The introduction of the OBD-2 interface provided standardized access to vehicle diagnostic data, enabling real-time monitoring of the system’s status. Effective utilization of this data requires the development of intelligent algorithmscapable of analyzing the obtained information while considering signal variability and the probability of faults. One of the approaches to improving analysis accuracy is the use of finite state machines (FSMs), which allow structuring the decision-making process based on a set of defined states and transitions between them. This study explores the method of integrating FSM into OBD-2 data analysis processes to create an automated diagnostic system that enhances fault detection accuracy and reduces the number of false-positiveresults. The proposed diagnostic model employs FSMs to build a flexible and scalable logic for analyzing a vehicle's condition. As part of the research, a mathematical FSM model was developed, considering the temporal variation of OBD-2 parameters and identifying critical deviations based on signal timing characteristics. A software package for system modeling and testing was created, allowing the verification of its effectiveness based on both synthetic data obtained in the MATLAB/Simulink environment andsimulated scenarios. A comparative analysis of fault detection accuracy using the proposed FSM model versus traditional threshold methods demonstrated an increase in diagnostic reliability. The testing results showed that fault detection accuracy increased to 92.2%, while the false-positive rate decreased to 4.1% compared to classical OBD-2 data analysis methods. The proposed approach reduced processing delay to 250 milliseconds per diagnostic cycle, making it applicable for real-time fault detection.