2306. Consultant Routing Problem: A Multi-Depot Vehicle Routing Problem with Multiple Visits and Trips
Invited abstract in session FA-12: Scheduling in Healthcare II, stream Health Care Management.
Friday, 8:45-10:15Room: H10
Authors (first author is the speaker)
| 1. | Jens Brunner
|
| Department of Technology, Management, and Economics, Technical University of Denmark | |
| 2. | Johannes Uhrmann
|
| Faculty of Business and Economics, University of Augsburg | |
| 3. | Christian Jost
|
| Operations Management, Technical University of Munich | |
| 4. | Sebastian Schiffels
|
| Wirtschaftswissenschaftliche Fakultät, Universität Augsburg |
Abstract
We study a real-world vehicle routing problem faced by pharmaceutical companies, where medical consultants regularly travel to meet physicians for product consultations. Given the long and frequent travel distances, as well as rising transportation costs, these visits represent a significant expense for companies. Therefore, efficient planning of tours is essential. In practice, however, this is often done manually, typically by the consultants themselves. Working together with a practice partner, our goal is to analyze the potential of centralized planning using mathematical optimization. Several key constraints must be considered: (1) all tours must start and end at the consultant’s home; (2) physicians may be visited multiple times within the planning period; (3) each planning period has a maximum number of working days, with each working day corresponding to a single tour. We model the problem as a multi-depot vehicle routing problem with multiple visits and multiple trips to accurately capture these requirements. The objective is to minimize the total travel time of all consultants while adhering to various operational constraints, such as fair distribution of visits, compliance with working hours, and the strengthening of client relationships. Due to the high complexity of the resulting model, we propose a decomposition-based solution approach using column generation, in which the pricing subproblems are formulated as time-constrained shortest path problems. To efficiently solve these subproblems, we explore several variants of the labeling algorithm, and present preliminary computational results that illustrate their comparative performance and practical viability.
Keywords
- Health Care
Status: accepted
Back to the list of papers