The principal objective of this article is to examine, compare, and develop a time series model for forecasting the growth of the life insurance business in Thailand. The proposed forecasting models include the Seasonal Autoregressive Integrated Moving Average Exogenous (SARIMAX) model, the Multilayer Perceptron (MLP) model, and a combined model that integrates SARIMAX and MLP models. The dataset, spanning from January 2003 to December 2022 and comprising 240 rows, is sourced from the Office of the Insurance Commission (OIC) website, the Office of the Economic Development Council, and the National Society website, providing gross domestic product (GDP) data. This dataset is partitioned into two subsets: the first subset encompasses the years from January 2003 to December 2021, utilized for constructing predictive models, while the second subset consists of data from January 2022 to December 2022, employed for comparing the accuracy of various forecasting methods. Evaluation criteria such as the Mean Absolute Percentage Error (MAPE), Root Mean Square Error (RMSE), and the R-squared are used for assessment. Our analysis, conducted through Python in Google Colab, demonstrates that the Multilayer Perceptron (MLP) model consistently outperforms both the SARIMAX and MLP-SARIMAX models.