This study deals with the modeling of a diesel engine to predict the engine performance datas and pollutant via artificial neural network (ANN) method. To collect data for training and testing the neural network, the experiments were carried on a diesel engine. At first part of this work, a parameter packet was used as inputs for neural network and satisfactory regression was found with the outputs (over~95%) except total hydrocarbons. Second stage of this work deals with developing new networks with additional inputs to the first input packet for predicting total hydrocarbons with which the regression was raised to 75 % to 90 %. This study shows that ANN approach can be used for predicting internal combustion engine characteristic values accurately and if necessary the neural network performance that was designed to predict the engine out parameter can be increased with using additional related data.