| Abstract: | Medical devices play a critical role in care and treatment. The human-related failures can significantly affect the safety of patients in clinical use of medical devices. This study develops a comprehensive risk assessment model for identification and evaluation of failures which may occur in the clinical use of medical devices. First, the “Swiss cheese” model and SHEL model (the acronym of software, hardware, environment, and liveware) are integrated to comprehensively identify the potential human errors. Then, a new failure mode and effects analysis (FMEA) approach improved by rough set theory and grey relational analysis is developed to assess the risk of the identified failures. The proposed method integrates the strengths of the “Swiss cheese” and SHEL model in identifying human failures from both the vertical and horizontal perspectives of the system, and the advantages of the improved FMEA approach in flexibly manipulating vague information in risk evaluation without much priori information. Finally, the proposed method is applied in clinical use of respirator to verify its efficiency and effectiveness. |
