Graphical comparison of predictions for speed given by catastrophe theory and some classic models

Using a common data set, the speed estimates given by catastrophe theory are compared with those of some classic models. After a new proposed set of transformations to real traffic flow data, which include translation and rotation of axes, catastrophe theory is applied to information from the Queen Elizabeth Way in Ontario, Canada. The catastrophe theory model is used for predicting speeds on the basis of information on volume and occupancy. Several proposed models and a double-linear-regime model are also used for predicting speeds. The different estimates for speed are graphically compared with the observed values. A coefficient of determination (R²) is given as the measure of reliability for all the models. The results show that the catastrophe theory model performs better than the other models.