This paper proposes a novel framework for assessing the system safety of complex electromechanical systems (CEMSs). From the perspective of system topology, the fault pervasion probability (FPP) is first proposed to analyze fault propagation mechanisms in combination with historical failure data. This approach can easily identify all failure propagation paths and rapidly locate fault nodes, thereby providing a valuable reference for maintenance engineers. To overcome the influence of subjective factors, the hesitant interval-valued intuitionistic fuzzy element (HIVIFE) is used to describe the failure consequences of components and fault paths. Then, a system safety indicator is proposed to measure the system state and provide support to managers and operators through integration of the failure consequences based on the hesitant interval-valued intuitionistic fuzzy Choquet integral (HIVIFCI). The bogie system of a high-speed train is selected as a case study to verify the effectiveness and applicability of the proposed approach. The results indicate that the proposed approach can (i) achieve a more accurate result for system safety assessment and (ii) identify all possible fault propagation paths. This study provides a basis for formulating maintenance strategies and reducing accident losses, which have important theoretical value and practical significance.