Gasunie, Nedtrain, Essent, NLDA, SiTech, Sabic
This project is funded by World Class Maintenance and the companies involved in the project.
In this project two PDEng students work on the design of tools that assist in achieving World Class Maintenance. The two PDEng assignments focus on decision support methods for the following challenges:
Determination of Failure Mechanism (Karampelas)
Although maintenance tries to prevent them, failures regularly occur in practice. However, a system or process operator or maintenance engineer normally lacks the knowledge to assess the precise failure mechanism that caused a certain failure. But, a precise identification of the mechanism yields considerable benefits for (1) performing a root cause analysis of the failure, and (2) performing a RAMS analyses with collected failure data. The latter is now often difficult, while registration of the failures in the computerized maintenance management system (CMMS) is now typically done by entering a failure type / failure mode / cause or selecting from a drop-down menu. In this project a decision scheme and tool (e.g. app for mobile device) will be developed that assist (process) operators in assessing the appropriate failure mechanism. The final deliverable of this project will be a validated decision tool to determine the failure mechanism, preferably also implemented as an app on a mobile device.
Guideline for selecting a condition monitoring method (Mouatamir)
Many sensors and condition monitoring techniques (e.g. vibration analysis, oil analysis, thermography), as well as several non-destructive testing techniques are available nowadays. However, applying these techniques and translating the collected condition monitoring data into effective condition based maintenance concepts for a complete complex system or plant provides some challenges:
- Which (sub)systems should be monitored, based on bad actor list, integrity critical equipment, process safety critical systems and criticality analysis ?
- What are the best techniques for each application ? What are the limits of application of the different inspection / monitoring techniques ?
- How can a business case be made: does investing in CM / CBM provide sufficient benefits ?
- How can the collected (raw) data be processed into useful maintenance information ? Due to the wide application of sensors in industry, this is in many cases a ‘Big Data’ problem.
Several practical approaches and methodologies available at present cover these topics partly (e.g. RCM and the method from a previous WCM project for selecting systems suitable for CBM). However, in the present project we aim to integrate all the aspects mentioned above.