Naftamatika Artificial Intelligence analyzes the shape of the dynagraph and compares it with the typical dynagraph shapes corresponding to typical malfunctions. Neural network is trained on a large number of real dynagraph cards and is able to determine the overlap of several defects on each other. Recognition is done by pattern method and a new type of warning can be added
according to the specific requirements of the customer. Recommendations are issued with a note of the severity of malfunction.