A Stochastic Evolutionary Dynamic Game Model for Analyzing the Ride-Sourcing Market with Limited Platform Reputation

Dong Mo,Xiqun (Michael) Chen,Zheng Zhu,Chaojie Liu,Na Xie

DOI: https://doi.org/10.1080/21680566.2023.2248399

2023-01-01

Abstract:Ride-sourcing services have become increasingly vital in meeting on demand mobility needs. However, safety issues and public health concerns are obstacles for passengers' trust in ride-sourcing service providers. In this paper, we propose a stochastic evolutionary dynamic game approach to modeling and managing the ride-sourcing market with limited platform reputation, and interpret some intriguing phenomena in terms of passengers' and drivers' travel behavior, e.g. survivor bias, reciprocity, and tragedy of the commons. A stochastic state dynamic game model is formulated to capture complicated interactions among three games/states, i.e. trust game, complaint game, and rating-based game. The stationary distribution of each state and transitional probabilities are deducted from the stochastic game theory. We analyze the ride-sourcing system's evolution and reveal its intrinsic mechanism based on the model. We found that, if the governmental support for complaints is inadequate and passengers bear the high complaint costs, drivers will be more inclined to compensate passengers privately (i.e. reciprocity) to escape punishment for unqualified service. As a result, the service quality will deteriorate in the long run. To prevent the deterioration of service quality, we explore adaptive pricing strategies for the ride-sourcing platform and suggest managerial strategies for the government. This paper bridges the gap between the dynamic evolution and the mechanism of trust in ride-sourcing services, which has been seldom studied in the literature. All of the analytical results are illustrated by numerical experiments.