A Flexible Solution Framework for Optimal Decision in Logistics Systems
This paper concentrates on the methodology of using evolutionary computation (EC) for decision optimization problems in the context of logistics systems. We introduce a new perspective for practitioners to synthesize the problem-specific optimizer. A new EC model is proposed for effectively use of the evolutionary search framework for practical logistics applications. The general principle of “trade-off between exploration and exploitation is significantly operable under our proposed solution framework with decoupled exploration and exploitation operators. Furthermore, the typical requirements of different logistics problems are discussed and the guidelines for deign and analysis of the tailored algorithms, derived from our generic solution framework, are comprehensively provided based on a series of analytical results. As an example, a typical warehousing task in the rotary rack Storage/Retrieval system studied to demonstrate the implementation details for practical problems. Additionally, comparison experiments are carried out and related results illustrate the benefits of proposed solution framework and design methodology for practical optimization problem.
evolutionary computation scheduling optimization warehousing system hybrid genetic algorithms
Pan Zhang Lei Jia Guohui Tian Xiaolei Li
School of Control Science and Engineering Shandong University Jinan, Shandong Province, China
国际会议
2007 IEEE International Conference on Automation and Lofistics
山东济南
英文
2007-08-18（万方平台首次上网日期，不代表论文的发表时间）