Short term prediction works in traffic engineering : the state-of-the-art.

In order to alleviate the worsening expressway congestion problem, dynamic traffic management (DTM) has aroused a lot of attention. Travel time provision, as an important ingredient of DTM and especially in the context of advance traveler information systems (ATIS), has become a common practice in many countries. Various technologies applying different sources of traffic data to estimate/predict travel time with high accuracy have been developed. This paper presents a state-of-the-art survey of traffic engineering related applications in prediction. The purpose of the paper is to provide: (1) a synthesis of published research in this area; (2) insights on prediction issues; and (3) the future research directions.