OFFM-ANFIS analysis for flood prediction using mobile IoS, fog and cloud computing

Flood is one of the most destructive natural disaster that can affect hundreds of thousands people and infrastructure that can cost in the range billion of rupees. After so much research in this field, we still suffer from the disastrous effect of it due to lack of consideration of small scale flood affecting parameters affecting regional floods. In this research paper, an IoS based sensing network powered by mobile edge computing (MEC), fog computing and cloud computing is proposed for flood prediction and forecasting after analysis through modified multi-ANFIS architecture known as OFFM-ANFIS. The sensing layer includes various static and mobile sensors IoS nodes that pass the data to fog server via MEC. Both regional fog server and cloud is facilitated with the training, testing, analysis and decision making power that rates the chances of flood on an ordinal scale. The OFFM-ANFIS includes seven modified ANFIS models that make decision on flood forecasting on the basis of trained data and analysis of received senses data. The flow of raw and analyze data from OFFM-ANFIS is staged in a way that more influential parameters have strong impact in the final forecasting output. Evaluation of the proposed mechanism is done on the basis of data provided by Indian Meteorological Department and effectiveness is shown in the application of proposed mechanism in forecasting flood well before its occurrence.