The Exponential Scaling Law for Taxi Mobility Patterns

Abstract

Given the influence of different travel modes on urban transportation planning, such as private cars and public transportation, it is crucial to accurately identify the law of mobility under a specific travel mode. However, existing mobility laws, from the gravity law to the visitation law, describe the properties of human mobility based on various travel modes at a large spatial scale and do not capture the laws in a single mode. Here, based on GPS trajectory data of taxis, we reveal that taxi mobility patterns are a combination of activities occurring in bursts and inactivity over long periods. By capturing the flow of taxis arriving at a specific location in many activities, an exponential scaling law of taxi mobility is discovered. It reveals an exponential relationship between the number of taxis visiting a specific location and both travel distance and visitation frequency. Finally, we study the travel process of a single taxi and reveal the mechanism by which individuals move in the city, which provides a certain explanation for the discovered scaling law of mobility. Our research advances the understanding of human mobility patterns in the context of taxi travel and provides an example for other researchers to study specific travel modes.