Research On Flexible Job-Shop Scheduling Problem Based On Hybrid Estimation Of Distribution And Ant Colony Algorithm

As automation in manufacturing industry in China grow deeper,the scheduling and control mode of workshop has gradually developed from the traditional simplicity to the diversification and flexibility nowadays,which can be more adaptable to today's complex and variable production environment and various production needs.This paper mainly studies the static and dynamic scheduling problems in the relatively more complex flexible job-shop.Each process in the flexible job shop scheduling problem can be selected in multiple devices,and the order of work pieces processed on each device varies with the process selection.In actual production,multi-objective optimization scheduling problem is more in line with the production needs of enterprises,which makes this paper practically significant.The dynamic scheduling problem is based on the static scheduling problem.The disturbance of the initial scheduling scheme is analyzed,and then the appropriate dynamic scheduling scheme is adopted to solve the problem.The main contents of this paper are as follows:The first chapter describes the research background and significance of the paper,analyses the research situation of flexible job shop scheduling home and abroad,summarizes the research status of ant colony algorithm and distributed estimation algorithm in flexible job shop scheduling,and introduces the main research content and organization structure of the paper.The second chapter introduces the problem description and solution of flexible job shop scheduling,explains the theory of multi-objective optimization,and expounds the research methods and related performance indicators of flexible job shop scheduling under multi-objective.The third chapter describes the characteristics of ant colony algorithm and distributed estimation algorithm,and proposes a hybrid estimation of distribution and ant colony algorithm.The probability model based on population incremental learning is selected to generate new population in the distributed estimation algorithm,and the pheromone updating mechanism is also improved in the ant colony algorithm.Finally,the concrete steps and flow chart of the hybrid estimation of distribution and ant colony algorithm are given.The fourth chapter first establishes the mathematical model of multi-objective flexible job shop scheduling problem and the comprehensive objective function by weighting method.Then,according to the characteristics of flexible job shop scheduling problem,the hybrid estimation of distribution and ant colony algorithm is described in detail.Finally,the simulation experiment is carried out and compared with other algorithms.The effectiveness of the proposed hybrid algorithm is demonstrated.In the fifth chapter,the dynamic scheduling problem of flexible job shop is analyzed,and the related dynamic scheduling strategies are introduced.A dynamic scheduling strategy based on the combination of hybrid estimation of distribution and ant colony algorithm and rolling window rescheduling is designed,and the perturbations in two cases of emergency job insertion and machine failure are experimented and analyzed.The sixth chapter summarizes the research contents of the full text,and prospects the future research directions.