Citation

Huang Z, Ling X, Wang P, Zhang F, Mao Y, Lin T, Wang FY. Transp. Res. C Emerg. Technol. 2018; 96: 251-269.

Copyright

(Copyright © 2018, Elsevier Publishing)

DOI

10.1016/j.trc.2018.09.016

PMID

unavailable

Abstract

Even though a variety of human mobility models have been recently developed, models that can capture real-time human mobility of urban populations in a sustainable and economical manner are still lacking. Here, we propose a novel human mobility model that combines the advantages of mobile phone signaling data (i.e., comprehensive penetration in a population) and urban transportation data (i.e., continuous collection and high accuracy). Using the proposed human mobility model, travel demands during each 1-h time window were estimated for the city of Shenzhen, China. Significantly, the estimated travel demands not only preserved the distribution of travel demands, but also captured real-time bursts of mobility fluxes during large crowding events. Finally, based on the proposed human mobility model, a predictive model is deployed to predict crowd gatherings that usually cause severe traffic jams.

Keywords: Road transportation

Language: en

Keywords

Big data; Data fusion; Human mobility; Travel demand estimation