From Repairs to Predictive Maintenance

Since my dad recently passed away, I’ve been thinking about some of the things I learned from him. I reminisced about when my parents got a car for the three kids to share. Sharing the car was just the way things were back then, and we didn’t really have any trouble sharing. But about six months after we had the car, my dad asked me when I was going to change the oil in the engine. Being a kid, I was stumped at this question. I knew there was oil in the car, but why do you have to change it? After some re-education, I learned a lot about the internal combustion engine — including why the oil needs to be changed — and then I learned the “how-to” so that I could change the oil.

This was my first experience with what I now know is called preventive maintenance. As I went about my jobs as a quality leader, I realized that quality departments also get involved in plant maintenance. Way back in the beginning, most maintenance was just fixing things that were broken or breaking. Mechanics knew about changing the oil and greasing all the appropriate fittings. But I dare to bet that the phrase “predictive maintenance” was never uttered until around the year 2000.

Our quality audits always seemed to point out areas where paint was peeling, floors were damaged or some bearing was over-greased. To me, it seemed like we were just fixing things after they broke. After learning more about quality management that focuses on consistency and continuous improvement, I took some of what I learned to the maintenance manager, and we discussed the possibility of preventive maintenance.

Bruce Ferree’s first experience with preventive maintenance involved the family station wagon.

Just like changing the oil in that shared car, we had a way to know what needed to be done to the equipment so that it would run consistently for a long time — maybe even past the warranty. The maintenance manager found some software in a trade magazine designed just for this, and he built a grand program. It worked. The equipment ran for a long time and helped the company produce consistent product. The program even helped when tracking maintenance costs such that we knew routine maintenance was now saving the company money. We were pleased. A decade later, we were discussing this tool and asked ourselves: “Could we use the 10 years of data we had to predict when something was about to break?”

Turns out we could take all that data from preventive maintenance activities and transform it into predicting information. We learned when gaskets were getting to the end of their lives. We learned when to expect a machine to need a new part. Preventive maintenance told us when to replace or rebuild a part, but predictive maintenance told us that, for example, after 3,000 hours of use, that part is going to die.

Knowing this, we were able to identify when we should schedule to scrape and repaint areas of the plant and when to take a machine out of use. At this point, the maintenance program was driving consistency of operations and of product. The quality department had less product placed into a hold or rework status. It was now spending time working on more preventive actions than reactions. We became a winning tandem — maintenance and quality.

Thinking back to that first car for us kids, I wonder if I should’ve spent more time with my dad developing a predictive maintenance program instead of just doing the work that was called for in the owner’s manual. Maybe I’d still have that old wood-paneled station wagon. At least I can tell dad that I used some of the tools he taught me: Proper maintenance saves money and results in good quality.