Accurate assessment and optimal management of water resources are crucial in reservoir operations, where uncertainties necessitate the use of advanced methodologies. This study addresses data uncertainty in dam operations by employing classical fuzzy logic, contrasted with the crisp method optimized via genetic algorithms (GA), using Iran’s Alavian Dam as a case study. Performance metrics—reliability, vulnerability, and sustainability—were evaluated. Results demonstrated that classical fuzzy logic outperformed the crisp method, satisfying 89% of downstream demand (vs. 77%) by integrating expert insights and AI-driven inflow forecasts. The LUBE (Lower and Upper Bound Estimates) method, utilizing an ensemble AI model, generated rule curves that confined actual dam releases within predicted bounds. The findings underscore classical fuzzy logic’s efficacy in reconciling operational uncertainties and enhancing decision-making precision in dynamic reservoir environments.