Condition monitoring is a great concept and technology – the ability in some cases to gain an almost instantaneous insight into the health of your machinery is an essential tool in an effective predictive maintenance regime. We’ve covered ‘condition monitoring in 10 minutes’ here.
However, until recently getting the most out of condition monitoring has required a deep expertise in the techniques used and often amounts to little more than setting thresholds (e.g. vibration amplitude) and waiting for a breach – but by that time damage has already occurred.
The human expertise required to get the most out of condition monitoring is expensive – a quick google reveals that condition monitoring / reliability engineers tend to start at $45,000 although this could easily turn into $1000+ per day if you’re using external contractors. Humans also don’t scale well, is one human operator going to be able to accurately and rapidly assess condition monitoring information from 1,000 machines with 10 measurements each? (Protip, the answer is no).
Humans are slow
Condition Monitoring, unless extremely simplified is inaccessible to the untrained production or maintenance manager. These individuals don’t have time to inspect 10 different variables per machine on their production line. It’s a human impossibility that production lines with 500+ machines can be effectively and efficiently monitored by humans. If prognostics is needed to forecast machine failure, that’s another extremely difficult discipline requiring complex mathematics and astute use of a variety of algorithms.
Condition monitoring and prognostics of machines is better left to machines.
It’s a service based industry
Until now, condition monitoring has been serviced based – you purchase expensive equipment, with expensive proprietary software and expensive training to learn how to use it – all without developing tangible expertise in condition monitoring techniques. Whilst some benefits to your predictive maintenance program will be seen, this setup suits the solution vendor most of all.
Condition monitoring and prognostics are complex pattern matching and mathematical problems, the kinds of things best solved by computers rather than relying on ‘eyeballing’ something. Why pay for consultants to slowly analyse condition monitoring and try to forecast machine failure when all of that could be done automatically, in the cloud and without requiring manual intervention?
With the ubiquity of cloud-computing and the rise of the Industrial IoT (or Industry 4.0) providing more information at a greater quantity than before possible. there’s no need for humans to be doing something that is much better suited to machines. With the constant need to improve factory efficiency, removing a manpower bottleneck and introducing automated condition monitoring and prognostics makes sense to reduce operational costs whilst having a higher accuracy of spotting current machine failure, as well as the ability to reliably forecast future machine failures.
We’ve developed an easy to use, cloud-based condition monitoring and prognostics software tool to automatically forecast machine failure which can work with your existing systems. It provides you with warning of when machinery will stop being able to work for you and so helps you to save money and avoid downtime. Download our free flyer to find out more!