A static model for predicting disrupted network behavior

This paper compares actual and perceived travel times and presents a model for predicting traffic flows when there is a network disruption. The goal of this research is to demonstrate the necessity of accounting for possible differences in travel time perception and actual travel times, and also to show trends in how the route choices change based on the transformation of the perceived travel times. The model is separated into three phases: equilibrium assignment, link travel time transform, and logit assignment. The transform of the link travel times is best represented by an inverse cumulative Normal distribution, and the corresponding values provide quantifiable measure of the severity of a traffic network disruption. This methodology is then tested on a network to evaluate the resulting route choice patterns.