The results of a durability test performed during 1000 It on a proton exchange or polymer electrolyte membrane (PEM) fuel cell are analyzed using some well-suited design of experiment techniques. The response surface methodology is adopted to model the performance degradation over ageing time front various load current-fuel cell voltage curves recorded at regular time-spaced intervals and for various air utilization rates. The dual response surface approach is employed to determine the most convenient operating conditions for the cells (load current and air stoichiometry rate levels), leading to a tradeoff between elevated electrical efficiency and low voltage variability versus ageing time and cell positions in the stack. In addition, some electrochemical impedance spectra are used to provide some physical interpretations and to corroborate the observations made from the static measurements. The work shows notably the benefit of using some variable stoichiometry rates through ageing time to get high-power levels and stable performances as well.