It is difficult task to predict the trend of precipitation meteorology and environmental sciences. The present study considered the monthly rainfall from 1960 to 2016 obtained from Bangladesh Meteorological Department (BMD).To examine Rainfall in Bangladesh and find a suitable model for forecasting is the main intention of this study with the help of SARIMA Approach.Original data plot shows that it is stationary and then tests Auto-Correlation Function (ACF), Partial Auto-Correlation Function (PACF). After taking first difference original data was transformed to stationary. Kwiatkowski-Phillips-Schmidt-Shin (KPSS) test with p-value of 0.1and Augmented Dickey- Fuller (ADF) test with p-value of 0.01 proved the stationarity of Rainfall data. On the basis ofleast Akaike Information Criterion (AIC)value we suggest SARIMA (0,0,0) (2,1,1)12 model is the best to predict monthly rainfall data. After diagnostic checking on SARIMA (0,0,0) (2,1,1)12 model, the Auto Regressive parameter was found to be statistically insignificant and SARIMA (0,0,0) (2,1,1)12 model that best fit and was used to forecasting 120 months (January2017- December 2022) seasonal Rainfall in Bangladesh.