Accurate estimation of water needs manages both crop yield and water loss occurring due to imprecise water supply to the crops. In this context, this paper proposes an artificial neural network (ANN)-based model to estimate reference evapotranspiration (ET₀) which is crucial in deciding the water needs of a crop. The Penman–Montieth (PM) method is considered as a benchmark by the Food and Agriculture Organization of the United Nations (FAO), but it lacks usage in deployments due to its heavy input requirements. The model presented in this paper targets to mimic the PM method and succeeds in obtaining the same results using minimum input variables. Corresponding results have been mentioned, which show that an infinitesimally small error of a maximum 0.4 mm/day is present in the output predicted using the proposed ANN model. It is found that the predicted ET₀ has no ill-effect on the future computations.