Optimal dynamic pricing for public transportation considering consumer social learning.

Abstract

Effective public transportation pricing strategies are critical to reducing traffic congestion and meeting consumer demand for sustainable urban development. In this study, we construct a dynamic game pricing model and a social learning network model for consumers of three modes of public transportation including metro, bus, and pa-transit. In the model, the metro, bus, and pa-transit operators maximize their profits through dynamic pricing optimization, and consumers maximize their utility by adjusting their travel habits through social learning in the social network. The reinforcement learning algorithm is applied to simulate the model, and the results show that: (1) as consumers' perceived sensitivity to different modes of travel increases, the market share and price of each mode of travel adjust accordingly. (2) When taking into account consumers' social learning behavior, the market share of metros remains high, while the market shares of buses and pa-transit are relatively low. (3) As consumers become more sensitive to their perception of each travel mode, operators invest more resources in improving service quality to gain market share, which in turn affects the price of each travel mode. Our results provide decision support for optimal pricing of urban public transportation.