,but also obtains better flexibility,improves the utilization of production space through the flexible construction of Seru.Seru system operation mainly involves two key decisions: Seru construction and Seru scheduling,making a separate decision on one of them can only obtain a locally optimal solution.In order to adapt to the actual needs and obtain a global solution.This thesis studies the joint optimization problem of Seru production based on the learning effect of multi-skilled workers,and the main contents are as follows:(1)Taking the pure Seru production system as the research object.The multi-skilled worker are graded,the learning effect curve model for multi-skilled workers suitable for Seru production is determined.The structure and scheduling of Seru were jointly decided,with the optimization goal of minimizing the maximum completion time and minimizing labor cost,a mathematical model of Seru production joint optimization based on the learning effect of multi-skilled workers is established.(2)Aiming at the multi-objective optimization problem proposed in the thesis,an improved Artificial Bee Colony(ABC)algorithm is proposed to solve the problem.Including the following strategies: In the employment bee stage,it is divided into two parallel bee colonies for search,which effectively expands the search scope;the following bee stage to search in its three neighborhoods,and obtains a better local search effect;The reconnaissance bee stage avoids the population from falling into local optimization through the setting of the number of honey sources being discarded and the effective cross-variation method,so that the improved ABC algorithm has a good global optimization ability.(3)To investigate the impact of the number of constructed Seru and the allocation of batches on the completion time,this thesis conducts experimental analysis.The results show that to minimize the completion time,it is necessary to construct as many Seru as possible and obtain the most balanced batch allocation relationships as possible.To test the efficiency of the proposed algorithm in this thesis,the Fast Non-dominated Sorting Genetic AlgorithmsⅡ(NSGA-Ⅱ)is designed for comparative experiments.The results show that the improved ABC algorithm,solution efficiency designed in this thesis is higher,the target value of the solution set is wider,the C matrix index is generally dominant,and the solution effect is better for the problems proposed in this thesis.