Effect of Information Quality on Compliance Behavior of Commuters Under Real-Time Traffic Information

Abstract

The compliance behavior of commuters under different advanced traveler information systems (ATIS) information strategies of varying information quality and credibility is investigated. A behavioral experiment involving commuters’ departure time and route decisions in response to traffic information in a simulated traffic system is conducted by using an interactive dynamic multiuser, computer-based simulator. By actually simulating traffic conditions in response to commuter decisions, the simulator provides stimuli to the participants that are always consistent with physically realistic traffic behavior and with their previous decisions. Event-count (frequency) models and dynamic, discrete-choice models are employed to capture the principal effects of the commuters’ experience with the traffic system and with the real-time information on user compliance. The analysis reveals that the information quality of ATIS strongly affects commuters’ compliance behavior. Switching “cost” and recent experience are found to influence their trip-making behavior as well. The analysis further shows that a higher compliance rate can be achieved under prescriptive information than descriptive information. A hierarchy of achievable compliance rates appears to exist in response to the type of information supplied, under which the highest compliance is obtained for reliable, predicted information and the least compliance for random information. Moreover, by providing posttrip feedback with performance information on recommendation or ex–post-best decision, higher compliance is attained.