The under-five mortality rate (U5MR) is an important measure of child well-being, which includes health status, and, more generally, of social and economic growth.. This paper presents an empirical model to forecast the future under-five mortality rates in India. The predictive models in the time series components are employed to forecast the U5MR in India. In this study, the training dataset was utilized to develop a model, and the test dataset was used to evaluate its performance. The performance was determined by computing error measures such as Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE). The results of both models demonstrate that the Feed-forward Neural Network (FFNN) model has superior performance compared to the Auto Regressive Integrated Moving Average (ARIMA) model