Bus station’s choice of the elderly based on latent-class logit model

Regular public transportation is an important way for the elderly in China, and the bus station is an important part of the bus system. Right now, few papers analyzed bus stations’ impact on the bus choice behavior of the elderly. Based on the data from the stated preference method, the potential Logit model was used to study the influence of the existence of the real-time information and the stool at the bus station on the elderly’s bus choice behavior. Through analysis, it was found that the latent-class conditional Logit model has higher fitness than the traditional conditional Logit model. The elderly should be divided into two subgroups (class1, class2). The elderly in class1 didn’t take into account the existence of real-time information and the stool at the bus station. The relative value of the existence of the real-time information and the stool at the bus station for the elderly in class2, is equivalent to reducing the departure interval by 10.84 minutes and 28.08 minutes, respectively. Whether the elderly has to take care of a grandchild, and the subjective evaluation of the physical conditions, have a significant impact on the elderly’s latent class. While age, gender, income, living with child or not, bus use frequency, participating in senior activities or not, the number of chronic diseases and satisfaction don’t have significant impact on the elderly’s latent class.