Data mining based framework to assess solution quality for the rectangular 2D strip-packing problem
作者：
Highlights：
• Data mining framework capable of assessing the quality of heuristic solutions.
• Regression model fitted with bottom-left-fill solutions and 19 predictors.
• 30,000 problem instances were generated to represent different 2D-SPP variations.
• Random forest is the data mining technique with the best level of generalisation.
• Data mining framework is consistent and can be applied for other problems.
摘要
•Data mining framework capable of assessing the quality of heuristic solutions.•Regression model fitted with bottom-left-fill solutions and 19 predictors.•30,000 problem instances were generated to represent different 2D-SPP variations.•Random forest is the data mining technique with the best level of generalisation.•Data mining framework is consistent and can be applied for other problems.
论文关键词：Strip-packing problem,Cutting and packing problem,Knowledge discovery,Data mining,Heuristics,Regression analysis
论文评审过程：Received 7 November 2017, Revised 14 September 2018, Accepted 3 October 2018, Available online 4 October 2018, Version of Record 20 October 2018.
论文官网地址：https://doi.org/10.1016/j.eswa.2018.10.006
