We propose a machine learning approach based on the crop simulation model WOFOST to assess the crop yield and water use eﬃciency. In our research, we use weather history to evaluate various weather scenarios. The application of multi-criteria optimization based on the non-dominated sorting genetic algorithm-II (NSGA-II) allows users to ﬁnd the dates and volume of water for irrigation, maximizing the yield and reducing the total water consumption. In the study case, we compared the eﬀectiveness of NSGA-II with Monte Carlo search and a real farmer’s strategy. We showed a decrease in water consumption simultaneously with increased sugar-beet yield using the NSGA-II algorithm. Our app-roach yielded a higher potato crop than a farmer with a similar level of water consumption. The NSGA-II algorithm received an increase in yield for potato crops, but water use eﬃciency remained at the farmer’s level.NSGA-II used water resources more eﬃciently than the Monte Carlo search and reduced water losses to the lower soil horizons.