Refuse or Accept?: Analysis of Taxi Driver Operating Strategies in E-Hailing Platforms

Abstract

For an e-hailing taxi operation, we analyze a driver's profit-maximizing reactive strategy (best evaluation criteria to either accept or refuse a ride request) in response to the ride request broadcasted by the platform and an associated driver penalty for refusing a ride request. We analyze six operating modes, which is a combination of three reactive strategies: no refusal, refusal based on proximity, and refusal based on profitability index and two broadcasting methods. In e-hailing taxi operations, the effect of the platform's broadcasting policy and topology of the service region on a driver's reactive strategy is mostly unknown. We derive structural properties for six operating modes and determine the optimal threshold values for refusals with proximity distance and profitability index criteria. We adopt a two-stage methodology to answer the research questions. In the first stage, through analytical techniques, we obtain closed-form expressions of the expected total profit for each operating mode and topology of the service region. In the second stage, we obtain the profit-maximizing reactive strategy with a derivative-free optimization method. We provide structural properties that showcase the similarities and differences among the operating modes. For a real e-hailing taxi application, we find that a driver could follow refusal based on proximity and earn approximately 11% more than the baseline - no refusal strategy. We find that refusal based on proximity is the best reactive strategy under most scenarios. Our results provide the best driver reactive strategy matrix and operating mode for the driver in an e-hailing market. The platforms can understand a driver's refusal behavior and the effect of penalty on driver's reactive strategy.