Model And Algorithm For Separable Linear Guideway Cutting Problem

With the rapid development of machine tool industry,linear guideway has wide application in CNC machine,precision machinery and electronic processing device,etc.The customer demand for products of linear guideway is varied,how to arrange the separating and cutting plan reasonably according to the customer's requirement,make full use of the material linear guideway,which is of great significance to improve the economic benefits of enterprises.This thesis first states the background of the research,and through access to a large number of literature and books,analyzes relevant research status about the domestic and foreign,This thesis introduces the relevant knowledge of the linear guideway,and translates the linear guideway cutting problem into over-sized bin packing problem.Then this thesis summarizes the algorithms and theories related with slotting packing problem.The process requirement and the actual operation process of the linear guideway separating and cutting are analyzed.Secondly,according to the linear guideway cutting problem,in the comprehensive consideration of the minimum length of the linear guideway split resolution restrictions,separate position restrictions,splicing restrictions,length of linear guideway,holes distance,edge distance,wastage of cutting,cutting direction and other requirements under the premise of customer demand,a nonlinear multi-objective integer programming model with the least number of lines and the least amount of waste is established.Combined with the characteristics of the linear guide way,in the comprehensive consideration of the linear guideway separating and cutting,designed fast heuristic algorithm with scrap priority and the least wastage.Among them,the heuristic algorithm of scrap priority to use the minimum number of the linear guideway,full use of scrap as greedy criteria,the priority use of scrap split the over-sized linear guideway,then,according to the principle of least wastage cutting linear guideway;the heuristic algorithm of the least wastage to produce wastage at least as greedy criteria,the priority use of non-over-sized linear guideway separate the over-sized linear guideway,then,according to the principle of least wastage cutting linear guideway.Against to the multi-objective mathematical program of linear guideway cutting,genetic algorithm and genetic tabu search algorithm are designed.Designed the two segment coding method which considering the product separating and cutting.For the two conflicting objectives of the model,we came up with the concept of wastage preference rate.The optimization objectives are converted to single objective by using wastage preference rate to convert wastage into raw material of preference.It provides an adjusting method for decision maker to achieve the desired separating and cutting programs.Combining the characteristics of the problem and the chromosome coding method and the heuristic algorithm,we design the crossover and mutation method for product separating,product cutting and cutting direction respectively.In order to avoid the premature phenomenon of genetic algorithm,the tabu search algorithm is embedded in the genetic algorithm.We designed the strategy of the heuristic neighborhood search method for linear guideway cutting,and the neighborhood search method for the minute adjustment of holes numble and the genetic process embedded in taboo search.These methods can effectively improve the performance and convergence of the algorithm.Finally,based on Excel,doing the model and algorithm for linear guideway cutting simulate experiment by VBA technology.Experiments include the contrast experiments of the algorithm parameters,different crossover and mutation operations,neighborhood search range,analysis the result of separating and cutting and the contrast experiments of the heuristic algorithm and genetic algorithm,tabu search,genetic tabu search algorithm.Through the above experiments,we verify the validity and feasibility of the proposed slotting optimization models and related algorithms.