A knowledge-based decision support system in reliability-centered maintenance of HVAC systems

Wong, Daniel (2000) A knowledge-based decision support system in reliability-centered maintenance of HVAC systems. Doctoral (PhD) thesis, Memorial University of Newfoundland.

[English] PDF (Migrated (PDF/A Conversion) from original format: (application/pdf)) - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. Download (27MB) [English] PDF - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. (Original Version)

Abstract

Studies have shown that in large physical systems, it is possible to eliminate or reduce costly machine failures, equipment downtime, lost production and decreased revenues by keeping abreast of the most effective and current maintenance techniques available. -- The purpose of this thesis is to propose a multi-faceted approach to maintenance which can address the short comings of conventional approaches to maintenance. -- The proposed methodology combines the reliability-centered maintenance technique (RCM), a fault tree analysis, a database system, and the Weibull analysis. The integration of these techniques produces an innovative system which increases the reliability and availability of the system. To the author's knowledge, this integrated approach has not been done before. -- As an example, the heating, ventilating and air conditioning (HVAC) of large buildings was used to illustrate this methodology. Failure data was collected from the Biotechnology. Arts and Administration Extension and Earth Resources Buildings of Memorial University of Newfoundland (CERR) over a six year period. The data included the time to failure and failure modes for each component within the central HVAC system. The collected data was used to quantify the reliability of the system. A probabilistic analysis based on the Weibull distribution was used to analyze the time to failure data. -- Using reliability-centered maintenance to identify the causes and impact of failures, the information acquired was used to develop fault trees. Failure modes identified in the fault trees were coded as identifiers to be used in a knowledge-based system for improving the reliability and availability of the system and its components. -- It was shown that system reliability can be improved by increasing the reliability of each component utilizing the proposed multi-faceted approach. Failure data analysis enabled us to quantify the reliability for many sub-components within the major components that constitute the HVAC system. -- It is concluded that the developed knowledge-based system enables us to troubleshoot causes of failure at a much faster rate and this will decrease the down time and increase the availability of the system.

Actions (login required)

View Item