Summary For a proton exchange membrane fuel cell (PEMFC), the pressure uniformity of fluid in the stack is a significant factor influencing the cell performance. This study employs the multi‐objective genetic algorithm (MOGA) to optimize the pressure uniformity in a PEMFC stack by adjusting the geometric design of inlet and outlet flow channels. The flow channel geometry is divided into three parts or named factors: tube length, tapered tube length, and channel height. The target is to find the minimum pressure difference between the inlet and outlet flow channels and optimize the pressure uniformity in the stack. The Latin hypercube sampling (LHS) method is used to organize the simulations for the design of experiments (DoEs), and the genetic aggregation (GA) method is employed to generate the response surfaces of the three factors. The resultant response surfaces are utilized to analyze the effects of the three factors on two objective functions, namely, pressure uniformity and pressure drop. The analysis of variance (ANOVA) method is also used to evaluate the influences of the factors, and the results are consistent with those obtained by the response surface method. The results show that the channel height produces the greatest impact on pressure uniformity. An increase in channel height can improve the pressure uniformity and reduce the pressure drop between the inlet and outlet channels. The MOGA analysis determines the optimal geometric design where the maximized pressure uniformity is 0.97 and the minimized pressure drop is 31 kPa. The number of cells in the stack is also taken into consideration. It is found that the influence of the inlet/outlet channel geometry on pressure uniformity is more pronounced when the cell number increases. The results benefit the design of inlet/outlet flow channels and the improvement of pressure uniformity in a PEMFC stack.