With so many organizations moving from a reactive maintenance approach to a preventive maintenance (PM) approach, while predictive maintenance (PdM) still only in its infancy in the vast majority of organizations, it begs the question as to why I want to write about what comes after PdM. Click here to speak with Dan
Too often hype overshadows market realities and to many, the idea that technology is going to cause a rapid shift in enterprise behavior is just hype. However, in a world where today’s four year olds may never have to learn to drive a car, the idea that manufacturing and resource processing plants can remain rooted in the current way of caring for our production assets is simply a path to failure. LNS Research’s survey data shows there is a building base of Industrial Internet of Things (IIoT) enabled devices and the rapid shift to Cloud data storage and analytics that will enable a new model we call Smart Connected Assets.
Why the Reactive-Preventive-Predictive Evolution has Been a Slog
I have been in the Asset Performance Management (APM) space for over 45 years now, and certainly have seen the long and slow evolution that has taken place as we have moved away from purely reactive maintenance to a point where the standard practice is PM. With the growth of programmable logic controllers (PLCs), distributed control systems (DCSs), and digital communications in the late 1980’s through the 1990’s, condition-based maintenance (CBM) started the shift towards predictive maintenance. The onset of statistical analysis tools in the last 15-20 years has enabled reliability-centered maintenance (RCM). Yet, all of these advances have come at a relatively slow pace. Part of the reason for that has simply been the cost-benefit ratio has been low enough that it is difficult to justify the investment. First it was the cost of the sensors, but digitization drove that down over time. Then it was the cost of connection, but wireless has driven that down. After that it was the cost of configuration, but self-aware devices have driven that down. Analytics were difficult because a deep understanding of statistical methods and domain expertise was scarce. Today’s analytical engines, developed by vendors to address the big data issues associated with customer data and the proliferation of data scientists at most suppliers, has now made the analytics barrier a moot point.
In essence we have reached the tipping point. The convergence of the IIoT, Cloud, Big Data and Analytics, all facilitated by the growth of smart handheld devices. Enabling mobility has created the perfect storm that will drive shifts over the next three plus years. So, by 2020 when self-driving cars become a common sight, although not the majority of cars on the road, a new model of maintenance driven by Smart Connected Assets will emerge.
Prescriptive Analytics Will Drive Prescriptive Maintenance
Just as analytics as a science is evolving from post-event analysis of historical data, to real-time event analysis, to predictive analytics, there is a new frontier on the horizon. This evolutionary step in analytics is referred to as prescriptive analytics. The idea is that the analytical tool not only can predict what is likely to occur, but it can offer “what-if” analysis of alternatives to provide a scenario that can alter the outcome. Since we already have too many P-M acronyms, I offer up the new acronym for prescriptive maintenance: RxM. I define RxM as maintenance activities that originate from within the technology stack itself. IIoT data is fed into either Cloud-based analytics or potentially distributed analytics within the Smart Connected Asset train to optimally define the prescribed maintenance activities based on the “best” outcome.
This shift will radically change the industry and represents a threat to many business models in place today. Consultants that make a living by teaching statistical methods and use of analytic tools or processes that enable RCM will need to change. The teaching will need to shift to defining optimal for your business and configure the next generation of tools. No longer will you need an ensemble of experts to tell you how and when to maintain your assets, as the assets themselves will tell you what they need if they are unable to fix themselves. The first suppliers that offer this capability and the early adopters that use it can impact productivity the same way Uber and Airbnb have transformed their industries. It won’t happen this year or the next, but it is inevitable, just as much as self-driving cars will be.
Join us on Tuesday, March 29, at 1:00 pm EDT as Andrew Hughes presents the results from the fourth iteration of the Metrics that Matter research study conducted between LNS Research and MESA International, and place particular focus on how the deployment of IIoT, Cloud, and Analytics are transforming manufacturing.