A Hybrid ARIMA-Intervention Modelling for Forest Fire Risk in The Dry Season

This study explores the time-related patterns of forest fires and assesses the impact of measures implemented during the dry season. Special focus is directed towards the effects of these interventions on the frequency and intensity of fires. This study highlights the importance of combining temporal analysis with spatial data to identify high-risk locations and optimize resource allocation for fire prevention. This study develops an ARIMA model to forecast fire risk before intervention. The findings indicate that integrating intervention factors into the ARIMA model will enhance the model's accuracy. The satisfactory MAPE values and the value data plots effectively demonstrate the data patterns. This method establishes a solid basis for predicting and reducing the risk of forest fires in the dry season, thereby enhancing the fire resilience of ecosystems considered at risk. The findings indicate that the onset of the dry season significantly elevates the risk of forest fires, especially in areas near bodies of water.