Loading...

Development of an Intelligent Reliability Centered Maintenance System

Mardani Shahri, Majid | 2022

209 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 55152 (01)
  4. University: Sharif University of Technolog
  5. Department: Industrial Engineering
  6. Advisor(s): Eshraghniaye Jahromi, Abdolhamid; Rafiee, Majid; Houshmand, Mahmoud
  7. Abstract:
  8. Reliability-centered maintenance (RCM) is one of the main maintenance methodologies that ensures the continuity of the organization's physical assets in order to meet the expectations of its stakeholders. Despite the widespread use of artificial intelligence in maintenance, the use of these tools to improve the RCM implementation process is a new issue. To this end, in this dissertation, using artificial intelligence and soft computing tools, an upgraded RCM system is introduced so that in addition to increasing the accuracy in the implementation of RCM processes, the time and cost required to implement these processes has reduced, and made it possible to record and store the knowledge of maintenance experts for use in future decision making. In this regard, the necessary RCM processes have been identified and the desirable solutions to improve these processes have been provided. The processes under consideration include the process of determining the criticality of the assets and the process of analyzing failure modes and effects (FMEA). In the proposed solution to improve the first process, users' expectations of the system, the importance of factors affecting the criticality, ambiguity and uncertainty in the process and modeling the process of the experts induction were considered. For this purpose, the concepts of fuzzy logic, Mamdani fuzzy inference systems, multi-criteria decision making methods and risk matrix have been used. In order to improve the second process, a solution to overcome the weaknesses of the process of failure modes and effects analysis, including risk assessment of failure modes and their prioritization and the difficulty of using and complexity of FMEA worksheets are presented. For this purpose, fuzzy logic, Pythagorean fuzzy numbers, developed fuzzy clustering algorithms and VIKOR method have been used. In order to evaluate the performance of the proposed solutions, a case study was conducted using data from a gas refinery. Comparison of results with other methods showed favorable flexibility, performance and efficiency of the proposed approaches
  9. Keywords:
  10. Reliability Centered Maintenance ; Fuzzy Logic ; Fuzzy Inference System ; Multicriteria Decision Making ; Failure Mode And Effect Analysis (FMEA) ; Clustering ; Assets Criticality Analysis

 Digital Object List

 Bookmark

No TOC