Statistical mechanism of passenger mobility behaviors for different transportations

Abstract

Passengers’ boarding time interval is of great significance for analysis of collective mobility behaviors. In this paper, we empirically investigate the boarding time interval of mobility behaviors based on three large-scale reservation records of passengers traveling by three different types of transportation from a travel agency platform, namely airplane, intercity bus and car rental. The statistical results show that similar properties exist in the passengers’ mobility behaviors, for example, there are similar burstiness [Formula: see text] and memory [Formula: see text] for different time interval distribution, which indicates that the passengers’ mobility behaviors are periodical. Furthermore, we present a probability model to regenerate the empirical results by assuming that the passengers’ next boarding time interval will generate between a short time of 1–7 days with probability [Formula: see text] and a random long time with probability [Formula: see text]. The simulation results show that the presented model can reproduce the burstiness and memory effect of the boarding time interval when [Formula: see text] for three empirical datasets, which suggests the periodical behaviors with the probability [Formula: see text]. This work helps in deeply understanding the regularity of human mobility behaviors.