It is difficult and challenging to design high-performance fuel additives in an industrial-design setting where data are sparse and noisy, and fundamental knowledge is often limited. An automated framework is presented for the design of such fuel-additive molecules that minimize the intake-valve deposit in the automobile. A hybrid model that combined functional descriptors from a first-principles degradation model with a statistical/neural-network model was developed to predict additive performance, given the additive structure. The results of the predictive model are discussed for differential industrial case studies. An evolutionary method using specialized representation and constrained operators to enforce formulation constraints was used to generate optimal additive molecules that meet desired performance criteria. The evolutionary design strategy in combination with the hybrid prediction model was successful in identifying novel additive molecules that also possess good synthesis potential.