Research On Production Scheduling Of Vulcanization Workshop Based On Estimation Of Distribution Algorithms

Facing the boiled market competition,the tire manufacturing industry faces new challenges,and the traditional production mode cannot meet the demand of the market.It is known that tire manufacturing is a large and labor-intensive industry.As the bottleneck of the tire production process,the vulcanization scheduling scheme has a direct impact on the economic benefit of the tire enterprise.This article focuses on problems of tire vulcanization process,which consist of taking a long time,high cost and other issues.Combined with characteristics of tire vulcanization workshop production and vulcanization process,express in mathematical formulas with the relationship between production variables.Take the various factors affecting the vulcanization plant production into account,a mathematical model of vulcanization workshop scheduling problem is built.To solve the vulcanization workshop scheduling problem,a distribution estimation algorithm is applied,which is benefit for increasing the production efficiency and reduce the production cost.The main content of this paper is as follows:Firstly,based on the characteristics of optimization scheduling in single-target vulcanizing shop,an improved discrete distribution estimation algorithm is proposed,which is used to solve the optimization problem of the vulcanization workshop scheduling with the goal of minimizing the makespan.The algorithm is integrated with the mutation operator to initialize the population and enriches the diversity of the population.According to population fitness evaluation function,calculate the fitness values of the individuals in the population,and select dominant group construct probability model.The probabilistic model is used to express the distribution information of the problem solution in space and the evolutionary trend of the population.From the known probabilistic model,a new population is generated using the random sampling rule,so that iteratively realizes population evolution.Secondly,aimed at minimum makespan and minimum cost of the mold changing and machine idle rate for multiple target vulcanization workshop production scheduling problem,utilize a variety of strategies to improve the estimation of distribution algorithm.The algorithm designs two kinds of coding methods,which are the encoding of sequence—number of vulcanization and encoding of vulcanization matrix.What's more,two methods of encoding mutual conversion are given.Besides,according to the rules of random initialization of population,use proportional increase strategy to select dominant groups.With the number of iterations of the algorithm increases,more proportions of dominant groups are selected to construct the probability model to more accurately fit the distribution information of the problem solution in the search space.Considering the lack of the search ability of the algorithm,binary search and dynamic fine tuning strategies are introduced to improve the algorithm convergence and accuracy.Finally,an improved distribution estimation algorithm is used to solve the vulcanization shop scheduling problem,and compare optimal scheduling results with improved discrete acoustic search algorithms and particle swarm optimization algorithms with examples.The experimental results show that the improved algorithm has faster convergence speed and higher precision,and imply that the effectiveness and feasibility of improved estimation of distribution algorithm for solving the vulcanization workshop scheduling problem.