Industry 4.0 and the Internet of Things (or Industrial Internet of Things if you prefer) are so closely related that we may as well use the terms interchangeably. Every day there seems to be some news about the Internet of Things and how it’s going to change everything – connected cars, connected people and even connected washing machines. Yet those connections don’t mean a whole lot without real use cases – yes, it’s now possible for your fridge to talk to your phone but that’s answering a question that nobody is asking, solving a problem which nobody cares about (although I’m prepared to eat my words in 5 years’ time when we are all sharing our fridge stats on whatever is the hottest fridge-based social network).
As an industry, Manufacturing has some very well defined problems – customers want goods quickly, at lower costs and at high quality. If these demands aren’t being met because of unplanned downtime, the consequences are expensive. It’s easier to make a case for using IoT technology to solve these issues; for example, better supply tracking using RFID tags can help to reduce supply-chain delays and improve quality. But it’s not what is really interesting.
With data storage being cheap, bandwidth ever-increasing and the cost of sensing devices steadily coming down whilst their abilities increase (being able to push straight to the cloud, being able to sense more than one thing at a time, etc.), doing intelligent things with machine data, beyond basic analytics is increasingly viable.
If a manufacturing business is not producing when it needs to be, it is not fulfilling its purpose – it is not making money. By applying diagnostic and prognostic techniques to machine data, you can reveal how machines are currently performing (diagnostics) and how they will perform in the future – will they be able to produce when they need to be (prognostics)? As we’ve mentioned before, prescriptive, calendar-based maintenance is a great way to spend money unnecessarily – predictive maintenance is the smart way to save money.
Prognostics –the science of predicting when a machine will stop being able to perform its intended function is a direct enabler of predictive maintenance. The data from your machines, processed by the right application can tell you when your machines will fail, so that you can maintain them ahead of this time and save yourself from downtime, failing your customers and spending more than you should to keep the machine performing. This knowledge of when a machine will fail is calculated as Remaining Useful Life (RUL).
With an accurate prediction of RUL, maintainers are better able to prioritise which machines need maintenance and in what timeframe; significantly reducing the risk of machine failure and avoiding costly over-maintenance. Think of it as another tool to help boost the effectiveness of maintainers.
Prognostic models giving RUL have traditionally been difficult to calculate and have needed talented data scientists to implement effectively on a bespoke basis. The high costs of this approach have harmed widespread adoption. Thankfully with the advent of advanced machine-learning and inexpensive cloud-computing, prognostics is becoming more accessible and in-turn predictive maintenance is achieving scalability.
Security, compliance, integration, process awareness and improvement are all important things that Industry 4.0 enables. Yet none of those matter without effective machinery, delivering reliable throughput and with minimal downtime – all enabled by prognostics.
Tell me more!
Want to learn more about prognostics and Sesnseye, our easy-to-use product for predictive maintenance? Book a demo and discussion here: