It’s All in the Mix: Technology choice between driverless and human-driven vehicles in sharing systems
研究了共享出行运营商如何最优配置无人驾驶与有人驾驶车辆的比例，以提升利润，发现混合车队在需求波动时优势显著，利润提升可达20.4%。
Operators of vehicle-sharing systems such as carsharing or ride-hailing can benefit from integrating driverless vehicles into their fleet. In this context, we study the impact of optimal fleet size and composition on an operator’s profitability, which entails a non-trivial tradeoff between operational benefits and higher upfront investment for driverless vehicles. We analyze a strategic fleet sizing and composition problem, integrating a rebalancing problem, which we formalize as a Markov decision process. We incorporate the rebalancing problem with a time-dependent fluid approximation to devise a scalable linear programming solution approach, which we improve by state-dependent emergency rebalancing. We present a numerical study on artificial and real-world instances that reveals significant profit improvement potential of driverless and mixed fleets compared to human-driven fleets. For real-world instances, the profit improvement amounts up to 20.4% over exclusively human-driven fleets. If both vehicle types incur equal operational costs, operators optimally mix a small number of driverless vehicles with a large number of human-driven vehicles. Mixed fleets are particularly beneficial if demand varies over time, and operators consequently shift rebalancing to lower-demand periods.