DEVELOPING TOUR-BASED DATA FROM MULTI-DAY
GPS DATA
Yun Zhang1, Peter Stopher1, Qingjian Jiang2
1Institute of Transport and Logistics Studies, 2The University of Sydney, and, Parsons
Brinckerhoff, Australia
ABSTRACT
The purpose of the research reported in this paper is to understand travel patterns by applying
tour-based analysis and using sociodemographic variables to characterise travel patterns to explore
new opportunities of developing activity-based and tour-based models. The data used in this
research is from an Australian panel where 200 households provided GPS data for a period of 7 days
with a small sub-sample (43 households) for 28 days, with a total of 388 persons. This paper
presents the results of tour analyses of the above data, which include the distribution of tours
per day and the trips per tour, the distribution of tour duration and the starting times, followed
by a summary of important considerations when dealing with tour-based data. We further introduce an
extended tour classification, using twelve tour types based on a hierarchy of trip purposes of
work, education, shopping, and other. With the application of the new tour classification, we
present findings concerning the composition of the tours (simple or complex tours) and
sociodemographic characteristics, such as employment or education status and the stages in the
family life cycle.