Impact analysis of actual traveling performance on bus passenger’s perception and satisfaction

Customer satisfaction has always been a pivotal indicator of public transport service quality. However, most existing studies on public transport satisfaction focused on passenger perception without considering the actual traveling performance. Thus, whether passenger perception veritably reflects the actual performance remains to be investigated. This paper adopts the gradient boosting decision trees approach to reveal the relationship between actual on-board performance and passenger satisfaction driven by individual perception. The work utilizes a mixed dataset combining customer satisfaction survey (CSS) with actual traveling data collected in Shijiazhuang, China. Instead of a direct equivalence, the results indicate the complex non-linear relationship between passenger perception and actual on-board performance. These findings enable deriving strategies for administrators and operators to improve the public transport service quality and customer satisfaction in a more effective and targeted way