Conversion of Volunteer-Collected GPS Diary Data into Travel Time Performance Measures: Final Report

Conventional travel-survey methodologies require the collection of detailed activity-travel information, which imposes a significant burden on respondents, thereby adversely impacting the quality and quantity of data obtained. Advances in the global positioning system (GPS) technology have provided transportation planners with an alternative and powerful tool for more accurate travel-data collection with minimal user burden. The data recorded by GPS devices, however, do not directly yield travel information; the navigational streams recorded by GPS devices have to be processed and the travel patterns derived from them. This research project developed prototype software to automate the processing of raw GPS data and to generate outputs of activity-travel patterns in the conventional travel-diary format. The software identifies trips and characterizes them by several attributes, including trip-end locations, trip purpose, time of day, distance, and speed. This final report documents the entire research performed as part of this project. Specifically, the authors present the conceptual overview of the software, the detailed algorithm for extracting travel diaries, the software implementation procedures, and the testing and validation of the software.