Phase change studies have been heavily researched due to their potential applications in many sectors such as heat exchangers, evaporators and condensersin air conditioning devices, oil and gas industries, and nuclear reactors. When a very hot surface quenched in a water pool, an undesired vapor layer isformed around it. This vapor layer acts as insulation that reduces heat transfer between the surface and the water pool. At a temperature called the minimumfilm boiling temperature (Tmin), the vapor film collapses therefore the water cools the surface dramatically during the direct physical contact. This explainsthe importance of this temperature in enhancing the heat transfer rate. In this work, a prediction of Tmin values for various substrate rods quenched in eitherhigh- or low-pressure distilled water pools is applied by an artificial neural network (ANN) technique. The ANN was trained using 379 experimentaldata collected from literature. Length to diameter ratio (L/D), fluid to the substrate material thermophysical properties (ρkcp), system saturated pressure(Psat) and liquid subcooling temperature (Tsub) are used as inputs, whereas Tmin is considered as the output. The number of neurons and hidden layers aredetermined based on the accuracy of results. The trained ANN is able to predict the experimental data with a mean absolute error (MAE) of approximately6%, and a determination coefficient (R2) greater than 0.91 for all data, using a configuration of 12 neurons within 3 hidden layers. The obtained resultsare within acceptable margin of error of all data. Future work will consider comparing ANN to the well-established literature correlations for Tmin demonstrates how important this tool is to predict Tmin accurately.