UCLA

Taxis play an important role in urban transportation by fulfilling people's demand for auto services without turning them into auto owners. In this thesis, I use Geographic Positioning System (GPS) data generated by 12,000 taxis in November, 2012 to analyze the spatio-temporal travel pattern of taxis in Beijing. ArcGIS and python programming are employed for data processing and computing. The result of the passenger pick-up and drop-off analysis on November 30th, 2012 confirms the geographical imbalance phenomenon mentioned in previous studies. The thesis also examines taxi driver's behavior by income and finds that drivers of higher daily incomes operate in larger territories. Additionally, time and monetary costs of the same trips made in transit mode and taxi mode are compared using travel time estimated by Google Maps. In extremely smooth traffic, when people consider their time worth more than an average of 30 Yuan an hour, they are likely to use taxis instead of transit.