Route Choice models have long been based on the assumption that travelers choose the shortest travel time path between their origin and destination, despite increasing evidence that is not the case. Robust models based on empirical data of how travellers do select routes remains an open question. This research will use empirical data on actual route choices from GPS and econometric methods to better predict actual routes selected.
This research will analyse existing GPS data sets of traveler routes between homes, workplaces, and other destinations and use statistical (econometric) and machine learning methods to develop empirical models of route choice. The models will be applied to large networks and compared with traditional equilibrium models and observed data. Datasets on highway and bicycle routes are available.
The opportunity ID for this research opportunity is 2226