In this study, a modeling of a mobile air conditioner system in different amounts of refrigerant and in different compressor revolution speeds was carried out via artificial neural network (ANN). The three-layered ANN that is utilized in order to be able to model this system has 2 cells in its input layer and 3 cells in its output layer. The least problem-yielding ANN structure was analyzed by examining the number of cells in hidden layer from 6 to 19. The best result was obtained in the ANN with 10 hidden cells. A 0.945 was obtained in estimating coefficient of correlation cooling capacity, 0.985 was found in the power consumed in compressor, and 0.994 was established in COP estimation. The obtained correlation coefficients showed that ANNs can be used with a high precision in guessing the performance parameters of mobile air conditioner (MAC) systems. (C) 2010 Elsevier Ltd. All rights reserved.