Learning to Solve Multiple-TSP With Time Window and Rejections via Deep Reinforcement Learning

Journal Article (2023)
Author(s)

Rongkai Zhang (Nanyang Technological University)

Cong Zhang (Nanyang Technological University)

Zhiguang Cao (Singapore Institute of Manufacturing Technology)

Wen Song (Shandong University)

Puay Siew Tan (Singapore Institute of Manufacturing Technology)

Jie Zhang (Nanyang Technological University)

Bihan Wen (Nanyang Technological University)

J.H.G. Dauwels (TU Delft - Signal Processing Systems)

Research Group
Signal Processing Systems
Copyright
© 2023 Rongkai Zhang, Cong Zhang, Zhiguang Cao, Wen Song, Puay Siew Tan, Jie Zhang, Bihan Wen, J.H.G. Dauwels
DOI related publication
https://doi.org/10.1109/TITS.2022.3207011
More Info
expand_more
Publication Year
2023
Language
English
Copyright
© 2023 Rongkai Zhang, Cong Zhang, Zhiguang Cao, Wen Song, Puay Siew Tan, Jie Zhang, Bihan Wen, J.H.G. Dauwels
Research Group
Signal Processing Systems
Issue number
1
Volume number
24
Pages (from-to)
1325-1336
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

We propose a manager-worker framework (the implementation of our model is publically available at: https://github.com/zcaicaros/manager-worker-mtsptwr) based on deep reinforcement learning to tackle a hard yet nontrivial variant of Travelling Salesman Problem (TSP), i.e. multiple-vehicle TSP with time window and rejections (mTSPTWR), where customers who cannot be served before the deadline are subject to rejections. Particularly, in the proposed framework, a manager agent learns to divide mTSPTWR into sub-routing tasks by assigning customers to each vehicle via a Graph Isomorphism Network (GIN) based policy network. A worker agent learns to solve sub-routing tasks by minimizing the cost in terms of both tour length and rejection rate for each vehicle, the maximum of which is then fed back to the manager agent to learn better assignments. Experimental results demonstrate that the proposed framework outperforms strong baselines in terms of higher solution quality and shorter computation time. More importantly, the trained agents also achieve competitive performance for solving unseen larger instances.

Files

Learning_to_Solve_Multiple_TSP... (pdf)
(pdf | 3.85 Mb)
- Embargo expired in 05-10-2023
License info not available