Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Bi-Objective Approach to Last-Mile Delivery Routing Considering Driver Preferences

Juan Pablo Mesa, Alejandro Montoya, Raul Ramos-Pollán, Mauricio Toro

TL;DR

The paper addresses last-mile routing under driver preferences by formulating a bi-objective MOVRP that minimizes routing cost and aligns with drivers' historical routes. It compares two preference-encoding strategies—visual attractiveness metrics and data mining of historical routes—and demonstrates, on real Amazon/MIT data, that data mining more accurately predicts driver route sequences, enabling stronger performance in practice. A two-stage GRASP with heuristic box splitting is proposed to approximate the Pareto front of the bi-objective CATSP, yielding a small, decision-friendly set of non-dominated solutions. The results suggest substantial practical value in balancing cost with driver familiarity for scalable last-mile operations, with potential for integration into broader logistics planning.

Abstract

The Multi-Objective Vehicle Routing Problem (MOVRP) is a complex optimization problem in the transportation and logistics industry. This paper proposes a novel approach to the MOVRP that aims to create routes that consider drivers' and operators' decisions and preferences. We evaluate two approaches to address this objective: visually attractive route planning and data mining of historical driver behavior to plan similar routes. Using a real-world dataset provided by Amazon, we demonstrate that data mining of historical patterns is more effective than visual attractiveness metrics found in the literature. Furthermore, we propose a bi-objective problem to balance the similarity of routes to historical routes and minimize routing costs. We propose a two-stage GRASP algorithm with heuristic box splitting to solve this problem. The proposed algorithm aims to approximate the Pareto front and to present routes that cover a wide range of the objective function space. The results demonstrate that our approach can generate a small number of non-dominated solutions per instance, which can help decision-makers to identify trade-offs between routing costs and drivers' preferences. Our approach has the potential to enhance the last-mile delivery operations of logistics companies by balancing these conflicting objectives.

A Bi-Objective Approach to Last-Mile Delivery Routing Considering Driver Preferences

TL;DR

The paper addresses last-mile routing under driver preferences by formulating a bi-objective MOVRP that minimizes routing cost and aligns with drivers' historical routes. It compares two preference-encoding strategies—visual attractiveness metrics and data mining of historical routes—and demonstrates, on real Amazon/MIT data, that data mining more accurately predicts driver route sequences, enabling stronger performance in practice. A two-stage GRASP with heuristic box splitting is proposed to approximate the Pareto front of the bi-objective CATSP, yielding a small, decision-friendly set of non-dominated solutions. The results suggest substantial practical value in balancing cost with driver familiarity for scalable last-mile operations, with potential for integration into broader logistics planning.

Abstract

The Multi-Objective Vehicle Routing Problem (MOVRP) is a complex optimization problem in the transportation and logistics industry. This paper proposes a novel approach to the MOVRP that aims to create routes that consider drivers' and operators' decisions and preferences. We evaluate two approaches to address this objective: visually attractive route planning and data mining of historical driver behavior to plan similar routes. Using a real-world dataset provided by Amazon, we demonstrate that data mining of historical patterns is more effective than visual attractiveness metrics found in the literature. Furthermore, we propose a bi-objective problem to balance the similarity of routes to historical routes and minimize routing costs. We propose a two-stage GRASP algorithm with heuristic box splitting to solve this problem. The proposed algorithm aims to approximate the Pareto front and to present routes that cover a wide range of the objective function space. The results demonstrate that our approach can generate a small number of non-dominated solutions per instance, which can help decision-makers to identify trade-offs between routing costs and drivers' preferences. Our approach has the potential to enhance the last-mile delivery operations of logistics companies by balancing these conflicting objectives.
Paper Structure (19 sections, 23 equations, 6 figures, 6 tables, 7 algorithms)

This paper contains 19 sections, 23 equations, 6 figures, 6 tables, 7 algorithms.

Table of Contents

  1. Introduction
  2. Related works
  3. Motivation
  4. Problem description
  5. Solution method
  6. Two-stage GRASP algorithm for the CATSP
  7. Adapting the CATSP solver with visual attractiveness metrics
  8. Adapting the CATSP solver with data mining from historical routes
  9. Adapting the CATSP solver for solving the bi-objective routing problem
  10. Computational experiments
  11. Experiments of visual attractiveness metrics versus historical patterns mining
  12. Experiments of bi-objective vehicle routing problem
  13. Conclusions
  14. Acknowledgments
  15. Conflicts of Interest
  16. ...and 4 more sections

Figures (6)

  • Figure 1: Different routing solutions for the same instance of the Amazon dataset.
  • Figure 2: Distributions of the scores obtained with different routing methods on the evaluation dataset.
  • Figure 3: Approximated Pareto front for instance 9.
  • Figure 4: Approximated Pareto front for instance 31.
  • Figure 5: Approximated Pareto front for instance 184.
  • ...and 1 more figures