Integrating user values assessment with the augmented RFM model to enhance bus transit systems’ market segmentation

Abstract

Understanding “who deserves retention” is pivotal to save the recent bus ridership decline. Despite efforts in segmenting bus riders into various behavior statuses, which riders who need, and even deserve, intervention policies to retain are largely unknown, making it impossible to allocate limited retention resources in a precise manner. To address this critical issue, we augment traditional RFM models with retention policy-sensitivity and transit mode-preference dimensions to capture bus riders' user value index by employing the multi-category factor analysis to evaluate riders’ behavioral loyalty, intentional loyalty, consumption potential, and recent activity level with the Analytic Hierarchy Process and expert interviews. Based on the four evaluation metrics, we cluster the riders into five segments: High-potential, Possible churn, Regular, Bus-dependent, and Loyal and recommend targeted retention policies to suit the characteristics of these five segments. The proposed transit rider segmentation framework is supposed to provide references for transit operators to find the potential churners and those who need to be retained among the identified potential churners.