End-to-end Deep Reinforcement Learning for Stochastic Multi-objective Optimization in C-VRPTW

TMLR Paper6721 Authors

Abstract:

In this work, we consider learning-based applications in routing to solve a Vehicle Routing
variant characterized by stochasticity and multiple objectives. Such problems are repre-
sentative of practical settings where decision-makers have to deal with uncertainty in the
operational environment as well as multiple conflicting objectives due to different stakehold-
ers. We specifically consider travel time uncertainty. We also consider two objectives, total
travel time and route makespan, that jointly target operational efficiency and labor regula-
tions on shift length, although more/different objectives could be incorporated. Learning-
based methods offer earnest computational advantages as they can repeatedly solve problems
with limited interference from the decision-maker. We specifically focus on end-to-end deep
learning models that leverage the attention mechanism and multiple solution trajectories.
These models have seen several successful applications in routing problems. However, since
travel times are not a direct input to these models due to the large dimensions of the travel
time matrix, accounting for uncertainty is a challenge, especially in the presence of multiple
objectives. In turn, we propose a model that simultaneously addresses stochasticity and
multi-objectivity and provide a refined training mechanism for this model through scenario
clustering to reduce training time. Our results show that our model is capable of construct-
ing a Pareto Front of good quality within acceptable run times compared to three baselines.
We also provide two ablation studies to assess our model’s suitability in different settings.

Submission Type: Regular submission (no more than 12 pages of main content)

Previous TMLR Submission Url: https://openreview.net/forum?id=RbuzrGD62Z

Changes Since Last Submission: Margins were adjusted in accordance with the tempelate. This is was a mistake on my side (the corresponding author) that went initially unnoticed.

Assigned Action Editor: ~Matteo_Papini1

Submission Number: 6721