Research On Flexible Job Shop Scheduling Considering Learning Effect And Optimization Of Energy Consumption

The problem of flexible job-shop scheduling has always attracted the attention of scholars,and the research in this area has matured.But in the actual production of the workshop,there are always workers,the learning effect is therefore widespread,considering the impact of the worker learning effect on the completion time,the scheduling optimization accuracy is higher;and with the implementation of the national energy conservation and emission reduction policy,enterprises pay more and more attention to energy consumption in daily operation.Therefore,in the common multi-objective flexible workshop scheduling,considering the impact of the learning effect on the processing time of the production process,and taking into account the optimization of energy consumption in the production process,enterprises can be more scientific and accurate in terms of actual scheduling problems to achieve further optimization of the goal.In this paper,theoretical analysis and NSGA-II.and MOPSO algorithms are used to study the multi-objective flexible workshop scheduling problem.Firstly,a multi-objective flexible shop floor operation scheduling model with the goal function of the maximum completion time and minimum energy consumption is constructed;secondly,NSGA-II algorithm based on the greedy decoding algorithm is designed,which performs double-layer real number coding based on the process and the machine,inserts the greedy algorithm into the decoding process,and obtains the sequence number of the process to be processed in the current equipment and the processing gap existing in the current equipment based on the workpiece operation number,and determines the position where the workpiece operation can be inserted in the equipment gap matrix.In addition,the start and completion time and machine energy consumption usage are updated;an external container R is added to the traditional MOPSO algorithm to record the number of machine usages,record the particle swarm information of the cumulative generation,and correct the particle swarm information of the new generation in combination with the learning function.Finally,based on the actual production situation of a coating flexible workshop of YT Company,the constructed model and the improved algorithm are applied,the actualcalculation is compared and analyzed,and the flexible scheduling workshop model considering the learning effect and energy consumption optimization is verified to be established,and the impact of different learning rates on the scheduling results,that is,the maximum completion time and the minimum energy consumption,is obtained.Select the group with the most obvious impact on the learning rate,when β = 70%,each of the three schemes of the two algorithms is obtained,namely the maximum completion time is optimal,the total energy consumption is optimal and the compromise scheme is obtained,and the algorithm iteration number change graph and result analysis,it is concluded that in this model and the study solution,NSGA-II.has better performance,faster optimization and more uniform distribution,of which the NSGA-II generated scheduling 1(time optimal scheme)has a minimum completion time of 3754 min and a minimum energy consumption of 16095996 J.Compared with the unoptimized initial plan,it can save the workshop a maximum completion time of15.4% and total energy consumption of 9.6%,improve the accuracy of the actual scheduling to a certain extent,help the company to save energy and efficiency more accurately,and the dispatcher can also choose a more suitable program according to their own needs.