Hydac Technical Training Manager Paul Marley continues on from his previous podcast on digital transformation in the form of Industry 4.0. Mr Marley has been involved in the fluid power industry for over 30 years as a hydraulic fitter and technical trainer.
Industry 4.0 technology enablers are a group of diverse technologies – spanning machine learning and robots to additive manufacturing and cloud technology – which work together in an ecosystem.
And yes, I know these technologies sound like they would more broadly supplement the fluid power industry than impact it directly, which brings to the fore the question as to how Industry 4.0 does in fact impact the industry.
It’s a fair question which depends on the enabling technology as some are more relevant and more direct in their effect than others.
Fluid power industry technologies of interest
When it comes to, for example, cloud technology, we use it but don’t make it, and this applies to robots which we don’t produce but use in our manufacturing.
So, which of these technologies is going to be of more interest. From our perspective, only a combination of these technologies will work as a solution for Industry 4.0 – ultimately what it’s all about. And within these technologies, predictive maintenance is of the most interest as it’s where we have the most impact.
Predictive maintenance is where it’s at
Broadly speaking, the different types of maintenance include reactive maintenance which centres around repairing breakdowns when they occur and preventative maintenance which centres around mobile or stationary machine service interventions at scheduled intervals before breakdowns occur.
Predictive maintenance service work is done on an as-needed basis, with the assessment as to when to intervene based on sensor technology.
It’s all about predicting a machine’s future state, which in a nutshell entails collecting data on how the machine behaved in the past to apply to the present and then to predict a future state – a form of time travel.
This has traditionally been called ‘condition monitoring’ in the context of industry.
And yes, condition monitoring has been around for a while, with the question arising as to what has changed to enable it to form part of Industry 4.0? From the outset it does simply sound like a change of terms and therefore not of much interest to us. But it’s a lot more than that.
Difference between predictive maintenance and condition monitoring
Predictive maintenance is different to condition monitoring – which basically looks at the current state of the machine and takes readings on it – as it hinges on the technology of accurate forecasting. Another difference is that more is done with information gleaned.
Traditionally it was about recording the state of the machine, but now with Industry 4.0 it’s about collecting data on the machine passively all the time and in real time, which was not the case previously.
And the fact that information is recorded in real time means that the period between readings is less, enabling increased accuracy when assessing the machine and pinpointing failures at their starting point.
This can be applied to all machines and not only new machines. It’s important to note that putting sensors on an old machine which is always breaking down isn’t going to reduce breakdowns. In this case maintenance practices and lubricant storage have to be repaired and users, operators and maintenance personnel have to be trained.
As to condition monitoring, it’s expanded what’s possible in this domain, with the accuracy of information coming through providing answers that differentiate Industry 4.0 from Industry 3.0.
Benefits of predictive maintenance
Then there’s obviously the question of whether condition monitoring has a greater benefit and return on investment, taking into account the cost of training people and placing sensors onto machines.
The answer here is a definitive yes as long as the predictive maintenance works, keeping in mind that anything applied runs the risk of not working.
To work it needs buy-in from all business personnel. A common problem with condition monitoring is that a slot set aside to service a machine is cancelled due to the production department piping up that it needs the machine to keep production schedules.
However, if the maintenance department has the authority to adhere to plans, then the data collecting will bear fruit and there will be a return on investment.
The next question is why would condition monitoring be required on a machine that is more reliable as a result of the processes in place to ensure its reliability?
Here the important point to note is that it’s not as simple as addressing reliability first and then putting sensors on a machine. The idea is that once a machine has become reliable, putting sensors on it and collecting data helps to predict future performance, taking into account that all machines need interventions and repairs. The advantage of the sensors is they enable the collection of data for more accurate machine predictions.
Through my training I’ve discovered that what works is to get people thinking about maintenance in the first place, which revolves around mechanics repairing machines in the Australian setting. However, of fundamental importance is the idea that the function of the maintenance department is to maintain machines. Otherwise it would just be called the repairs department. So it’s all about having the knowledge, skills and reliable technology. That’s what our industry demands and is able to enhance.