Online clustering reduction based on parametric and non-parametric correlation for a many-objective vehicle routing problem with demand responsive transport
作者：
Highlights：
• An online cluster approach transforms an 8-objective problem onto a 2-objective one.
• A Pearson/τ-Kendall hierarchical cluster approach generates the 2-objective problem.
• The approach coupled with a state-of-art algorithm is tested on a realistic dataset.
• Comparing with the original formulation, the approach shows competitive performance.
• The approach shows a favourable impact on search efficiency and computational cost.
摘要
•An online cluster approach transforms an 8-objective problem onto a 2-objective one.•A Pearson/τ-Kendall hierarchical cluster approach generates the 2-objective problem.•The approach coupled with a state-of-art algorithm is tested on a realistic dataset.•Comparing with the original formulation, the approach shows competitive performance.•The approach shows a favourable impact on search efficiency and computational cost.
论文关键词：Vehicle routing problem with a demand responsive transport,Many-objective optimization,Dimensionality reduction techniques,Cluster analysis,Kendall’s correlation,Pearson’s correlation
论文评审过程：Received 16 August 2020, Revised 13 November 2020, Accepted 5 December 2020, Available online 16 December 2020, Version of Record 14 January 2021.
论文官网地址：https://doi.org/10.1016/j.eswa.2020.114467
