基于排队博弈的网约车服务模式选择研究

Motivated by different carpooling strategies in ride sharing platforms, this work studies two models: A solo-ride which offers solo rides only and a mixed model which offers both solo and shared rides. We conduct queueing economics analysis under different information revelation policies (observable and unobservable queues), for which we derive the customers’ equilibrium strategies and the platform’s revenue-optimal service prices. We have several interesting discoveries. First, when the market size is relatively small (large), hiding (revealing) the queue-length information helps improve the platform’s revenue. Second, the introduction of shared rides in the mixed model may cause unexpected backfires and be detrimental to the system’s revenue, especially when the market size is small. These results help gain useful insights into the management of ride sharing platforms, which in turn can be used to better inform the managers of the most effective carpooling strategies and information provision policies.