In 2017, Canada presented its intent to buy C$90 million worth of fighter jets. While these jets need regular modifications to be kept up to date with the airworthiness standard of Canada, they rely on older aircraft architecture like the MIL-STD-1553B. In this thesis, we investigate the MIL-STD-1553B technology used in aircraft systems and explore how aircraft components can be automatically classified. We propose a novel a two-step active scanning approach to establish message timing, built-in test responses and memory contents in order to classify the devices. OMAP is able to classify device types and versions. We compared the accuracy of multiple Machine Learning classification algorithms when exposed to different test case scenarios as well as compared: One-step vs Two-step classification, Joined vs Separate spoofed class, Timing features granularity effects. Finally, using ANN and SVC we obtained a classification accuracy of 95% for device type and 88% for device version.