The development of a zero-dimensional, multi-zone combustion model is presented for predicting the performance and nitric oxide (NO) emissions of a spark ignition (SI) engine. The model is validated against experimental data from a multi-cylinder, four-stroke, turbocharged and aftercooled, SI gas engine running with syngas fuel. This alternative fuel, the combustible part of which consists mainly of CO and H-2 with the rest containing non-combustible gases, has been recently identified as a promising substitute of fossil fuels in view of environmentally friendly engine operation. The basic concept of the model is the division of the burned gas into several distinct zones, unlike the simpler two-zone models, for taking into account the temperature stratification of the burned mixture during combustion. This is especially important for accurate NO emissions predictions, since NO formation is strongly temperature dependent. The multi-zone formulation provides the chemical species concentrations gradient existing in the burned zones, as well as the relative contribution of each burned zone to the total in-cylinder NO formation. The burning rate required as input to the model is expressed as a Wiebe function, fitted to experimentally derived burn rates. All model's constants are calibrated at one operating point and then kept unchanged. Zone-resolved combustion related information is obtained, assisting in the understanding of the complex phenomena occurring during combustion in SI engines. Combustion characteristics of the lean-burn gas engine tested are provided for the complete load range, aiding the interpretation of its performance and knocking tendency. Computed NO emissions from the multi-zone model for various values of the engine load(i.e. air-fuel ratios) are presented and found to be in good agreement with the respective experimental ones, providing confidence for the predictive capability of the model. The superiority of the multi-zone model over its two-zone counterpart is demonstrated in view of its more realistic in-cylinder NO emissions predictions when compared to the available experimental data. (c) 2008 Elsevier Ltd. All rights reserved.