EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

1933. Leveraging Contextual Information for Robustness in Vehicle Routing Problems

Invited abstract in session WB-3: Interpretable Optimization Methods and Applications, stream Data Science Meets Optimization.

Wednesday, 10:30-12:00
Room: 1005 (building: 202)

Authors (first author is the speaker)

1. Irfan Mahmutogullari
Dept. of Computer Science, KU Leuven
2. Tias Guns
KULeuven

Abstract

We investigate the benefits of using contextual information in data-driven demand prediction to solve the robust capacitated vehicle routing problem with time windows. Instead of estimating the demand distribution or its mean, we introduce contextual machine learning models that predict demand quantiles even when the number of historical observations for some or all customers is limited. We investigate the use of such predicted quantiles to make routing decisions. We also evaluate the efficiency and robustness of the decisions obtained by both exact and heuristic methods. Our extensive computational experiments show that using a robust optimization model and predicting multiple quantiles is promising when extensive historical data is available. In scenarios with limited demand history, using a deterministic model with only a single quantile shows greater potential. Interestingly, our results also indicate that using appropriate quantile demand values within a deterministic model leads to solutions with robustness levels comparable to those of robust models. This is important because, in most applications, practitioners use deterministic models as the industry standard, even in an uncertain environment. In addition, because they are less computationally demanding and require only a single demand value prediction, deterministic models coupled with an appropriate machine learning model hold the potential for robust decision making.

Keywords

Status: accepted

Back to the list of papers