The transformer is an indispensable asset in railway network infrastructure. Condition monitoring and asset management is therefore a key concern in all electrical railway utility providers especially for railway infrastructure manager. This paper is a result of an industrial project in collaboration with SNCF Réseau dedicated to the determination of traction transformer faults based on the dielectric fluid analysis, and analysis of the active parts. These methods are the most prevalent techniques for detecting progressing faults in transformers immersed in oil. Different methods have been employed for analysis of dissolved gas in oil. With these methods, typical faults that manifest in in-service transformers have been identified. Furthermore, in this paper, a statistical study has been done to find the methods between the group which have the same fault code. This study has been applied for different categories of transformers for fixed railway traction electrical substations of SNCF Réseau. The results show the ability oof Duval triangle and CIGRE SC-15 methods to generate more fault code compared to other methods. Also, a validation on several case studies shows the robustness of the method of Duval triangle.