This study presents a data-driven approach to determining the diagnostic significance of freight wagon components using Shannon entropy method combined with multi-criteria decision analysis. The research is based on operational data from 24888 incidents recorded on five railway lines in south-western Poland during 2009-2013. Six principal wagon systems were evaluated: wheel sets, axle bearings, suspension, brakes, wagon coupler and wagon body. A decision matrix was developed taking into account five criteria: participation in incidents, percentage of incidents without confirmation of emergency status, diagnostic error probability, potential need for re-homologation, and the number of components that can be diagnosed using a single system. Following normalization and application of Shannon entropy method, weights were assigned to each criterion. The results show that wheel sets represent the most diagnostically significant component (score: 0,98 on a scale of 0-1), followed by axle bearings (0,61) and suspension (0,43). The use of Shannon entropy minimizes the subjectivity inherent in expert assessments. The presented method allows the determination of the diagnostic significance of individual freight wagon components, which allows for effective prioritization during the implementation of diagnostic systems.