Study On Group Scheduling For Optimal Setup In Job-shop

Manufacturing is the main body of the national economy,the founding conviction of a country,and the base stone of a power country.To meet the personalized demand of customers and improve the flexibility to market,the production mode of Customer to Manufacture(C2M)becomes the development trend of manufacturing enterprises.In this production mode,features of many varieties and small batch production are more and more obvious,which leads to frequent setup activity and a lot of setup time,and thereby increasing the complexity of shop-floor scheduling.Classical scheduling theory mainly considers the jobs‘ queue waiting time,and the jobs‘ setup time is rarely considered.Hence,to C2 M production mode,these will waste greater time.Through the experimental analysis,for some process,the setup time are dozen times higher than the processing time in C2 M production mode,it reduces the utilization rate of equipment and other resources and affects the production cycle of the product.It was found that the different scheduling scheme of shop-floor scheduling led to different setup time.It adds the difficulty of the production planning and implementation control.Therefore,in this paper,a set of shop-floor scheduling scheme for optimal setup time was constructed based on group technology.This research has important scientific research value and engineering practice significance for the discrete manufacturing shop-floor scheduling management.The research contents include the following five points:Firstly,in this paper,we analyzed the current situation of shop-floor scheduling and setup and the influencing factors of setup time.Based on it,the shop-floor scheduling problem with optimal setup based on group technology is presented,and the holistic model of group scheduling problem for shop-floor manufacturing is established.Secondly,the parts cluster grouping genetic algorithm based on processing resources was studied.It also put forward that the processing resources in tooling CNC shop mainly includes machine,the clamping methods,processing precision,CNC program,employees’ knowledge level,etc.Firstly,it classified the needed resources of processing parts,then divided these resources into child class for different categories,and took 0-1 integer coding to show whether the resource is needed in processing.It determined the weight of the core processing and general processing resources depending on the importance of setup time from processing resources.The "Similarity" between the parts was obtained by using the Jaccard coefficient,and this paper applied grouping genetic algorithm to determine the parts classification and group.The case study proved the feasibility and effectiveness of the methods proposed.Thirdly,single machine group scheduling for optimal setup was studied based on group technology.The optimization goal was to minimize the total tardiness time.The scheduling scheme of sequence depending on setup time was used to reduce the setup time.First,jobs were clustered grouping according to the similarity of job needed processing resources.Second,EDD-SDST-ACO heuristic rule was applied,and the SNR of taguchi design method was used to optimize the parameters of algorithm,and through case studied,we have compared the maximum,minimum,average,and the number of times of searching optimal solution of optimization goal respectively with the optimization rules proposed in this paper,Ant Colony Optimization algorithm,and Genetic algorithm.The running results verified the feasibility and effectiveness of the EDD-SDST-ACO heuristic rules.The results of the case study showed that the proposed scheduling rule in this paper,makespan was shortened by 22.9%,the total tardiness was shortened by 99% and the utilization rate of equipment was increased by 21.87%.than the scheduling rule which was currently adopted by the enterprise.Fourthly,the unrelated parallel machine scheduling for optimal setup was studied based on group technology.The optimization goal was to minimize the total tardiness time.The scheduling scheme of sequence dependent on setup time was used to reduce the setup time.Because the processing speed factors of each machine are different,the first step is to study all jobs group‘s allocation on each machine,and the second is to research the optimized sequence of each job group on the same machine,and the permutation order of jobs group is different so that the setup time and the total tardiness time are different.The setup time is assumed as 0 seconds in the same set of assignments.The mathematical programming model was also established)for this problem,and the target optimization was optimized by GATS algorithm.And the parameters of the GATS algorithm were optimized by the SNR of taguchi-method,and took case studies for different scale problems,and compared with the artificial colony(ABC)algorithm and genetic simulated annealing algorithm(GASA),respectively.It proves the feasibility and effectiveness of the GATS hybrid algorithm.Fifthly,the flexible job shop scheduling problem for optimal setup was studied based on group technology.The minimizing the maximum completion time was the optimization goal.Considering the constraints of sequence dependent on setup time,the processing time and machine load,it first choose processing machine according to the machines load and machining time and second makes processing sequence on the machine according to the setup time and processing time.An improved QCSO hybrid algorithm combined with quantum bit and cat group algorithm is proposed,it introduced quantum coding and completed the cat group position iterative update by updating quantum rotation angle.According to the change of the algorithm iterations number,the value of the dynamic MR was chosen,which expanded the solution of the space and improved the operation efficiency and speed of the algorithm.By simulation experiment,which the running results between the improved QCSO hybrid algorithm and Parallel Genetic algorithm(PGA)were compared.Results show that the improved QCSO hybrid algorithm has better optimization results and robustness.The results of the case study showed that the proposed scheduling rule in this paper,makespan was shortened by 38.14% and the utilization rate of equipment was improved by 31.55% than the scheduling rule which was currently adopted by the enterprise.