Modeling And Optimization Of Integrated Process Planning And Scheduling Based On Network Graphs

Integrated Process Planning and Scheduling(IPPS)can make full use of the complementary properties between process planning and shop scheduling to improve the overall manufacturing system performance efficiency and quality,and therefore has important academic and engineering values.This paper is dedicated to the study of modeling and optimization methods for IPPS problems.In this paper,the mathematical model of process planning,as well as the mathematical model of IPPS problems,encoding and decoding methods,Multi-MILP Model Collaborative Optimization(MMCO)method,are systematically studied together with standard test sets and engineering application examples.The main research works are as follows:For process planning problems,the superiority of the process network graph representation is proved by the study comparing the optimal solution of the process network graph with the optimal solution of the process feature table.By analyzing the selecting logic of OR nodes to operation nodes,the necessary and sufficient conditions for selecting operations are proposed;the essence of operation sequencing is analyzed through three t precedence matrices,and the model considering the transmission time constraint between machines is established accordingly.Finally,the model is applied to solve open problems and compared with the existing approaches to verify the superiority and effectiveness.For the modeling method study of the IPPS problem,the integrated MILP model of the IPPS problem is established based on the model of the process planning problem proposed in the previous text.Since the solving speed is slow due to the limitation of the problem scales,a further study of modeling methods is carried out to improve its solving efficiency.In terms of the manufacturing flexibility contained in the IPPS problem,the flexibility is divided according to the types of decision sub-problems.And the modeling idea of flexibility decomposition is proposed to establish a hierarchical MILP model of IPPS.Finally,the two models are applied to solve the IPPS problem examples respectively,and compared with existing models and algorithms to verify their superiority and effectiveness.For the study of optimization algorithms for IPPS problems,an integrated encoding and decoding method is proposed.The method can realize the coding of IPPS problems by adding OR node sequences to an individual.The process planning and the scheduling parts can be represented simultaneously in one individual.Moreover,a Modified Genetic Algorithm(MGA)is designed based on the proposed coding methods.As for the precedence constraints between operations,the specifically designed operators can guarantee the feasibility of the operation sequence during the searching procedure.Then,the superiority of MGA is verified by testing on 37 well-known open problems.To further study the optimization algorithm of IPPS,the MMCO method is proposed.Exact methods including MILP fail to solve the problem in a short computing time.The current solving methods that have achieved good results are almost metaheuristics.But due to the randomness in the metaheuristics,the quality of solutions cannot be guaranteed.Therefore,this paper proposes the MMCO method composed of four sub-models that are responsible for solving four types of sub-problems respectively in an IPPS task.The submodels optimize the same individual and collaborate under a greedy searching framework.The MMCO is faster than the traditional MILP method and can obtain better solutions than metaheuristics.The experiments on the well-known Kim benchmark show that the MMCO obtains all optimal solutions which have not been found before by the existing approaches.And these new records prove that the proposed method can effectively solve the IPPS problem.For a domestic packaging machine forming module machining workshop and an aerospace KT machining workshop,the theory and method proposed in this paper are applied to the practice production scheduling of the above workshops.Based on the analysis of the workpiece process information in the two workshops,the above two cases are categorized into typical IPPS modes.The corresponding models and algorithms in this paper are used to solve the cases verifying their effectiveness.Finally,the above work is summarized,and future research directions are discussed.