The optimization of batch chemical reactors is strongly affected by the accuracy of the Tendency model used. In this work, the effect of this process-model mismatch on the optimization of the epoxidation of oleic acid using Tendency modeling is studied. The primary objective is to quantify the effect of the model’s uncertainty on process optimization so as to select the less uncertain between two Tendency models resulting far the experimental data. By placing confidence limits on the optimal operating policy and the expected performance index predicted by each model, it is possible to predict the poor performance that results when using a power-law kinetic model to optimize this process. The improved performance that results from using an alternate kinetic model, featuring inhibition kinetics, can also be predicted through this type of uncertainty analysis.