A combustion model to simulate the oxidation of multi-component fuels has been developed and applied to model autoignition combustion processes. The model is called the Physical Surrogate Group Chemistry Representation (PSGCR) method, which combines a multi-component fuel model to describe the physical properties of the fuel with reduced reaction mechanisms to represent the chemistry of the fuel components. A group of physical surrogate (PS) components are used to describe the fuel's physical properties, such as the distillation profile, specific gravity, lower heating value and hydrogen-to-carbon (H/C) ratio. The oxidation processes of the fuel are modeled using skeletal reaction mechanisms for a group of chemical surrogate (CS) components that employ generic reactions as well as detailed reaction kinetics. The generic reactions were built to model unavailable reaction pathways from physical surrogates to the base CS components or their intermediate species. The model includes the 6 major chemical classes of typical hydrocarbon fuels, i.e., n-paraffins, iso-paraffins, olefins, naphthenes, aromatics and oxygenates, and 65 representative hydrocarbon components are included in a data base for the physical surrogate components and 43 chemical surrogate components are considered for the group chemistry representation. Six types of generic reaction sets are proposed to model the oxidation of 15 physical surrogate components that are called extended-CS (chemical surrogate) components. The present generic reactions are modeled such that the major characteristics of the oxidation process of the fuel components are included in a consistent manner. The final version of the PSGCR reaction mechanism consists of 264 species and 1292 reactions. Validation of the reaction mechanisms was performed by comparing predicted ignition delay times with experimental measurements and/or predictions from comprehensive mechanisms available in the literature. The model was also validated against HCCI experimental data of the FACE fuels. Then the model was applied to simulate practical diesel fuel (F76) spray combustion in a constant volume chamber, and the predicted ignition delay times of the surrogate model fuel were compared with measured values. The results show that the present PSGCR method captures the combustion characteristics of complex multi-component fuels, giving reliable performance for combustion predictions, as well as computational efficiency improvements for multi-dimensional CFD simulations through the use of reduced mechanisms. (C) 2015 The Combustion Institute. Published by Elsevier Inc. All rights reserved.