The upcoming change from fossil to biorenewable feedstock requires, among others, the identification of new liquid fuels to be used in mobile transportation. Although several molecules similar in structure to current petroleum-based fuels have already been proposed as biofuel candidates, a targeted search should identify molecules which are tailored to the numerous requirements posed by modern combustion engines. Such molecular design approaches strongly depend on reliable predictive models which describe the relation between the fuel's molecular structure and its combustion properties. To this end, we present here a targeted quantitative structure-property relation (TQSPR) for the laminar burning velocity, which is a fundamental indicator for the quality of fuels for spark ignition engines. We present the methodology and demonstrate its feasibility by the correct prediction of the laminar burning velocity of ethanol. We show that the experimental value is predicted well within measurement accuracy for various conditions. (C) 2010 Elsevier Ltd. All rights reserved.