42. An rVNS metaheuristic to optimize airport ground-handling: a focus on catering operations assignment.
Invited abstract in session FC-2: Logistics, stream Logistics.
Friday, 13:30 - 15:00Room: M228
Authors (first author is the speaker)
| 1. | Alexandre Dupaquis
|
| APO, IRIT TOULOUSE | |
| 2. | Ronan Guivarch
|
| IRIT, INP Toulouse | |
| 3. | Sandrine Mouysset
|
| IRIT, INP Toulouse | |
| 4. | Daniel Ruiz
|
| IRIT, INP Toulouse | |
| 5. | Vincent Charvillat
|
| IRIT, INP Toulouse | |
| 6. | Frédéric Bouilhaguet
|
| IRIT, INP Toulouse |
Abstract
As processing power grows and discoveries are made, previously difficult problems surrounding airport management become solvable. Specifically, automatizing ground-handling operations to reach yet unmatched efficiency through optimization. We are interested in the catering problem, a blend of vehicle routing problems and scheduling problems, which can be categorized as a simultaneous supply & delivery problem. The MTCVRSPTW-MB (Multi-Trip Capacitated Vehicle Routing and Scheduling Problem with Time Windows and Meal Breaks) is extended to a heterogeneous fleet of trucks and drivers, with qualifications and different load patterns. We introduce a greedy heuristic to solve a relaxed version of the problem, as a baseline. Then, we implement a cooperative rVNS-based metaheuristic. We compare our algorithms on multiple real-life instances at San Francisco International Airport. Our computational study shows the effectiveness of each approach and brings out the strength of our metaheuristic. Our algorithms will help ground handlers specialized in catering operations, to speed up the decision-making process of assigning hundreds of tasks every day. In particular, we show the multiple extra benefits of the metaheuristic approach, including minimizing fuel consumption and allowing dynamic task re-assignment.
Keywords
- Scheduling
- Routing, location and capacity planning
- Heuristics and meta-heuristics
Status: accepted
Back to the list of papers