Network Route-Choice Evolution in a Real-World Experiment: Necessary Shift from Network- to Driver-Oriented Modeling

Route choice models are a cornerstone in many transportation engineering applications. Two main types of route choice models can be found in the literature: first, mathematical network-oriented models such as stochastic user equilibrium and, second, behavioral driver-oriented models such as random utility models. Although the former models are much more widely used in the transportation engineering realm, evidence of their inadequacy is growing continuously. The degree of their inadequacy, however, remains debatable. Two major criticisms of the theory are the unrealistic assumptions of human perceptions and the inability to incorporate driver heterogeneity. However, attempts to incorporate driver heterogeneity into the behavioral driver-oriented route choice models, too, are still inadequate. Another major limitation in the literature is that because of cost and past technological limitations, only a few studies are based on real-world experiments. Most studies are based on either stated preference surveys or travel simulators. This work analyzes results of a real-world route choice experiment with a sample of 20 drivers who made more than 2,000 real-world choices. Network and driver learning evolutions were recorded and analyzed. Findings of the experiment include the following: (a) with learning and network experience, real-world route choice percentages seem to be converging to specific values; however, these values are mostly different from those derived by using stochastic user equilibrium expectations; (b) four types of heterogeneous driver learning and choice evolution patterns are identified; and (c) the identified learning patterns are modeled and found predictable on the basis of driver and choice situation variables.