Diesel fuels are complex mixtures of thousands of hydrocarbons. Since modeling their combustion characteristics with the inclusion of all hydrocarbon species is not feasible, a hybrid surrogate model approach is used in the present work to represent the physical and chemical properties of three different diesel fuels by using up to 13 and 4 separate hydrocarbon species, respectively. The surrogates are arrived at by matching their distillation profiles and important properties with the real fuel, while the chemistry surrogates are arrived at by using a Group Chemistry Representation (GCR) method wherein the hydrocarbon species in the physical property surrogates are grouped based on their chemical classes, and the chemistry of each class is represented by using up to two hydrocarbon species. The developed surrogate models were applied to predict conventional and low temperature combustion (LTC) characteristics of the three fuels in a single cylinder diesel engine using the KIVA-ERC-CHEMKIN code incorporated with a MultiChem mechanism having 120 species and 459 reactions. The predictions compared well with the measured data and engine trends with the three fuels with cetane numbers ranging from approximately 40 to 57. To examine the advantages of using the present multi-component models, the results were also compared with a conventional single component model, viz., n-tetradecane representing physical properties and n-heptane representing chemistry. It was found that although the single component fuel model reproduces the combustion and emission characteristics of the three diesel fuels under conventional combustion operation, it was unable to predict combustion and emission processes in LTC operation, especially, for lower cetane diesel fuels. It was also observed that there was a lack of sensitivity to changes in fuel properties in conventional combustion due to favorable charge conditions in terms of higher oxygen fraction and higher temperatures as compared to LTC. Copyright © 2013 SAE International.