Multi-objective Flexible Job Shop Scheduling Problem Based On Evolutionary Algorithm

With the development of technology,market demand is shifting to individualization and diversification,and companies have to be forced to accept production methods of multiple varieties and small-medium batches.This discrete production method causes the complexity and controllability of the information in the production process,which is easy to cause the quality is not guaranteed,the delivery cannot be scheduled,and the economic benefits are reduced.In view of this,flexible job shop scheduling problems have become an important development trend in practical complex manufacturing systems.Firstly,this paper introduces the research status of flexible job shop scheduling and its application methods of evolutionary algorithm.Then,the multi-objective flexible job shop scheduling is introduced in detail,and the mathematical model with the minimum completion time,maximum machine load and total machine load as the optimization objectives is constructed,and the reasons for using genetic algorithm are analyzed.Finally,we improve the genetic algorithm and apply it to flexible job shop scheduling problems in different situations.The main work of this paper is as follows:(1)Aiming at the problem that multi-objective flexible job shop scheduling problem can produce illegal solutions by using genetic algorithm,the non-dominated sorting genetic algorithm ?(NSGA-?)based on elite strategy is improved,which speeds up the search speed,and its effectiveness is verified by several benchmark examples.Through a large number of experiments,the variation of parameters in the improved NSGA-II algorithm for multi-objective flexible job shop scheduling problem is analyzed.(2)Aiming at the problem that there is no large-scale flexible job shop scheduling problem data set at present,this paper analyses its significance to the actual industrial manufacturing and introduces how to design and produce large-scale flexible job shop scheduling problem data set.In order to avoid the problem of local optimum due to the increase of data size,an improved multi-population NSGA-? algorithm is proposed,and the effectiveness and feasibility of the algorithm are verified by the data set generated for large-scale flexible job shop scheduling problem.