This paper puts forward a method of preliminary driver classification, which applies two-criteria based analysis of the phenomenon of driving style. The resulting rank enabled the author to order the drivers according to their driving style, from the most active to the extremely mild. The most active driver in the sense of the two-criteria analysis is the one who covered a test stretch of the road the fastest while changing the position of the accelerator pedal most intensively. The classification of the drivers, along with the measurement data registered during road tests, will subsequently provide the foundation for later modeling of a dynamic classification of the driving style, in the form of a recurrent learning neural network of the Elman’s type.