Information Systems Faculty Publications
Document Type
Article
Publication Date
7-13-2026
Abstract
The growing commercialization of autonomous vehicles (AVs) is reshaping consumer service preferences and prompting ride-hailing platforms to redesign fleet structures that accommodate the coexistence of human-driven vehicles (HVs) and AVs. This article develops a queueing game framework that incorporates vehicle heterogeneity and consumer preference differences to systematically compare three fleet configuration strategies: the pure HV (PHV) strategy (HVs only), the pure AV (PAV) strategy (AVs only), and the hybrid strategy (both HVs and AVs). The analysis highlights how consumer mismatch losses, AV operating costs, and service rates jointly shape equilibrium outcomes. Results show that when consumer mismatch losses are moderate, the optimal fleet strategy transitions from PAV to hybrid as AV operating costs increase—and may even revert unexpectedly to a PHV configuration. From a multistakeholder perspective, when mismatch losses are high and AV operating costs are relatively low, the hybrid strategy can generate win–win–win outcomes for consumers, drivers, and platforms. Moreover, increasing either the HV or AV service rate improves platform profitability and can attract more consumers, although an excessively high HV service rate may depress driver wages. Robustness checks—incorporating heterogeneous driver opportunity costs and fixed AV deployment costs—confirm that the central insights hold across broader settings. Overall, the findings offer practical guidance for platform managers and policy-makers navigating the operational and strategic challenges of AV integration.
Recommended Citation
Li, W., Zhang, Y., Sun, J., Yang, Z., 2026. Human-Driven, Autonomous, or Hybrid? The Optimal Fleet Configurations for Ride-Hailing Platforms. IEEE Transactions on Engineering Management 73, 4282–4296. https://doi.org/10.1109/TEM.2026.3712815
Publication Title
IEEE Transactions on Engineering Management
DOI
10.1109/TEM.2026.3712815
Comments
© 2026 IEEE. All rights reserved, including rights for text and data mining, and training of artificial intelligence and similar technologies. Personal use is permitted, but republication/redistribution requires IEEE permission. See https://www.ieee.org/publications/rights/index.html for more information.