The acid catalyzed transesterification of rapeseed and jatropha oil with 1-butanol was studied by varying temperature, reaction time, butanol/triglyceride ratio and the amount of sulfuric acid. A full factorial design supporting non-linear regression models was used to combine the four design parameters. The measured yield, viscosity and acid number were quantitatively described by PLS-regression as functions of the four design variables for each vegetable oil. Product purity was controlled by H-1 NMR. The PLS-regression model was used for optimizing the three response parameters. High (100%) yield is easily obtained within the experimental domain, whereas viscosity and acid number, with different optimum conditions, have to be compromised in order to keep both as low as possible. While rapeseed reached optimum within the experimental limits, the results indicate that for jatropha further reduction in viscosity and acid number requires increasing temperature and/or the butanol/triglyceride ratio beyond the upper limits of the design. (C) 2014 Elsevier Ltd. All rights reserved.