Research On Task Scheduling In Automated Storage And Retrieval System Based On Ant Colony Algorithm

Automated Storage and Retrieval System (AS/RS) is an important constituent of Modern Logistics System. With the optimizing of task scheduling in AS/RS, task time could be decreased and the system efficiency could be improved without adding investment of equipment. So the optimizing of task scheduling in AS/RS is a research focus. The optimizing of task scheduling in AS/RS refers some practical problems, so there is no uniform solution to solve all problems. Currently, the solution of the optimizing of task scheduling in AS/RS could be classified into two categories: One kind of solution focuses on the theoretical research. It pre-designs ideal premise and attempts to instruct the optimizing of scheduling. The shortcoming of this kind of solution is the pre-designed premise is too ideal to use in the practical application. Another kind of solution puts forward optimizing scheme for practical problems of factories. It conducts research by simulation method. And then the simulation method will upgrade to theory. Limited application domain is the flaw of this kind of solution. In the essay, background and current research status of AS/RS is analyzed firstly. And then, research on the task scheduling in Automated Storage and Retrieval System based on Ant Colony Algorithm according to disadvantages of current research and the actual situation of AS/RS. The research is consisted of the optimizing of stacking crane order picking and the optimizing of the transportation system.After the analysis of the flow and characteristics of stacking crane order picking, according to the unpredictability of the number of tasks, the constrain of container cubage and the constrain of stacking crane rated load, a novel optimizing scheduling problem model of Automated Storage and Retrieval System stacking crane order picking(ARSCOP)is put forward. Then, advantages of solving the optimizing of task scheduling in AS/RS by Ant Colony Algorithm (ACA) are explained. To improve the convergence time of ACA, avoid falling in local best and enhance the quality of solution, a new Non-Fixed Length Little Window based Self-adaptive Ant Colony-Genetic Hybrid Algorithm (NFLWSACG) is proposed to solve ARSCOP in terms of characteristics of ARSCOP. Comparing with ACA, the simulation result on ARSCOP shows that NFLWSACG has better performance, such as the convergence performance, the quality of results and the stability. Regardless the number of tasks, results of NFLWSACG is steady. And it proves that NFLWSACG can be used in various scale AS/RS.In the research about the optimizing of the transportation system, disadvantages of current research status and traits of scheduling problem are explained firstly. And then the ordering scheduling priority factor of stacking crane and transportation system is brought forward to resolve the bottleneck problem that the ordering scheduling problem of stacking crane and transportation system. Further, a new optimizing scheduling problem model of transportation system (TSOP) is proposed. The ordering scheduling priority factor of stacking crane and transportation system is used in the transition probability formula of ACA. Considering traits of TSOP and learning characteristics of Particle Swarm Optimization Algorithm, an original Dynamic Parameters Ant Colony Algorithm with Particle Swarm Characteristic (DPAPA) is presented to settle TSOP. The simulation result shows that the comprehensive performance indicator of DPAPA is better than ACA in TSOP. And the DPAPA has good convergence performance. The result also proves that DPAPA is more suitable to be used in the large-scale AS/RS.The essay is consisted of seven chapters. In the first chapter, the background and current research status of Automated Storage and Retrieval System is introduced. And then the optimizing scheduling problem of stacking crane order picking and transportation system is summarized. The optimizing scheduling problem model of Automated Storage and Retrieval System stacking crane order picking is proposed in the second chapter. The idea that solves ARSCOP by ACA is explained in the third chapter. In the fourth chapter, Non-Fixed Length Little Window based Self-adaptive Ant Colony-Genetic Hybrid Algorithm is presented to solve ARSCOP. In the fifth chapter, the optimizing scheduling problem model of the transportation system (TSOP) is brought forward. Then, Dynamic Parameters Ant Colony Algorithm with Particle Swarm Characteristic (DPAPA) is put forward to solve TSOP in the sixth chapter. The summary and expectation is showed in the seventh chapter.