Forecasting of Traffic Jams at Disturbed Sections of High-ways

Igor Grabec

Abstract

Disturbances of driving conditions on high-ways usually lead to evolution of traffic jams. Disturbances are often caused by traffic accidents, installed bottlenecks, adverse weather, etc, and result in a decreased road capacit. By using an estimate of a desired speed in the disturbed section the traffic information providers can forecast quantitatively the evolution of jams and inform the population about them in advance. The article presents a new mathematical method for this purpose. The corresponding intelligent unit first forecasts the traffic flow at a disturbed road section based upon records of traffic flow in the past. Forecast data are next mapped to characteristics of evolving jam by using the desired speed value and a new fundamental diagram of traffic flow. Performance of the method is demonstrated by forecasting the evolution of jam at the point of maximal traffic activity on a high-way in Slovenia.

Relevant Publications in Applied & Computational Mathematics