A semi-empirical model, with two adjustable parameters, has been developed for predicting the cloud points following the blending of diesel fuel components. The model is based on a kinetic argument deduced from the cloud-point dependence of the cooling rate. By either using a constant cooling rate or standardizing the cloud point to a constant cooling rate, blended cloud points can be accurately predicted from the equation [GRAPHICS] where T-j are the component cloud point (K), nu(j) are the component volume fractions, and T-c is the blended cloud point. The two adjustable parameters, alpha and beta are associated with the concentration of nucleating sites in the components. The contribution of the beta term to the prediction is small and is insignificant if component cloud points are evenly distributed. In general, the larger the cloud point the larger the number of nucleating sites, but this also appears to be dependent on the type of molecules involved in the nucleating sites.