TY - GEN
T1 - Multi Ontology-Based System-Level Software Fuzzy FMEA Method
AU - Hu, Xuan
AU - Liu, Jie
AU - Wang, Yichen
N1 - Publisher Copyright:
© 2020 IEEE.
PY - 2020/1
Y1 - 2020/1
N2 - Failure Mode and Effect Analysis (FMEA) is a method for identifying and analyzing potential failures in systems and has been widely used for reliability and safety analysis of hardware and software systems. However, there are some shortcomings when the traditional method is applied to the system-level software FMEA, e.g., the relevant domain knowledge is scattered and not systematic, which makes the analysis result greatly depend on the experience and the familiarity of the domain to be analyzed of the analyst. Moreover, traditional methods are usually based on textual descriptions and have no tool support. These shortcomings greatly hinder the sharing and reuse of system-level software FMEA knowledge. Besides, the traditional method uses the risk priority number (RPN) to determine the priority of the failure mode, ignoring the objective attributes of the system itself, which is not reasonable enough. This paper presents a multi ontology-based system-level software fuzzy FMEA method. This method realizes the sharing and reuse of domain knowledge through the ontology. In addition, the failure mode rating method based on entropy weight and fuzzy TOPSIS overcomes the shortcoming of the traditional method and can improve the rationality of failure mode rating.
AB - Failure Mode and Effect Analysis (FMEA) is a method for identifying and analyzing potential failures in systems and has been widely used for reliability and safety analysis of hardware and software systems. However, there are some shortcomings when the traditional method is applied to the system-level software FMEA, e.g., the relevant domain knowledge is scattered and not systematic, which makes the analysis result greatly depend on the experience and the familiarity of the domain to be analyzed of the analyst. Moreover, traditional methods are usually based on textual descriptions and have no tool support. These shortcomings greatly hinder the sharing and reuse of system-level software FMEA knowledge. Besides, the traditional method uses the risk priority number (RPN) to determine the priority of the failure mode, ignoring the objective attributes of the system itself, which is not reasonable enough. This paper presents a multi ontology-based system-level software fuzzy FMEA method. This method realizes the sharing and reuse of domain knowledge through the ontology. In addition, the failure mode rating method based on entropy weight and fuzzy TOPSIS overcomes the shortcoming of the traditional method and can improve the rationality of failure mode rating.
KW - entropy weight
KW - FMEA
KW - fuzzy TOPSIS
KW - knowledge reuse
KW - ontology
UR - https://www.scopus.com/pages/publications/85085513549
U2 - 10.1109/DSA.2019.00015
DO - 10.1109/DSA.2019.00015
M3 - 会议稿件
AN - SCOPUS:85085513549
T3 - Proceedings - 2019 6th International Conference on Dependable Systems and Their Applications, DSA 2019
SP - 49
EP - 59
BT - Proceedings - 2019 6th International Conference on Dependable Systems and Their Applications, DSA 2019
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 6th International Conference on Dependable Systems and Their Applications, DSA 2019
Y2 - 3 January 2020 through 6 January 2020
ER -