Automatic detection of the best performing priority rule for the resource-constrained project scheduling problem

作者：

Highlights：

• The decision tree is used to classify the best performing priority rule for RCPSP.

• Two models are proposed to map indicators onto the performance of priority rules.

• Three computational experiments evaluate the effectiveness of the models.

• Both classification models outperform any single priority rule.

摘要

•The decision tree is used to classify the best performing priority rule for RCPSP.•Two models are proposed to map indicators onto the performance of priority rules.•Three computational experiments evaluate the effectiveness of the models.•Both classification models outperform any single priority rule.

论文关键词：Project scheduling,RCPSP,Machine learning,Classification model,Performance prediction

论文评审过程：Received 18 May 2020, Revised 15 September 2020, Accepted 9 October 2020, Available online 21 October 2020, Version of Record 10 February 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.114116