Mobile Phone Data Reveal the Spatiotemporal Regularity of Human Mobility

Recent advance on human mobility are mainly based on mobile phone data since mobile phone records are the most detailed information across a large segment of the population in the modern society. With the spatiotemporal regularity missing in the individual and group level, we investigate the statistics of human mobility pattern using the mobile phone data provided by telecom in Guangdong, finding that the human activity pattern exhibits a heavy-tailed interval time distribution and regression property. We further demonstrate that the spatiotemporal characteristics can contribute to real-time travel prediction of human mobility and be applied in OD survey which is meaningful in traffic planning and management.