Fog is a complex meteorological phenomenon that poses significant challenges for accurate prediction, particularly because of its nonlinear dependence on a number of atmospheric variables, including such as air pressure, temperature, humidity, and wind speed. Traditional statistical models often struggle to capture these intricate relationships, leading to suboptimal fog forecasts. In order to improve the accuracy and interpretability of fog prediction, this paper investigates the use of machine learning models, particularly random forest and decision tree classifiers, which are improved by explainable artificial intelligence (XAI) techniques such as SHapley additive exPlanations and local interpretability model-agnostic explanations, are employed. By analyzing a comprehensive dataset of historical meteorological data, this study demonstrates how these models can effectively predict fog events and elucidate the key factors driving their formation. The integration of explainable AI methods provides critical insights into these models’ decision-making procedures, improving their transparency and dependability in practical applications. The results suggest that the suggested methodology not only improves forecast precision but also provides a more profound comprehension of the fundamental meteorological processes, rendering it an advantageous instrument for industries significantly affected by fog, including aviation and transportation.