The activity of Cu-based mixed oxide catalysts for oxidative steam reforming of methanol (OSRM) was investigated. In order to find a binary Cu-X mixed oxide catalyst with high methanol conversion, high H(2)selectivity, and low CO selectivity, an artificial neural network (ANN) was applied to relate the physicochemical properties of additive element X and the catalytic performance of the Cu-X mixed oxide catalyst. Experimental results of 14 Cu-X catalysts were used to train the ANN; the trained ANN predicted that Cu-Ca, Cu-Ce, and Cu-Pr oxides are the best candidates among all Cu-X binary catalysts. The same method using physicochemical properties and ANN was applied to find a good third additive for each binary oxide: (i) For the Cu-Ca system, any additive resulted in inferior catalytic performance. (ii) For the Cu-Ce system, Cu-Ce-Mn showed the best performance. (iii) For the Cu-Pr system, Cu-Pr-Ti was best. In the final step, ANN was also applied to improve the performance of the Cu-Pr-Ti catalyst by optimizing the catalyst composition and the preparation conditions (calcination temperature and total metal salt concentration). For the optimization, an L-9 orthogonal array, and a grid search (GS) were applied with the ANNs. The optimum Cu-Pr-Ti catalyst was: Cu/Pr/Ti = 58/16/26, calcination temperature = 623 K, total metal salt concentration in preparation step = I M. The catalyst showed good performance comparable with the best Cu-Zn-based catalyst ever reported: methanol conversion, H-2 selectivity, and residual CO concentration were 96%, 78%, and 510 ppm, respectively. (C) 2008 Elsevier B.V. All rights reserved.