Over the last thirty years, there has been progress in reducing the infant mortality rate in Gujarat; nevertheless, the state is anticipated to not meet the Sustainable Development Goal (SDG) target for infant mortality. This necessitates increased efforts to tackle this challenge. This research used infant mortality data obtained from the Sample Registration System (SRS), the World Bank, and various reports to evaluate the situation in Gujarat. The study identified trends in infant mortality and developed predictive models to provide an accurate forecast of the mortality rate in the state. Two time series models were examined and assessed: the Autoregressive Integrated Moving Average (ARIMA) and Neural Network Autoregression (NNAR). The comparison of these models was conducted based on five different metrics: Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Percentage Error (MPE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE). Both models revealed similar trends and were considered appropriate for forecasting; however, the NNAR model surpassed the ARIMA model in all five-evaluation metrics. The analysis shows a declining trend in the infant mortality rate in Gujarat, with the NNAR model projecting a reduction of roughly 5-6% in the rate over the next decade