The equations that govern the various transport phenomena occurring in a polymer electrolyte membrane fuel cell (PEMFC) were formulated and implemented in a commercial finite element software, in order to predict the fuel cell current density with respect to the operating conditions. The numerical model showed polarization curves in accordance with literature. The catalyst utilization was then improved by optimizing the platinum distribution (design variable) in the fuel cell, so as to maximize current density (objective function) for a fixed total amount of platinum (constraint). The first analysis showed that, for equal anode and cathode catalyst layer thicknesses, maximal current density was achieved by placing more catalyst in the cathode than in the anode. The second analysis showed that, for equal anode and cathode catalyst layer density, maximal current density was achieved by using a catalyst layer that is thicker on the cathode side than that on the anode side. Finally, a topological optimization of the platinum density within the cathode catalyst layer was performed with a gradient based algorithm, and the results showed that at a high stoichiometric ratio, the best design has most of its platinum placed where the reaction rate is the highest, i.e., close to the membrane layer. Copyright (C) 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.