The invention relates to a public transportation and customized public transportation collaborative dynamic pricing method and
system. The method comprises the following steps: acquiring and fusing passenger behaviors, a vehicle
Internet of Things, an external environment, a competition situation and operation cost data; constructing a dynamic user portrait based on the fused data, carrying out clustering and grouping, and generating price sensitivity,
time sensitivity, comfort preference and service loyalty portrait tags; predicting
payment willingness of the user based on the portrait; a collaborative pricing and travel recommendation strategy for different groups is generated by taking the
system state and the
user group portrait vector as input through a
reinforcement learning pricing engine; executing the strategy and carrying out differentiated
ticket price adjustment and service recommendation; passenger feedback data are collected, and user portraits and
reinforcement learning model parameters are updated. According to the invention, the conversion from static splitting pricing to dynamic collaborative pricing is realized, the problems of poor
collaboration and lack of elasticity of the
system are solved, and the revenue capability and
resource allocation efficiency of the system are remarkably improved.