Abstract Manager

Arguably one of the many strengths of the family of travel demand models generally referred to as activity-based models is that they provide a behaviourally sound computational framework for modelling trip mode choice within an explicit tour-based context. A tour (also referred to as a trip-chain) is defined as a connected sequence of trips use first origin and last destination are the same point; i.e., the sequence of trips forms a closed loop that begins and ends at the same point. This is usually the trip-maker’s home (i.e., a home-based tour), although other tour anchor points, notably work or school locations may also exist.

Adopting a tour-based approach to mode choice is very important, since the choice of mode for one trip is often contingent upon the mode, purpose and/or location choices of other trips on the same tour. Most notably, the decision to use the auto-drive mode for one trip on a tour necessarily requires that the auto be used on all other trips on the tour – if the car leaves the driveway, it must, inevitably, return to the driveway at the end of the tour. Trip-based models, in which the mode choice of each trip is determined independently of all other trips quite obviously can lead to illogical predictions (e.g., use the car to drive to work and then use transit to return home). A similar argument holds for bicycles (in the absence of a bikeshare option, in which the tour constraint may be broken). Similarly, if one was only going to and from work (a simple home-work-home) tour, one might use choose transit as the best mode for both trips. But if one wishes to shop on the way home from work, at a location that can only be reasonably accessed by car, then, again the car will need to be used to go to work, even though, in isolation, transit may be the preferred mode. Thus, tour structure, which emerges for the person’s desired daily activity pattern, conditions individual trip mode choice.

Operational tour-based mode choice models have become increasingly common in recent years. This paper briefly reviews and critiques common structures for tour-based mode choice models. It then presents in detail the TASHA tour-based model which is being employed in numerous Canadian and international urban regions.