Predicting the Remaining Useful Life of Assets

The industrialisation of machine learning and artificial intelligence is very challenging. In the context of health management, the optimisation of asset performance throughout its lifecycle, ML and AI struggles in dealing with scalability. Scalability can include challenges such as variance in the component or systems manufacturing or integration, variability in direct usage and influencing ambient conditions, human interaction and interference. In collaboration with the research team, Smart Systems Group: , the student will have access to lifecycle data from component testing. This will include mechanical, electromechanical, storage and electronic components.

The knowledge and skills in the field of Prognostics and Health Management are highly relevant to all industrial sectors, and represent a valuable addition to your career development. Examples of PHM can be found via the Smart Systems Group webpage.

The student will;
- review reliability analysis,
- review prognostics and health management,
- evaluate data sets on chosen components or products from lifecycle test data,
- optimise models for predicating remaining useful life,
- explore robust frameworks for evaluating models e.g. skewed data sets, missing data etc.
- prepare a scientific paper on this topic.

Supervisor name: 
Professor David Flynn
Supervisor and Deputy email addresses:
Project location: 
Remote working
Deputy name: 
Dr. Jonathan Swingler