4.4 Article

Elite solutions and Tabu assisted variable neighbourhood descent for rescheduling problems in the steelmaking-refining-continuous casting process

期刊

出版社

SPRINGER
DOI: 10.1007/s10696-022-09465-8

关键词

Steelmaking-refining-continuous casting; Rescheduling; Iron and steel production; Variable neighbourhood descent; Tabu; Metaheuristics

资金

  1. Natural Science Foundation of Hubei Province [2021CFB368]
  2. Research Project of Hubei Provincial Department of Education [Q20201105]
  3. National Natural Science Foundation of China [51705177, 51905199, 71701156]
  4. China Postdoctoral Science Foundation [2021M692778]
  5. Humanity and Social Science Foundation of Ministry of Education of China [21YJAZH050]

向作者/读者索取更多资源

This paper proposes a mathematical model and a method to solve the problem of SCC rescheduling considering charge start-time delay. By designing three Tabu based neighbourhood structures and special strategies, the experimental results demonstrate the effectiveness of the proposed method.
Steelmaking-refining-Continuous Casting (SCC) is a bottleneck in the iron and steel production operation. In order to enhance production efficiency, SCC scheduling is employed to find an optimal schedule. Unfortunately, dynamic events such as charge start-time delay may occur in a real-world SCC process, which will invalidate the optimal SCC schedule, i.e., making the schedule not optimal or inexecutable. To cope with such a situation, SCC rescheduling is significant for generating a new optimal schedule. This paper proposes a mathematical model of the SCC rescheduling problem considering charge start-time delay, and further presents an Elite solutions and Tabu assisted Variable Neighbourhood Descent (ETVND) method to tackle the problem. The main framework of the ETVND method is Variable Neighbourhood Descent (VND). In the ETVND method, three Tabu based neighbourhood structures are elaborately designed. Moreover, three distinguished features are incorporated, i.e., an elite solutions based exploration strategy, two-layer local search based on the Fruit fly Optimization Algorithm, and multi-type perturbation. The first two features are devised to enhance the intensification abilities while the third is devised to improve the diversification abilities. Experimental results have demonstrated the effectiveness of the ETVND method by comparing with several algorithms in the literature. Further comparison experiments have validated the efficiency of the Tabu based neighbourhood structures and specially devised strategies.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.4
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据