Alimentation Deep Multiple Optimal Ant Colony Optimization to solve Vehicle Routing Problem with Time Windows
Open Access
- 14 July 2024
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 17-18
Abstract
Recently, there has been a noticeable increase in the adoption of deep learning techniques to tackle time window routing issues, which are prevalent in various real-world applications. Our paper introduces Alimentation Deep Multiple Optimal Ant Colony Optimization (AMO-ACO), a new neural-enhanced method for solving the Vehicle Routing Problem with Time Windows. Our model handles the problem as a multi-objective task, using transfer learning across constraints and exploring various paths for the ant colony from starting points.Keywords
Funding Information
- Office of Naval Research (ONR) Global (ONRG) and the U.S. Army Combat Capabilities Development Command Indo-Pacific (DEVCOM Indo-Pacific) (N62909-23-1-2018)
This publication has 6 references indexed in Scilit:
- DeepACO: Neural-enhanced Ant Systems for Combinatorial OptimizationarXiv, 2023
- Learning the Travelling Salesperson Problem Requires Rethinking GeneralizationarXiv, 2022
- Half a Dozen Real-World Applications of Evolutionary Multitasking, and MorearXiv, 2022
- POMO: Policy Optimization with Multiple Optima for Reinforcement LearningarXiv, 2021
- Explicit Evolutionary Multitasking for Combinatorial Optimization: A Case Study on Capacitated Vehicle Routing ProblemIEEE Transactions on Cybernetics, 2020
- Attention, Learn to Solve Routing Problems!arXiv, 2019