Since, the dawn of time, flooding is the most frequently affectedoccurring natural disastersdisaster in the world. There are numerous publications with applicable to flood forecasting for enhancing flood rescue operations. It may be the major danger to each culture across the planet. 95% Asian and African Combined floods [1]. A minimal loss of $605 million was suffered by theflood disaster in 2007 [2], [3]. Due to its nature, flood prediction is extremely complex and vastly non-linear time-series;. Time-series forecastforecasting developed diverse stages formfrom the start formof the late 70s till today, there. There are numerous Artificial Neural Network (ANN) used for solving non-linear time seriestime-series such as the Floodflood prediction clues tofor risk management and reducesreducing the natural disaster hazard Indexindex [5], [6]. The research carriescarried out in 2015 modelling flood water level prediction using Neural Network Auto regressive with Exogenous Input (NNARX) has proof thatproved 7 hour predictions with low Root Mean Square Error (RMSE) withof 0.0059 m and over 85% confidence onon an auto correlation plot of 3 hours NNARX prediction model [7].