| With the rapid development of modern industry and intelligent manufacturing,personalized customization has become a brand-new production mode and has developed rapidly.The core technical problem in this process is production scheduling.This article mainly takes the personalized production scheduling problem as the research object,and combines the specific cases of flexible job-shop scheduling to carry out the research of this paper:First,the flexible job-shop scheduling problem(FJSP)is taken as the main research point to analyze the optimal scheduling method of the flexible workshop with uncertain processing time.In order to accurately describe the degree of interference of random variables and the degree of permitted violation of constraints,this paper also introduces two uncertain parameters to study the robust optimization method of general linear programming problems where the processing time is subject to probability distribution,and it will be difficult to solve the random The FJSP model of time is changed to a deterministic robust peer-to-peer model,which can be solved quickly.Second,according to the conditions required by FJSP to effectively meet the decision requirements of machine allocation and workpiece sequencing,adopt a two-layer coding method based on machine coding and process coding,and use greedy algorithm to solve the problem of active scheduling decoding.To this end,the corresponding fitness function and chromosomes for testing unstable scheduling problems are designed.Deep fusion neighborhood search technology builds robust scheduling related algorithms.Finally,use standardized test cases and typical cases to test the experimental results.Experiments show that when the disturbance of processing time is uniformly distributed in [-1,1],the robust optimization method designed can obtain the robust model corresponding to FJSP.This kind of algorithm can quickly get the best answer,reduce the loss of performance,and get a confidence of up to 95%. |
PDF Full Text Request