Knowing how to predict equipment failure, rather than just reacting to failure, can mean significant benefits for your business, from reduced downtimes to extended equipment life! When looking at raw sensor data it can often be hard to distinguish between healthy and unhealthy operation. Using Condition Indicators, you can make this distinction in your data, thus bringing you a step closer to developing a model that can predict the next failure!
Condition Indicators are often identified in the time-domain by visual inspection. However, they could also hide in other domains, such as the frequency– and time-frequency domain. It is therefore important to visualise your data in other domains. In this second eBook we take a look at
- What are Condition Indicators?
- Why are they important?
- How to select them for your model?