The 13th Annual Postgraduate Research Conference, hosted by the Doctoral College.
Kazeem Balogun (PhD, FST) with this poster entitled: Overview of blackblaze HDD analysis for predictive maintenance.
Click the poster below to enlarge.
With recent studies showing viability of machine learning (ML) among other tools as a better approach in performing Hard Disk Drive (HDD) analysis and enable fault detection models, the adoption of ML in the field of predictive maintenance (PM) becomes more complex. The complexity is not farfetched to the increase in sensor deployment, large volumes of generated data, variances in attributes or parameters and differences in characteristics of data sets. We provide an overview of HDD analysis using Backblaze as a case study and literature review of different ML approaches to HDD analysis for PM. We observed reasonable amount of studies in this direction, but the application of ML towards HDD analysis for PM needs more focus to improve model accuracy and stimulate further work in this area. We shall conclude this study by proposing our ground-breaking state-of-the art HDD analysis and model enhancement.
You can view the full poster exhibition and pre-recorded presentations on the conference webpage.
If this research has inspired you and you’d like to explore applying for a research degree please visit the postgraduate research web pages or contact the Doctoral College dedicated admissions team.
