Drought events have an impact on various sectors such as the economic, agricultural, environmental and social sectors. The impact of drought needs to be minimized by designing a drought prediction system, this is an effort to disseminate early warning information based on climate and hydrological aspects. Drought prediction uses Multi-Layer Perceptron (MLP) algorithm model to predict drought based on Standardized Precipitation Index (SPI) for 1 and 3 months. SPI is one of the drought indices calculated through rainfall analysis. Predictions were made using rainfall data obtained from the FY-4A QPE satellite, then processed into monthly rainfall accumulations for the 2019-2020 period. FY-4A QPE data is corrected based on rainfall observation data at the observation rain gauge. The corrected FY-4A QPE data is used to predict drought using the SPI1 (1 Month) and SPI3 (3 Months) indices. SPI1 shows fluctuations and a very small Nash-Sutcliffe Efficiency (NSE) value of -0.11, while SPI3 shows a model that is able to follow the peak and valley fluctuations of the actual value as well as an increase in the NSE value of 0.65 and a decrease in the Root Mean Square Error (RMSE) value of 1.43 in SPI1 to 0.77 in SPI3. Drought modeling using MLP was successfully implemented and it showed that as the time span increases, the SPI prediction performance gets better. © 2024 IEEE.