A neural network based prediction model for flood in a disaster management system with sensor networks

A disaster management strategy may be divided into two sequential phases, namely, pre-disaster management and post-disaster management. Prior to a disaster, management activities are pre-disaster planning, and disaster prediction. A good disaster prediction technique plays a crucial role in an efficient mitigation of disasters such as flood. In this paper, we have proposed a flood forecasting technique that is based on an artificial neural network (ANN) model, namely, multi-layer perceptron (MLP). We have shown the relative importance of different environmental parameters used to predict flood and it is found that underground water level is the most significant parameter for the prediction model. We have also shown that the proposed technique produces a statistically significant forecasting result in the test data set.