Research On Dynamic Flexible Job Shop Scheduling Problem Under Urgent Order Disturbance

Dynamic events,such as machine breakdown,urgent orders,random arrival of jobs,delay in delivery and emergency order,occur frequently in the actual production scenario,which has an increasing impact on the production efficiency and stability of the manufacturing industry.As an effective solution to the interference problem of dynamic events,Dynamic Flexible Job Shop Scheduling Problem(DFJSP)becomes one of the most popular research areas in manufacturing field.In this paper,the urgent order is selected as a dynamic event and a single objective model with scheduling efficiency as the performance index is established.Furthermore,considering both efficiency and stability,a multi-objective model with dynamic event is constructed.Finally,these two models are optimized by the hybrid of Particle Swarm Optimization and Genetic Algorithm.The main research contents are as follows:(1)The rescheduling strategies and characteristics of different disturbances in DFJSP are deeply analyzed.Combined with the research status of flexible shop dynamic scheduling in the domestic and overseas,the important and difficult points in the current dynamic scheduling problems are analyzed.On this basis,the Dynamic Interaction Layer(DIL)is used to complete the rescheduling process.Then the rescheduling process and information interaction mode are systematically formulated.In addition,the operation logic of Pareto multi-objective optimiza-tion algorithm is analyzed,and the calculation method of each index in DFJSP is given.(2)In order to solve the single objective dynamic scheduling problem,a single objec-tive DFJSP model is constructed with the makespan as the initial scheduling objective,and the makespan and completion time of the urgent order as the dynamic scheduling objective.Based on the single objective model,the Particle Swarm Genetic Algorithm(PSGA)is designed.This algorithm combines the position update strategy of Particle Swarm Optimization(PSO)with the genetic evolution strategy to improve the global and local search ability of the algorithm.Then,DIL and PSGA are adopted to solve the unexpected situation of urgent order in workshop pro-duction.Finally,the simulation experiment verifies the ability of DIL to process urgent orders and the effectiveness of PSGA algorithm.(3)For the multi-objective dynamic scheduling problem,a multi-objective flexible job shop dynamic scheduling model with makespan,machine load variance and energy consump-tion as the optimization objectives is firstly constructed in this paper,and a Multi-objective Particle Swarm Genetic Algorithm(MOPSGA)is designed based on the PSGA.The algorithm includes elite gene serialization strategy,gene pool selection strategy,and individual crowd-ing degree elimination mechanism based on neighborhood.In addition,in order to solve the problem of urgent order,the complete rescheduling strategy and the insert offset rescheduling strategy are used to carry out several groups of controlled tests.By comparing the performance data of the two rescheduling schemes,the applicable scenarios of rescheduling strategy are ob-tained.Finally,the experimental results verify the effectiveness of the rescheduling strategy and MOPSGA.In this paper,the single-objective and multi-objective optimization algorithms are designed respectively for the dynamic scheduling problem of flexible workshop.Combined with DIL and rescheduling strategies,the interference problem of urgent orders is solved,and the ability of the workshop to deal with the interference of dynamic events is also improved.