Indonesian wildfires are disasters that occur annually, inflicting substantial ecological and economic devastation, jeopardizing public health, and even straining international relations. In response to these escalating challenges, this study explores the development of data-driven disaster management strategies that prioritize proactive wildfire prediction. To address the growing need for effective wildfire mitigation strategies, this study investigates the application of machine learning for wildfire occurrence prediction. Machine learning can handle medium complex data and adapt to unforeseen circumstances, such as changes in environmental conditions, makes it ideal for developing robust wildfire prediction models. This study aims to develop an accurate prediction model for wildfire occurrence that considers natural factors, such as climate and peatland vegetation. The prediction model was built using supervised learning algorithms, namely K-Nearest Neighbors (KNN), Decision Tree (DT), and Random Forest (RF), which were evaluated using metrics accuracy, sensitivity (recall), precision, and F1-Score. The prediction model results with Random Forest showed the best performance with an accuracy of 98.82%, sensitivity of 93.18%, precision of 88.26%, F1-Score 90.22%, and an AUC score of 97.43%. The temperature has the greatest influence feature on wildfire occurrence in Indonesia. © 2024 IEEE.