Automating Web Collection and Validation of GPS Data for Longitudinal Urban Travel Studies

Traditional paper and phone travel surveys are expensive, time consuming, and have problems of missing trips, illogical trip sequences, and imprecise travel time. Global Positioning Systems (GPS)-based travel surveys can avoid many of these problems and are becoming increasingly popular in major cities worldwide. However, methodologies have not been developed to catch up with the enormous amount of data generated by the GPS. This project established a Web-based geographic information systems (GIS) prototype to speculate travel modes and trip purposes from GPS data and to validate the results from the survey participants. The prototype has the functions to collect GPS data from participants through the Web, run algorithms to speculate travel modes and trip purposes from the GPS data and other transportation and land use data, send back the results to participants for verification or modification, and finally update the data. While reducing the burdens on participants by using GPS, this Web-based GIS prototype has the potential to provide accurate and validated travel information for transportation modeling and planning.