Analysis of Washington, DC taxi demand using GPS and land-use data

Taxis remain a key asset for urban mobility despite the tremendous growth of modern mobility-on-demand service providers such as Uber and Lyft. A fundamental understanding of the factors that impact the taxi demand is essential for planning an effective multi-modal transportation system, and can also shed lights on new on-demand services. This study addressed a gap in literature by investigating the correlation between demand for taxi, land use patterns, and accessibility to other modes using detailed GPS and GIS information collected from the Washington D.C. metropolitan area. The results of the models showed a strong link between demand for taxi, land use patterns, and accessibility to other modes. Mixed land use did not show a strong correlation with taxi demand. The study also found that the taxi mode is likely to complement metro trips, but compete with bus trips, although both of these modes of travel are considered public transit. Airport trips were found to be the most important source for taxi travel. These findings were further supported by the time-of-day and seasonality analysis.