Inferring Unmet Human Mobility Demand with Multi-source Urban Data

摘要

As the sharing economy has been increasing dramatically in the world, the mobile-hailed ridesharing companies like Uber and Lyft in the US, Didi Chuxing in China has begun to challenge traditional public transportation providers such as bus, subway or taxis. Ridesharing companies have shown their ability to provide the mobility services where public transit has failed. The human mobility demand that cannot be satisfied by traditional transportation modes (unmet human mobility demand) can be served by the ridesharing companies. In this paper, we provide a 'hydrological' perspective for inferring unmet mobility demand patterns in cities with multi-source urban data. We observe that the unmet human mobility demand is proportional to the met mobility demand by examining the yellow taxi and the Uber data in New York City. Based on this observation, a Single Linear Reservoir (SLR) model has been proposed for modeling unmet human mobility demand from multi-source urban data.