Abstract: In order to research the scaling law of the passenger travel time interval and the formation mechanism for the scaling law, the particle swarm optimization algorithm is used to fit the probability distribution of the travel time intervals of the passengers’ group. It is found that the time intervals of the group do not obey the unified power law distribution. Based on the results obtained by the fitting, the analysis is made from three aspects: the burstiness, the regularity of the passenger’s travel time and the purpose of travel. The results show that the 78.2% of time interval is concentrated within the 14 hours and the 21.8% of time intervals is greater than 14 hours, which makes the two ranges correspond to different power exponents. The value of the standard deviation of the time intervals is greater than its average value, resulting in a fat tail in the distribution. A large number of time intervals of random travel passengers causes the time intervals of passengers’group deviate from the poisson distribution. Passengers traveling for different purposes have different time interval ranges, resulting in uneven time intervals of the passenger’ group. These results indicate that passenger travel time intervals are mainly concentrated within 14 hours, which provides a new way for the traffic management department to predict the passenger travel time.

Key words: passenger, time interval, particle swarm optimization, power law distribution, fat tail phenomenon, poisson distribution, travel time, probability distribution