دانلود رایگان مقاله تعادل بازار تاکسی با خدمات ارزیابی شخص ثالث

abstract

 With the development and deployment of new technologies, the oligopolistic taxi industry is transforming into a shared market with coexistence of both traditional taxi service (TTS) and app-based third-party taxi service (ATTS). The ATTS is different from TTS in both entry policy and fare setting, and brings competition into the market. To account for the revolution of the taxi industry, in this study, we analyze the characteristics of the TTS and ATTS, model the taxi market as a multiple-leader-follower game at the network level, and investigate the equilibrium of taxi market with competition (TMC Equilibrium). In particular, passengers are modeled as the leaders who seek to minimize their travel cost associated with taxi rides. Followers involve TTS and ATTS drivers, who compete for passengers to maximize their revenue. The network model captures selfish behavior of passengers and drivers in the taxi market, and we prove the existence of TMC Equilibrium for the proposed model using variational inequality formulations. An iterative algorithm is further developed to find the TMC Equilibrium, which corresponds to the strongly stationary point of the multi-leader-follower game. Based on numerical results, it is observed that fleet size and pricing policy are closely associated with the level of competition in the market and may have significant impact on total passengers cost, average waiting time, and fleet utilization.

6. Conclusion

In this paper, we model the taxi market with TTS and ATTS as the multi-leader-follower game at the network level and investigate the TMC Equilibrium. Specifically, taxi drivers and passengers are considered as players in the game and are assumed to have greedy behavior. We formulated the problem using quasi-variational inequality, analyzed the game structure and model properties, and proved that the TM-L/F game has a TMC Equilibrium. Seeing the difficulties of solving multi-leader-follower game, we decomposed the problem as solving two parameterized augmented VIs, and developed an iterative algorithm to find the strongly stationary point of the problem.