Talking Predictive Maintenance with Ralph Rio

predThe convergence of operational technology with information technology has been a hot topic for several years at SAP-Centric EAM, and as software companies like SAP move from industrial Internet of Things (IIoT) concepts to actual IIoT products and services, this creates real opportunities for asset management strategy growth, particularly around predictive maintenance.

I talked with Ralph Rio, Vice President of Research at ARC Advisory Group, about his advice for launching a predictive maintenance program.  Ralph’s 40+ years’ experience with manufacturing and industrial applications at companies like General Electric, Digital Equipment Corp., Motorola, and Texas Instruments inform his research in the areas of Enterprise Asset Management (EAM) and Field Service Management (FSM), continuous improvement programs, and his work with Global Service Providers.

Q: Adoption of predictive maintenance processes and the enabling technologies – moving up along the asset management maturity curve – has been slow.  Why do you think that is?

A: Part of it is an awareness issue.  In the past predictive maintenance was hard to do.  It required investment into a special project team to develop custom code, which took a lot of time and money. This code had to integrate data from control systems into their EAM system, and that custom code breaks down during upgrades. So predictive maintenance was difficult to implement, costly and fragile. That’s the ingrained memory.  Now it’s much less expensive because it is an actual platform that integrates with your EAM and ERP systems. These systems (SAP’s Predictive Maintenance & Service (PdMS) product for example) are now two or three years old and are becoming very robust platforms that are proven. People have – or should have – more confidence.

Q: So how do people begin moving toward predictive maintenance – how do they get there?

A: The first thing people need to do is to educate themselves to understand what is available from a technology standpoint.  People just entering this area are no longer “early adopters” so there is plenty of information out there.  Get comfortable with the platforms and the business processes.

Then they need to find a “tall tree” and what I mean by that is to identify a critical asset where unplanned downtime has serious business implications.  A lot of people make the mistake of doing a pilot on a less visible piece of equipment, but in many of these cases the results don’t really matter or get much attention and the project dies. Do a pilot with a critical asset!   Bring control data together with equipment data into a predictive maintenance platform like SAP’s PdMS and do analytics on it.  Run these over a long enough time so that you can fine-tune and learn to identify false positive failure notifications.

Make sure to structure business processes and integrate the workflows so that when there is an imminent failure notification, a work order is generated to prevent that failure.

And finally, when an alert prevents a failure, report on that.  The original business case is built on the ramifications of preventing unplanned downtime on critical assets and expanding the pilot will absolutely depend on the ability to accurately report on financial outcomes.  If you are an oil and gas company, for example, and a mission-critical compressor fails once a quarter on average, and you can prevent these through predictive maintenance, you can demonstrate savings in the multi-millions. That gets the attention of the executive suite.

Q:  You will be talking more about this in your session at SAP-Centric EAM, why should conference attendees come to your session?

A: I will go into much more detail about how to develop your predictive maintenance strategy, the industrial internet of things and how to structure successful projects, and ultimately how this knowledge can enhance the careers of maintenance managers and leaders.


Ralph Rio
Vice President, Research ARC Advisory Group





Subscribe to our Blog