Route choice is determined by some function of mean travel time and distance on the routes available in most traffic assignment models. Increasing traffic volumes on a route increases delay, making a particular route less desirable.
The NZ Transport Agency (2010) Economic evaluation manual allows the benefits of improved network reliability to be monetised. However, our network models are unable to provide a convenient means of calculating the road user responses to travel time variability.
Route choice is a more complex issue than a comparison of relative travel times and distance. It appears that road users are also considering travel time variability in their route choice. Variability may occur as a result of congestion in cities or on any network as a result of road geometry, a high volume of heavy vehicles on narrow steep roads, or other reasons.
This research was carried out during 2008 to 2011 using Wellington data and sought to identify a methodology that best incorporated travel time variability into route choice models. The research determined the most useful formulation for use in models and the appropriate measure of travel time variability.