Abstract: Car may be classified by a number of different standards and targets. Even, a broad classification is difficult, because a vehicle may fit into multiple categories. The proposed work provides a computer-based tool capable of classifying cars, as closely as possible to classifications performed by skilled operators. Such a tool is capable of extracting a number of numerical parameters characterizing the cars in areas like Value for Money, Design and Function and On the Road Performance. Such parameters are, then, used for training and learning an Artificial Neural Network (ANN) with the aim of classifying them by using different training and learning functions which reflects its performance level. From the results, it is revealed that TRAINLM, LEARNGDM & LOGSIG give comparatively good performance for this problem. So it would be easy to classify a car, whether it is family car or sports car or any other car within one second.