Resarch On Hybrid Flow Shop Scheduling With Unrelated Parallel Machine

Hybrid Flow Shop Scheduling Problem with Unrelated Parallel Machine(HFSP-UPM)is a special kind of Hybrid Flow Shop Scheduling Problem(HFSP),which is widely used in flow production factories with mixed use of new and old machines.This paper takes HFSP-UPM as the research object,considers the heterogeneity of factory resources and batch constraints under distributed conditions,and studies DHHFSP-UPM(Distributed Heterogeneous HFSP-UPM)and BDHHFSP-UPM(DHNWFSP with Batch).Combining the characteristics of the problem,the Fruit Fly Optimization Algorithm(FOA)is used as the research method to solve the above problems.First of all,an IFOA(Improved FOA)algorithm is designed for HFSP-UPM,which aims to minimize the maximum completion time(Makespan).Based on the characteristics of the problem,a job-based encoding method is adopted,and the decoding adopts forward and reverse decoding strategies.In the IFOA algorithm,a new neighborhood structure based on the critical path is proposed based on the neighborhood structure of N7 and K-insertion,and the above three neighborhood structures are combined with the variable neighborhood search strategy to improve the local search ability of the algorithm.Compared with the test results of the existing literature,the stability and superiority of the IFOA algorithm are verified.Secondly,for the DHHFSP-UPM,a mathematical model was established,and an improved IFOA algorithm was designed to solve the problem.Based on the distributed characteristics,a double-layer encoding method based on jobs and factories is adopted,and a greedy decoding strategy is proposed by combining forward and reverse decoding strategies.The greedy search method based on critical factories is adopted to effectively improve the local search ability of the algorithm.The superiority of the IFOA algorithm is verified by solving calculation examples of different scales.Thirdly,the batch constraint was introduced into DHHFSP-UPM,comprehensively considering the two goals of minimizing the makespan and the total flow time,the BDHHFSP-UPM mathematical model was established,and a MOFOA(Multi-objective FOA)was proposed.Based on the batch characteristics,the coding based on the batch size of the sub-batch is introduced,a three-layer coding method is designed,and a sub-batch adjustment strategy is designed.The critical path neighborhood solution structure in the factory and the greedy search strategy between factories are extended to all factories,and the completion time and total flow time are fully reduced.The test results of calculation examples show that the comprehensive performance of the proposed MOFOA algorithm is better than other comparison algorithms.In addition,combined with the actual production process of a mobile phone case,the above theoretical research results are applied to actual projects,and the feasibility of improving the FOA algorithm to solve the actual scheduling problem is verified.Finally,the work of the full text is summarized,and several prospective directions are put forward.