Research On Flexible Job Shop Scheduling Problem Based On Intelligent Optimization Algorithm

With the rapid development of multi-variety small-batch production mode,lot streaming has become an effective technology widely used in modern batch production.In recent years,researches on lot streaming flexible job shop scheduling problem mainly focus on the fixed number or equal quantity consistent sub-lots,but the optimal sub-lots division is difficult to determine directly.In addition,due to the complex production environment,machine failures,temporary order events,and frequent disturbances can easily lead to low efficiency or infeasibility of the original scheduling plan,affecting the stability of the production system.Therefore,this paper focuses on lot streaming flexible job shop scheduling problem in static environment and multi-objective flexible job shop scheduling problem in dynamic environment.The works are as follows:(1)In static lot streaming flexible job shop scheduling problem,an improved variable neighborhood search algorithm is proposed to solve the problem of sub-problem coupling,large search space,and the algorithm is prone to fall into local optimization.The matrix-vector form encoding is designed to deal with lot segmentation and lot sequence simultaneously.Multiple neighborhood structures are designed according to the characteristics of lot streaming problem,different sub-problems are coordinated and optimized by switching neighborhood structure,and save the solution to the elite set and the poor set.The critical process optimization mechanism is designed.This mechanism performs right-insert-left shift operations on sub-lots and critical process moves to make full use of machine idle time and make up for coding defects.In order to prevent solutions from falling into local optimization,a neighborhood scheme based on JAYA idea is designed,which makes full use of the valid information in the two archives.Finally,the effectiveness of the algorithm structure is verified through experiments.(2)In multi-objective dynamic flexible job shop scheduling problem,a migrating birds optimization algorithm based on game theory is proposed to solve the problem of using weighting methods to deal with multi-objective problems in the scheduling process,which tends to ignore the potential relationship between objectives and is difficult to determine the weight.Firstly,a two objective game mechanism is designed to handle machine allocation problems,ensuring Pareto optimality and fairness between targets.When solving the game matrix,a solution method that approximates the Nash equilibrium solution is designed for situations where there is no unique perfect Nash equilibrium solution.In the improved migrating birds optimization algorithm,neighborhood operators based on path relinking and machine age are designed to fully utilize the effective information in the non-dominated solution and improve the search ability.Based on the attributes of the multi-objective problem,a multiple similarity metric method is designed,which combines Hamming distance and Euclidean distance to select and replace solutions.Finally,the effectiveness of this algorithm in solving multi-objective dynamic flexible job shop scheduling problems is verified through simulation experiments.