Value of Information in Bayesian Routing Games

Abstract

Over the years, travelers are increasingly relying on navigation services to inform their route choices and save time. How much value—in terms of saved time—can these services indeed provide to their users? The answer is not as simple as one may think because the value of traffic information not only depends on its accuracy, but also on how widely it is shared among travelers. For instance, information on an incident has the highest value for a traveler when the traveler is the only person who knows about it and takes a detour. However, this information becomes less valuable when it is shared with more travelers, who may take the same detour and cause congestion. Our work uses a game-theoretic approach to analyze how the relative value of information provided by different navigation services changes with their market share and how travelers may choose from different available services.