In Part 1 of this 2-part series, we’ll look at changing our worldview about what defines failure.
Many years ago, while completing my degree in engineering, I was forced to take a class in philosophy. To this day I can’t give a good reason for why this was required; most likely it was to provide some needed funding to that department. Two concepts have stuck with me from that class, though they never seemed of any value until I began working in the realm of reliability.
The first concept is that our “view of the world” influences our perceptions and actions. Basically, our professor taught that we are all biased (in the statistical sense) by our experiences. A great example of this is in the debates between Albert Einstein and Neil Bohr. Bohr was a physicist who worked in quantum mechanics, where events seem strange and random. Einstein’s work was in belief and pursuit of order. Einstein believed that the entire universe was ordered, patterned, and predictable; he once stated, “The Almighty does not play dice with the universe.” Bohr’s reply was basically, “Who are you to tell the Almighty how to run the universe?” Each man’s view of the world undoubtedly influenced his actions. Einstein’s view was order and Bohr’s view was probability.
The second concept I took away from that philosophy class was that if something had a beginning, it must have an ending. If there is a creation, there must be a destruction; if there is a start, there is a finish.
That brings us back to the title of this article: failure is unavoidable. Assets and systems are designed, built, installed, operated/maintained, and fail. The only thing we can do is prolong the life cycle and/or avoid the consequences of failure. Anyone working the realm of reliability in the pursuit of preventing failure is on a fool’s errand. I know this flies in the face of much reliability industry thinking, because most folks are trying to stop things from failing—but consider your personal health. Everyone who reads this article will die; the question is whether it will be in minutes, days, or decades. All we can do is take actions that prolong our lives, and put things in place to mitigate the consequences when our time comes. That is what true reliability is all about—prolonging asset life and mitigating the consequence of the inevitable failure.
Over my years in the paper industry, I have heard machine managers demand, “I don’t want another bearing to fail.” This sets up an impossible demand on the organization. Every bearing will fail, no matter how loud the pronouncement against failure or how many threats are made against the operating/maintenance crews. I bet many of you reading this article have lived through rash demands made by frustrated managers. So how do we deal with these pronouncements?
It’s not easy. These types of demands come from an emotional reaction typically based in a lack of understanding of the degradation process, and from stress applied on managers from upper managers who are just as uninformed. Unfortunately, the task to educate these folks flows to the supervisor/planner/engineers. These folks are the “lowest common denominator” in any mill because they’re sandwiched between the hourly work force and management. As a former mill manager I knew once put it, these people make up “that thin line that is between a rock and a hard place.”
The task I set before them is to be vocal in the face of direct management hostility. This is a tough thing to ask, but if we don’t stand for what is right, who will? All it takes is guts, confidence, knowledge, and a little bit of a cavalier attitude. Every supervisor I have ever known possesses all of these traits. Hopefully, what I have said to this point sets the scene for where I’m going: We need to give the supervisor/planner/engineer a little more knowledge, confidence, and guts about addressing the inevitable failure that will come.
Where does that leave us, if failure is something we cannot prevent? We must develop a new definition of failure. We need to define a type of failure that can be prevented. We call this “functional failure.” This is a failure we can avoid.
Simply stated, functional failure is a point at which the user/owner has drawn a line that they do not want the degradation to cross. For example, let’s consider pump performance. If we have a pump that can pump 400 gal/min, but all we want is 300 gal/min, then our functional failure is when the pump can no longer pump 300 gal/min. Maybe it can pump 299 gal/min, but that is less than the performance standard we have set. So at 299 gal/min, we say the pump is functionally failed (Fig. 1).
Once functional failure is defined, we look for indications that the pump is heading toward functional failure. We might look for a decrease in flow rate below 400 gal/min, but above 300 gal/min. This way we have time to prevent the functional failure of falling below 300 gal/min.
As for the paper machine bearing discussed earlier—when the manager pronounces, “I don’t want another bearing to fail,” we must ask that manager how we define when the bearing has failed. The most common response I’ve heard is, “When it won’t turn!”—but that’s too late.
Okay, I know that everyone has jumped ahead of me and is thinking about all the technology/techniques we apply to bearings to detect their degradation (vibration, ultrasound, broomsticks, etc.) The most important takeaway at this point is to see that most components will follow a degradation curve. This degradation curve is known as the P-F Curve (Fig. 2).
In Part 2 of this column (scheduled for the May/June issue of Paper360°), I’ll discuss the P-F Curve and how to use it to determine the point of functional failure. I’ll also look at different causes of failure and how they develop. If you have any questions or comments about the concepts I have outlined in this article, I would really like to hear from you. Feel free to contact me directly by e-mail or through the editor.
Jay Shellogg spent the last 16 years of his career working at a large pulp and paper mill, primarily 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]