Application Research Of Ant Colony Algorithm In Multi-objective Integrated Scheduling Problem

The integrated scheduling problem of production and transportation is one of the essential branches of scheduling.However,the existing studies on integrated scheduling often neglect the effect of inventory.In real production,inventory has a significant influence on the overall operational efficiency,since the completed products have to be stored before they are transported.Therefore,research on integrated scheduling considering inventory has important theoretical significance and application value.In terms of solving the integrated scheduling problem,the production and transportation integrated scheduling problems in existing studies are usually divided into two sub-problems,i.e.,production and transportation,and algorithms are respectively designed s to solve the two sub-problems to obtain the solutions of the original problem.In this way,the information collected from each of the two stages cannot be combined and utilized fully and effectively.Hence,two algorithms are presented to make full use of the information of each stage to solve the integrated scheduling problem considering inventory.The main work of this thesis is as follows:(1)In this thesis,the problem of the single customer integrated scheduling of production and distribution considering inventory is investigated for minimizing the total cost of production,inventory and transportation,and the algorithm FJACO is designed to solve the problem.First,a specific problem description is given for the research problem.Then,a method for dynamically generating two-stage coding is designed according to the characteristics of the multi-stage problem.After that,a probabilistic decision mechanism is designed based on the binary decision characteristics of storing and being transported in batches.Then the probability formula is updated according to the information of the overall solution to form a positive feedback algorithm structure in order to make full use of integrated scheduling solution information at various stages to improve the quality of the solution.Besides,this thesis also designs a candidate list to improve search efficiency and a local optimization strategy to improve the quality of the solution.Finally,the simulation experiment results show that the overall performance of the proposed algorithm is better than other comparative algorithms.(2)This thesis further extends the study of the multiple customers integration scheduling problem,that is,in the case of multiple customers,the processing and transportation stages of the job need to consider the difference between the customer and the job preference.Based on this,this thesis proposes a Feedback Joint Ant Colony Optimization with Preferences(FJACOP).First,the coding method was redesigned according to the customer to which the jobs belongs.According to the preference attribute of jobs under multiple customers,heuristic information based on job preference is designed.Then,for improving the performance of the proposed algorithm,this thesis develops a probabilistic decision-making mechanism with preference,and on this basis,a multiple customers joint decision mechanism is proposed.Finally,the effectiveness of the algorithm is verified through a large number of simulation experiments.