A modeling approach to predict the performance of a commercial diesel oxidation catalyst (DOC) under actual vehicle operating conditions is presented in this study. Prior to completing this prediction, the performance characteristics of DOCs, as previously published, are examined to validate the currently developed in-house computational code. Steady-state experiments with DOCs mounted oil a light-duty four-cylinder 2.0-L turbocharged diesel engine then are performed, using an engine-dynamometer system to calibrate the kinetic parameters Such as activation energies and pre-exponential factors of heterogeneous reactions. The reaction rates for CO, HC, and NO oxidations over a fresh PtIAl2O3 catalyst are determined in Conjunction with a transient one-dimensional (ID) heterogeneous plug-flow reactor (PFR) model with diesel exhaust gas temperatures in the range of 150-450 degrees C and space velocities in the range of (1-5) x 10(5) h(-1). To determine the kinetic parameters that best fit the experimental data, a two-step optimization procedure is introduced. First, the results from the conjugated gradient method (CGM) with individual temperatures for each species are plotted in ail Arrhenius plot to generate proper intermediate guesses from initial guesses for all pre-exponential factors and activation energies. Here, the kinetic parameters for CO oxidation are calibrated to provide the best fits to the lowest temperature data with fixed initial activation energy without implementing the first-step tuning procedure, because of complete conversion of CO over the temperature range of 150-450 degrees C. Kinetic parameters for all species then are obtained simultaneously by searching the best fits to experimental data using the CGM from the intermediate guesses for all species. The prediction accuracy of the model through first step optimization procedure against experimental results is slightly improved by performing a second-step optimization procedure, and the simulation results of optimized kinetic parameters are compared to the experimental data obtained at both 1500 and 2000 rpm.