ePrints.FRI - University of Ljubljana, Faculty of Computer and Information Science

Methods and computer based tools for failure analysis

Daniel Brulc (2011) Methods and computer based tools for failure analysis. EngD thesis.

Download (8Mb)


    The purpose of this thesis is to present the most commonly used methods for failure analysis. Maintenance management can significantly affect the efficiency of a company – particularly in terms of allocating resources. Analysis, as the cornerstone of business decision making and using modern approaches and technologies, has progressed substantially. Using modern approaches, such as data mining, we can discover the relationship between data and find patterns in them. Root-cause analysis (RCA) is a reactive method used for finding the real cause of failure. We use RCA when failure has already occurred. The method uses different creative and structured approaches. Fault tree analysis is a proactive method, consisting of event and logical gates that present the relationship between events. We analyse the fault tree using the rules of Boolean logic. Qualitative and quantitative evaluation of the fault tree give important information about the probability of the events or the reliability of the component. The second proactive method presented is Failure Mode and Effect Analysis, which is based on a list of possible failures. For each failure, we calculate the risk priority number. The list of failures, sorted descending by RPN, is an indicator of the importance of each failure. In the last section, we present a prototype application for failure analysis. Using this application, users can describe the failure and make further analyses based on various criteria. This helps management make correct decisions for reducing failures and errors.

    Item Type: Thesis (EngD thesis)
    Keywords: failure analysis, quality assurance, efficiency, root cause, IT support
    Number of Pages: 58
    Language of Content: Slovenian
    Mentor / Comentors:
    Name and SurnameIDFunction
    viš. pred. dr. Damjan Vavpotič286Mentor
    Link to COBISS: http://www.cobiss.si/scripts/cobiss?command=search&base=50070&select=(ID=00008666452)
    Institution: University of Ljubljana
    Department: Faculty of Computer and Information Science
    Item ID: 1528
    Date Deposited: 23 Sep 2011 16:23
    Last Modified: 04 Oct 2011 09:12
    URI: http://eprints.fri.uni-lj.si/id/eprint/1528

    Actions (login required)

    View Item