Towards longitudinal activity-based models of travel demand

Existing activity-based models of travel demand consider a day as the time unit of observation and predict activity patterns of inhviduals for a typical or average day. In this study we argue that the use of a time span of one day severely limits the ability of the models to predict responsive behavior especially at the level of activity frequency and time-use variables. The only way to overcome this shortcoming is to extend the time frame from one-day to a multiweek period. Although this idea is not new and first steps have been made in this direction, there does not seem to be a coherent view of scope and objectives that would help to define the area of research and set out a research agenda. The purpose of this study is to fill this gap and propose such a view.