Field Deployment to Quantify the Value of Real-time Information by Integrating Driver Routing Decisions and Route Assignment Strategies

Advanced Traveler Information Systems (ATIS) have been proposed as a mechanism to generate and distribute real-time travel information to drivers for the purpose of improving travel experience represented by experienced travel time and enhancing the performance of the vehicular traffic networks. From the system operator’s perspective, it is desired that a driver would fully comply with such information. Traditionally, the prediction of system performance under real-time information provision has been studied using dynamic traffic assignment (DTA) models in which individuals are assigned to time-dependent routes from their origins or en-route locations to their destinations so as to satisfy some system-wide objective and/or individual user level constraints. However, these models primarily focus on prescribing the traffic flow propagation robustly, while the role of driver behavior in the evolution of network dynamics has largely been subsumed by making potentially restrictive a priori assumptions, which include one or more of the following: (i) travel time is the only basis for route choice decision-making, (ii) users are behaviorally homogeneous, and/or (iii) pre-specified behavior classes are available whose fractions are known in the ambient traffic stream. In this context, a comprehensive modeling framework is proposed to understand individual drivers’ behavioral responses in route choice under real-time travel information provision based on driving simulator experiment data. An interactive driving simulator experiment is developed to collect various data related to driving and decision-making with real-time travel information. The associated surveys are also precisely designed to measure drivers’ perception of the information and evaluation of the travel experience.