Efficiently routing a fleet of autonomous vehicles in a real-time ride-sharing system

The advent of new communication technologies (e.g., smartphone) and autonomous vehicles (AVs) is enabling real-time ride-sharing systems where the travel requests arrives in the system on very short notice or even en-route, i.e., when AVs are already serving other users. Each request specifies an origin, a destination and a time window of pick-up, and must be immediately either accepted or rejected. Each request accepted must be assigned to an AV and both scheduled and inserted in its route considering the other possible requests already assigned to the same AV. In a lexicographic way, first we want to maximize the total number of new requests accepted, then we want to minimize the total traveled distance and finally, the total time of serving the requests. The problem is formulated as a Mixed Integer Linear Program and solved by a rolling horizon approach (MILP-RH). To efficiently address medium/large-sized instances, a rolling horizon Local Search (RHLS) is also designed, with moves properly tailored for the problem Numerical comparisons show that, on both the small-sized and some medium-sized instances, the RHLS outperforms the MILP-RH concerning the total computational time. Instead, on some medium-sized and on the large-sized instances, the RHLS is the only viable method since the MILP-RH is not able to even find a feasible solution in the given time limit. A sensitivity analysis on possible variation of some parameters is also performed deriving some useful managerial insights.

• Record URL:
  • https://doi.org/10.1016/j.cor.2024.106668
• Record URL:
  • http://www.sciencedirect.com/science/article/pii/S0305054824001400
• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/oclc/1793974
• Supplemental Notes:
  • © 2024 The Authors. Published by Elsevier Ltd. Abstract reprinted with permission of Elsevier.
• Authors:
  • Bruglieri, M
  • 0000-0002-4517-873X
  • Peruzzini, R
  • 0000-0002-5630-6573
  • Pisacane, O
• Publication Date: 2024-8

Language

• English

Media Info

• Media Type: Web
• Features: References;
• Pagination: 106668
• Serial:
  • Computers & Operations Research
  • Volume: 168
  • Issue Number: 0
  • Publisher: Elsevier
  • ISSN: 0305-0548
  • Serial URL: https://www.sciencedirect.com/journal/computers-and-operations-research

Subject/Index Terms

• TRT Terms: Autonomous vehicles; Mixed integer programming; Ridesharing; Routing; Vehicle fleets
• Subject Areas: Highways; Operations and Traffic Management; Planning and Forecasting; Vehicles and Equipment;

Filing Info

• Accession Number: 01921454
• Record Type: Publication
• Files: TRIS
• Created Date: Jun 13 2024 9:00AM