Regression based neural network model for prediction of road traffic congestion : A case study of Bhubaneswar

Abstract

The prediction of road traffic congestion is the most important and essential aspect to reduce the suffering of population of urban cities which is primarily carried by roads. The lack of a traffic congestion data unavailability and evaluation standard makes the effect of traffic congestion prediction more difficult and worsen. Traffic congestion occurs due to increase in number of vehicles on roads which reduces speed of vehicles, increases delay time, and increasing vehicular queuing in traffic. Due to traffic congestion, not only delaying time increases but also its threat to slower down the economic growth rate of our country and also have high impact on our personal growth, living condition with high level of pollution and undesirable feature of overloaded streets. Traffic congestion predicting modeling plays very important role so we need a innovative approach to predicting the congestion on roads. In this paper, we aim to provide a model which studies the real time environment characteristics of the road, and analyzed the data, congestion location identification, directional movement of all those locations surveyed and forecasting traffic location where traffic congestion may occur in near future.