An Artificial Neural Network Model for Traffic Noise Predictions

The major environmental challenge encountered by metropolitan city today other than air pollution is traffic noise. So urban planning needs methods to aid in designing, planning, and forecasting in order to accommodate the increasing population and increasing traffic noise levels. Since the problem of traffic noise is nonlinear in nature, a model based on backpropagation neural network to counter this problem is suggested and examined. In order to have a clear and distinct insight on the magnitude of this problem, single-noise metrics L _Aeq is modeled. It is observed that the model is accurate in predictions and can be employed efficiently to predict traffic noise levels and conduct sensitivity analysis of factors affecting the noise levels. It can serve as an important tool for urban planning and development.