Stop, you don't need additional IoT sensors

Most companies looking to innovate by using AI predictive analytics for predictive maintenance get scared off by the thought of adding additional IoT sensors, which can sky-rocket implementation costs and delay projects by months (if not years).

Yes, data is like gold and the more data the better, however truth be told you probably have sufficient sensor data already.

What about legacy brownfield plants and facilities? Yes the same is normally true for you.

How can this be true?

Well, let’s explain how VROC’s AI Auto Machine-learning platform works. As it’s not a typical manual data science process.

VROC ingests all available live and historic data, not just sensors connected to your historian (we know you have more that you never connected, as it was too expensive), plus any technical information you have and even external data such as the weather.

Once you’ve entered the problem you would like to solve into the VROC platform, such as ‘tell me where my next failure is going to occur’, VROC’s machine learning technology learns how your whole facility functions. It learns what normal and abnormal operation looks like and will predict the failure with pinpoint accuracy down to the individual component.

Legacy plants do have less sensors available than brand new modern facilities. However you’ll often hear the opposite argument from operators at new facilities. They are drowning in too much data and can’t keep track of where their sensors are even located (imagine that!).

What if the sensor is not available, will the AI still predict my failure?

VROC has been able to predict failures where a specific sensor was not available (see our Generator case study), as other surrounding sensors indirectly identify an issue. If the sensor is unavailable we won’t be able to tell you the exact root cause (although the data will tell you a story of where to direct your attention). In most circumstances this is more than enough information, however if required, additional sensors can be installed to give your team and management 100% confidence in the predictions.

If needed, sensors can be installed that are extremely cost-effective with a data feed straight into the VROC platform.

If you’ve been holding off implementing a predictive maintenance solution, our recommendation is to start with what you’ve got. Commence with a proof of concept (POC) at minimum – as you gain confidence in the data that you have and it’s ability to give you valuable predictive insights. We think you’ll be surprised by the findings!

For a demo of the VROC platform or to start your own POC, contact us here.