Prediction of short-term traffic volume for applications in intelligent transportation systems

Intelligent Transportation Systems (ITS) can play a major role in resolving the traffic congestion problem faced by today's highway systems. Successful implementation of ITS depends upon the accuracy of estimates of current and near-term expected traffic situation. Effective short-term road traffic forecasting techniques and collection of realtime traffic data assume a significant role in this context. This study presents the results of traffic prediction models developed using multiple regression analysis (traditional statistical method) and neural network technique (intelligent modeling technique) to predict the next hour traffic volume for Highway No. 2 between Calgary and Edmonton in Alberta, Canada.; Most of the past research on short-term traffic volume forecasting has focussed mainly on intra-urban traffic situations. The present study addresses the problem pertaining to a traffic corridor of inter-urban nature and also prediction models using more number of input variables. The traffic volume patterns for Highway No. 2 between Calgary and Edmonton are analyzed in detail at four locations at Airdrie, Olds, Red Deer and Leduc.; Traffic prediction models using multiple regression analysis and artificial neural network (ANN) technique are developed to predict the next hour traffic volume (between 7 a.m. and 8 p.m.). (Abstract shortened by UMI.)...