Abstract Vehicle miles traveled (VMT) is one of the basic metrics with applications in almost every aspect of traffic engineering and safety. Traditionally, there are many methods available to estimate the VMT, e.g., traffic volume counts, odometer readings, household questionnaire surveys, and fuel sale data. However, limitations of these techniques include high data-collection costs, low accuracy, and failure to estimate the VMT at the finely disaggregated levels. The study aims to develop an approach to estimate the VMT of the operating taxis according to the readily available GPS data, which bears the advantage of being accurate, all-weather, continuous, and less expensive. The study consists of: (1) identifying and removing the abnormal data with the potential issues from the GPS database; (2) computing the total VMT by summing the distances of every two sequential GPS points; and (3) validating the estimated VMT data from the practical and statistical perspectives (by error analysis and two-sample Kolmogorov-Smirnov test). The GPS data of operating taxis in Nanjing, China are used for the empirical study. The study serves to provide a feasible solution for the VMT estimation.
Vehicle Miles Traveled Estimation Based on Taxi GPS Data: A Case Study in Nanjing, China
Advances in Smart Vehicular Technology, Transportation, Communication and Applications ; 4 ; 320-327
2018-12-01
8 pages
Article/Chapter (Book)
Electronic Resource
English
Potential of Taxi-Pooling to Reduce Vehicle Miles Traveled in Washington, D.C.
Transportation Research Record | 2018
|Estimating vehicle miles traveled combined with ITS data
Online Contents | 2011
|Vehicle Miles Traveled in Multivehicle Households
Transportation Research Record | 2005
|Vehicle Miles Traveled in Multivehicle Households
Online Contents | 2005
|