Toward Management and Imputation of Unavailable Data in Online Advanced Traveler Information Systems

Advanced traveler information systems (ATIS) deliver information derived from intelligent transportation systems data. ATIS continue to emerge and grow in popularity, as some of these systems are public and freely available. ATIS performance and perceived usefulness depends, among several design factors, on the reliability and quality of its data sources, especially when the systems are reporting current conditions. We present a sample scenario in which an inductive loop detector station along a freeway stops transmitting data, and an ATIS service presents an estimated state of the system instead of showing that no data is available. The estimation is based on sources of available and correlated data. Our first approach to implement an estimation framework consists of deriving a parametric description of the correlation from archived historical data. An example application is provided as well as suggestions for future research.