India is a country highly dependent on rainfall for its various activities which include agriculture, businesses and other direct metrics which directly affect the economy. The rainfall forecast is not only beneficial in examining the fluctuating patterns of rainfall, but it will also help in organizing the safety measures in case of catastrophe and its management. The altering patterns of rainfall lead to global warming; which is rising the earth’s temperature due to an increase of Chlorofluorocarbons that are emitted from air conditioners, refrigerators, deodorants, printers etc. producing melting of the snow caps, altered weather patterns have become a significant part of everyone’s life. The rainfall for a particular location is more dependent on the different factors like, relative humidity, specific humidity, temperature, wind speed, wind direction surface pressure, evaporation, transpiration, percolation, runoff, ocean currents, population, El Nino and La Nina etc. Rainfall forecasting or predicting is very vital because extreme and uneven rainfall can cause many effects like demolition of crops and farms, and damage of property. So, for rescuing the ecosystem, a better forecasting or predicting model is essential for a quick warning that can diminish the risk causing to the life and property and also for efficiently managing the agricultural farms. This project aims to create a distinctive and effective mathematical model and machine learning system for rainfall prediction. In this study, various rainfall metrics from Osman Sagar catchment of Gandipet mandal, Telangana are tested in order to measure the model's effectiveness and perseverance. This study focuses on the development of a mathematical model with the help of multi-linear regression analysis (MLRA) and artificial neural network (ANN) models. The data learning is performed using a hybrid and back propagation network approach. The relationship between the dependent and independent parameters of rainfall have been demonstrated graphically. The MLRA and ANN models are used in this study to achieve sustainable results through training and testing data. The accuracy of the models is checked by comparing the model's monthly rainfall forecasts with real data after training and testing. The study's findings show that the model is effective in forecasting the data of monthly rainfall with the specific parameters and statistical error analysis has been performed to evaluate the strength of the models. Finally, the developed model has been validated for the year 2022 and the future prediction has been done for the year 2025 and 2030.