The term Big Data is everywhere nowadays, but what does it actually mean? According to a definition provided by Google, big data is ‘extremely large data sets that may be analysed computationally to reveal patterns, trends, and associations.’
Sounds great, right? In theory, yes. Through big data, businesses can have a greater overview of their operations than ever before, allowing them to improve efficiency and improve their service. However, in practice achieving this is rarely the case.
The issue is that processing big data is incredibly complex, time intensive and usually expensive. And in many instances analysis of the data doesn’t lead to change. We regularly speak with businesses that are collecting huge volumes of data, but aren’t actually using it properly. And if the data isn’t being used effectively or for a purpose, what’s the point of collecting so much of it – and at such high cost – in the first place?
That’s why the focus needs to shift from collecting big data to collecting the right data, especially in the world of maintenance.
Now, unfortunately there is no Google definition for the term ‘right data’ (yet!), but we consider it to be the ‘critical pieces of data that drive competitive position’. We’re sure you’ll agree that optimising machine health and maximising uptime is key to productivity and profitability – and therefore competitive position.
So what does the right data look like? Well, it doesn’t have one form; instead, it’s likely to differ from application to application and site to site depending on different assets and the chosen maintenance programme.
Let’s take vibration monitoring using accelerometers as an example. With a relatively small volume of data, maintenance teams can easily see problems developing with their assets due to accelerometers picking up abnormal vibrations emitted from rotating parts. In this instance, the data collected that highlights the abnormal vibrations is what we’re after – nothing more, nothing less. There is no need for complex (and costly) software and algorithms; maintenance engineers are instead able to get to work, repairing assets or replacing problematic components before a more serious issue develops.
Fighting the urge to install systems and processes to collect big data is a challenge many businesses in the manufacturing and process industries are likely to face in 2017. But it’s important to stay vigilant. Trust in the right data, take the appropriate actions when necessary and remember at all times that the objective is to improve competitive position, not simply collect as much data as possible.
