Learning-Based Pricing and Service Planning for Shared Automated Vehicle Services in a Competitive Multi-Provider Market
Y. Zhang (TU Delft - Electrical Engineering, Mathematics and Computer Science)
J.T. van Essen – Mentor (TU Delft - Electrical Engineering, Mathematics and Computer Science)
Q. Fan – Mentor (TU Delft - Civil Engineering & Geosciences)
G.N.J.C. Bierkens – Graduation committee member (TU Delft - Electrical Engineering, Mathematics and Computer Science)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
Shared autonomous vehicles (SAVs) have the potential to support sustainable urban mobility, but their operation in competitive markets requires coordinated pricing, service planning and operational decisions. This thesis develops a reinforcement-learning-based framework for competitive SAV service planning with multiple providers. Each provider learns its base fare, distance-based fare, maximum waiting time and service rate through repeated interaction with a market environment. A passenger choice model allocates demand among providers based on their joint decisions, after which integer linear programming models optimise fleet size, passenger assignment and vehicle routing for each provider. Provider profit is used as the reward signal for reinforcement learning. The framework enables providers to learn competitive pricing and service strategies while accounting for their effects on demand and operational performance. It provides a unified approach for studying the interaction between strategic service decisions and operational optimisation in competitive SAV markets.