Algorithm And Simulation Research On A Complex Batching Scheduling Problem For Injection Molding Industry

Injection molding enterprises mainly produce plastic productions, which are related toaspects of national economy and people’s life. As a key link and weakness, shop schedulingproblems have made a great influence on the injection molding enterprises development.From the view of how machines process workpieces, shop scheduling problems can beclassified as single scheduling and batch scheduling problems, which include parallel batchscheduling problems and serial-batch scheduling problems. As for injection molding productionscheduling problems, on the one hand, there are usually more than one mold cavities in injectionmould, with parallel batch scheduling features; on the other hand, continuous process of injectionmolding typically is the same as serial-batch scheduling problems. In this paper, a kind ofcomplex batch scheduling is presented based on the background of injection molding production.Productions are continuously processed in batches with single operation and model constraints.With flexibility in choosing machines, the problem consider economic nature of processingmachines, status of moulds, amount of equal moulds, cavities of mould, releasing time anddelivery date of works, weightiness, colors and so on. Main contents of the study are as follows:Firstly, from the introduction of actual production in injection molding enterprisesworkshops, an injection molding production scheduling model, which can reflect the actualcharacteristics of the manufacturing system, is constructed research on representation andprocessing methods about constraint conditions and objective function of the scheduling problem.Secondly, with respect to three decision problems involved in the problem, including batchpartition optimization, sequencing, injection mould and machinery resources, the paper exploreda divided solution strategy, and proposed a hybrid algorithm combining genetic algorithm andheuristics scheduling rules. Concerning genetic algorithm, the paper utilized codes of workpiecesbatch for chromosome encoding, partial crossover mapping for chromosome crossover, andgenes swap for chromosome mutation. Optimal solution was saved in the evolutionary process.As for resource selection, the Earliest Tardiness Time&Best Matching Degree resourcedecision rules were proposed.Finally, a scheduling simulation system was designed based on Java, which is an objectoriented language, and its development platform called JBuilder2006. In order to illustrate andverify the problem and algorithm proposed in this paper, we designed an experiment caseconstituted by19product batches and conducted the scheduling simulation. In addition, we alsodesigned the two scheduling rules called SEBPT and LEBSL for sorting the processing ordersaiming to compare the effectiveness of the algorithm and its performance. Through detailedcontrastive analysis of scheduling results, it was further verified the rationality and effectivenessof the proposed algorithm in this paper, thus it provided a valuable reference for subsequentscheduling system development of injection molding enterprise.