TRAFFIC PARAMETER-ESTIMATION AND HIGHWAY CLASSIFICATION - ROUGH PATTERNS USING A NEURAL NETWORKS APPROACH

Neural networks provide more accurate estimations of traffic parameter s than conventional methods. This paper explores the possibility of us ing more sophisticated neural networks based on rough patterns for inc reasing the accuracy of estimations. A rough pattern is represented by upper and lower bounds of the input values. The paper compares four d ifferent data collection schedules and two different types of neural n etwork architectures for estimations of average and peak traffic volum es as well as classification of highways.