<p>This study optimized the design and operating parameters of proton exchange membrane water electrolyzers (PEMWEs) by using a computational approach. Initially, sensitivity analysis was used to identify optimal parameters, with liquid water saturation, hydrogen production rate, and ohmic resistance selected as performance indicators. Five parameters were then scanned parametrically to generate a large dataset for artificial neural network (ANN) training. Subsequently, the non-dominated sorting genetic algorithm with elite strategy (NSGA2) was employed to simultaneously optimize the three performance indicators, achieving values of 0.7839, 114.25 mΩ, and 3767.6 mol·s⁻¹·m⁻³, respectively. The optimized design exhibited a lower polarization curve than the original model, indicating reduced power consumption during electrolysis. The increased liquid water saturation enhances water transport and electrolysis efficiency; reduced ohmic resistance minimizes energy loss; and the higher hydrogen production rate supports improved fuel cell operation.</p>