Reliability engineering techniques have been used in the manufacturing environment for many years. However the reliability analysis of repairable systems is not so widely practised in the steel manufacturing environment. Many different analysis methods have been proposed for the modelling of repairable systems, most of these have had limited application in the manufacturing environment. The current reliability analysis techniques are predominantly used by engineers to construct a “snapshot” in time of a manufacturing system’s reliability status. There are no readily identifiable applications of reliability modelling techniques being applied to repairable systems over a long time period within the manufacturing environment The aim of this work is to construct a method which can analyse and monitor the reliability status of multiple repairable systems within the steel plant over an extended operating period. The developed analysis method is predominantly automated and is facilitated by applying standard reliability analysis techniques to all of the repairable systems failure data sets under review. This Thesis illuminates the methodology used to fulfil the remit of this research by the following sequential steps: Developing a new methodology for the application of reliability analysis techniques to repairable systems within a steel manufacturing facility Utilised an innovative step of combining three reliability analysis methods as complimentary activities Constructed an automated reliability analysis model which fulfils the project remit. In addition the model is capable of the long term monitoring of repairable system reliability The new reliability analysis method has been delivered to Tata Steel and is installed in the Port Talbot Technology Group with a direct link to the Hot Strip Mill (HSM) monitoring database. This reliability analysis method has been tested with four years operational data from the Hot Strip Mill manufacturing area and the analysis has shown that changes and trends in all systems reliability status can be easily identified.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:567218 |
| Date | January 2011 |
| Creators | Owen, Robert |
| Publisher | Cardiff University |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://orca.cf.ac.uk/25646/ |
Page generated in 0.0019 seconds