IIoT for R&M: The Real Struggle

In Part 1 of this article (available in the Oct/Nov Paper360°), we discussed the first major hurdle we must clear to reap the benefits of IIoT technology: For “big” data and the M&R IIoT to take root and grow, we must stop ignoring data over our own biases, tribal knowledge, and culture.

I did not write about this fact to beat up on operations and maintenance, but to point out that we are biased against using data we do not completely understand. For the M&R IIoT to take root, we must come to terms with—and give up—our biased views of operations and maintenance. We must trust the data and information the M&R IIoT can provide.

How can we learn to do that? The first step in organizational adoption of M&R IIoT is an organization-wide understanding of RCM softside principles.

Those principles are:
1. 80 percent (±) of all equipment failures occur randomly with respect to age of the equipment.
2. Most equipment failures follow a degradation curve known as the P-F interval.
3. The human senses are capable of detecting 80 percent (±) of failed states, and often human senses are all that are needed to detect equipment problems.
4. The people who work closest to failing equipment are the subject matter experts (SME) on that equipment.
5. Data are not required to begin reliability work, only the SME’s knowledge.
6. It is vital to consider how a failure affects safety, environment, quality, and/or production. Under no circumstance should the consequence of failure or criticality be allowed to determine frequency of inspection.
7. As risk is inherent in everything we do, we must define what level of risk is tolerable.
8. Assets can only perform as well as they are designed, installed, operated, and maintained. We must understand what our equipment “can” do vs. what we “want” it to do.
9. Failure modes (root causes of failure) occur in three ways: suddenly, over a period of time, or hidden.
Overcoming almost any type of bias requires education. For the M&R IIoT to take hold we must first educate ourselves on these basic principles of M&R and how they relate to the M&R IIoT. Yet there are two more hurdles we must overcome: accountability and cost.

In Part 1, I wrote about an amazing maintenance supervisor who was extremely skilled at understanding his production area’s operating process. He was able to track, trend, and see all the same data points as any operator. With this data at his fingertips, he began to rebut calls to repair equipment that he knew was not broken.

Yet operator bias against data was not the only problem at hand; the other problem was accountability. It was not uncommon for process upsets to last for several hours; production was not stopped, but the rate was slowed back. Put yourself in the position of the lowest level of operations management. You call this amazing maintenance supervisor and tell him that you have been struggling with the process for hours and that a control valve has failed. The maintenance supervisor reports back and tells you that the control valve has not failed—you only need to adjust a process set point. A set point change takes (maybe) minutes and that change could have corrected hours of slowed back production.

Now, everyone on the plant site knows the process has been slowed back, and everyone in upper management wants to know why. Are you going to tell upper management that maintenance found that an operational set point was wrong? Only an organization with very good trust will tell the truth on this one. The data are there, the knowledge is present, but not the trust.

To embrace the M&R IIoT, we must be pre-forgiven to make mistakes. Mistakes are one normal way we learn. The key is to learn from our mistakes and the mistakes of others. When there is no trust in an organization, we only learn from our own mistakes, because the mistakes of others will remain a secret.

Another example of this trust issue I have run into is with vibration analysis. In the early days of implementing vibration analysis on a plant site, most folks were skeptical about how well this technology would work. Vibration analysts would “prove” their worth by letting bearings degrade until there were plainly visible signs of defect. As years went by, trust developed between the vibration analysts and their area maintenance/operations folks and the need for showing the wear dissipated.

This trust took years to develop, but often only penetrated one layer deep. I have seen the trust in vibration analysis be very strong with maintenance/operation folks who were there at the beginning of a program, but with their inevitable replacement by new folks, that trust broke down. For “big” data and the M&R IIoT to take root and grow, we must address accountability, trust, and “CYA” culture.

For facilities being built, getting “wired up” during construction is pretty straightforward. Mills can budget for the cost of equipment, effort to install, and need for training. It’s not quite so easy for mills that have been operating for a while—from a few decades to a century or more. In these older facilities, many are challenged by not only the skepticism of M&R IIoT, but the cost of retrofitting equipment to gather and transmit the data.

As technology advances, the process of getting “wired up” is becoming easier and cheaper, especially with the use of wireless devices; but there is still a cost in manpower and dollars to install and set up even the simplest of “smart” devices and the number of devices needed can be staggering. The cost impediment to M&R IIoT adoption can only be overcome by proving the cost/benefit ratio and a willingness to try—which means there must be trust to fail.

Organization must also:
1. Understand the principles of reliability
2. Understand the business process of maintenance
3. Allow for open communications from employees
4. Trust that employees will do the right thing
5. Provide leadership/coaching/mentoring/training when and where needed.

For M&R IIoT to succeed in our older process and manufacturing plants, the organization must have an entrepreneurial spirit, a willingness to experiment, and be tolerant that sometimes they will fail. There must be a culture that pre-forgives mistakes. And first, mills must learn to overcome the biases, tribal knowledge, and culture that cause workers to mistrust the data these systems provide.

Without all of these elements in place, the M&R IIoT will struggle to take root.

Jay Shellogg spent the last 16 years of his career working at a large pulp and paper mill as a senior environmental engineer and maintenance/reliability superintendent. During that time he encountered many challenges; in his own words, “Some I overcame, and some I didn’t.” Contact him at [email protected].