<![CDATA[Weather prediction, particularly rainfall potential, plays a crucial role in supporting airport operations to minimize the risk of flight disruptions, particularly in coastal areas prone to extreme weather. This study aims to provide practical solutions for airport managers in anticipating operational disruptions due to rainfall by utilizing historical data and probabilistic models for weather condition classification. Rain potential prediction is performed using radiosonde index data representing atmospheric lability conditions, namely the Lifted Index (LI), K-Index (KI), Showalter Index (SI), and Total Totals Index (TT) as predictor variables. Data were obtained from the Tunggul Wulung Cilacap Meteorological Station database for the 2020–2024 period, then used in the model training and evaluation process by handling missing values and measuring performance using a confusion matrix. The results show that the Naïve Bayes machine learning method is capable of producing a rainfall potential prediction model with an accuracy of 71.56%. These findings are expected to support more timely and efficient decision-making and have the potential to be applied in an early warning system based on actual observations for disaster mitigation.]]>
Copyrights © 2026
