EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
1816. New Reinforcement Learning Algorithms for Pickup and Delivery Routing Problems
Invited abstract in session MB-58: Heuristics for Vehicle Routing 3, stream VeRoLog - Vehicle Routing and Logistics.
Monday, 10:30-12:00Room: S07 (building: 101)
Authors (first author is the speaker)
| 1. | King Tong Wong
|
| Business School, University of Edinburgh | |
| 2. | Tsung-Sheng Chang
|
| Department of Transportation and Logistics Management, National Yang Ming Chiao Tung University | |
| 3. | Jamal Ouenniche
|
| Management School and Economics, Edinburgh University | |
| 4. | Joosung Lee
|
| University of Edinburgh |
Abstract
The Pickup and Delivery Routing Problems (PDRPs) are challenging combinatorial optimization problems with significant real-world applications. This paper proposes new Reinforcement Learning (RL) methods to address this type of routing problems. The typical mathematical programming formulations and the existing solution methods are first presented. Then, the concepts of Reinforcement Learning (RL) and Deep Learning are introduced, and the PDRP is reformulated as a RL problem. Several RL algorithms including new proposals are discussed and their performance is compared to an exact solution. This research contributions include employing RL to solve PDRP and comparing various RL algorithms including new proposals.
Keywords
- Vehicle Routing
- Machine Learning
Status: accepted
Back to the list of papers