Surrogate-assisted automatic evolving of dispatching rules for multi-objective dynamic job shop scheduling using genetic programming
作者：
Highlights：
• A MO-GP-HH framework is proposed to solve the multi-objective DJSS problem.
• A surrogate-assisted fitness estimation is incorporated into the framework.
• Comparison of surrogates regarding their selection accuracy and execution time.
• Surrogate-assisted pre-selection significantly improved the algorithm.
• Computational costs of the MO-GP-HH framework are reduced up to 70%.
摘要
•A MO-GP-HH framework is proposed to solve the multi-objective DJSS problem.•A surrogate-assisted fitness estimation is incorporated into the framework.•Comparison of surrogates regarding their selection accuracy and execution time.•Surrogate-assisted pre-selection significantly improved the algorithm.•Computational costs of the MO-GP-HH framework are reduced up to 70%.
论文关键词：Multi-objective optimization,Surrogates,Hyper-heuristic,Genetic programming,Dynamic job shop scheduling,Machine learning
论文评审过程：Received 21 February 2022, Revised 30 June 2022, Accepted 15 July 2022, Available online 21 July 2022, Version of Record 2 August 2022.
论文官网地址：https://doi.org/10.1016/j.eswa.2022.118194
