Observing travel behaviour from GPS data - A tool comparison survey in the Torino metropolitan area

Travel surveys help researchers to paint a clear picture of specific aspects of travel behaviour. In the transport field, data quality is largely dictated by the data requirements of mathematical models, and by the rising complexity of individuals' travel behaviour. Beginning with an illustration of the most common transport models, this thesis will first present an overview of traditional survey tools, in order to understand their structural biases and current developments in the transport survey field. One of the recent solutions to common data collection problems has been the implementation of passive data collection tools in household and personal travel surveys. Passive data collection tools allow researchers to derive travel behaviour information from positional and navigational data, collected with devices that use location-aware technologies, such as GPS, GSM, and RFid. Passive data collection tools – in particular, GPS devices – have proven useful in household and personal travel surveys, and have shown themselves capable of providing researchers with high-quality travel data. The objective of this research is to evaluate the use of GPS as a survey tool in household and personal travel surveys. Technological advances and decreasing costs have helped GPS to achieve wide use in the survey field. Furthermore, GPS-equipped devices allow surveyors to collect high-quality data on the time and position of individuals and vehicles – data that are more difficult to ascertain using traditional survey tools, such as self-administered questionnaires and telephonic interviews. A research team at the Politecnico di Torino designed and carried out a multi-instrumental personal travel survey, in order to assess the context-specific problems of a GPS-based survey in the metropolitan area of Torino. Survey methods included both a paper-and-pencil travel diary, and locational data collected using GPS devices. The survey effort consisted of a 4-day pilot survey with a sample of 4 individuals, and a successive 14-day GPS survey with a sample of 8 individuals. Results from self-administered travel diaries and GPS-derived data provided surveyors with valuable data for assessing the quality and completeness of travel information, and for determining the data’s ability to accurately describe respondents’ travel behaviour. The final outcomes of the GPS survey effort and of supplementary passive data collection tests allowed researchers to identify strengths and weaknesses of the implementation of passive data collection tools. Actual trends and future developments in the field will supplement the overview.