Benchmarking maintenance

When you start feeling good about yourself, it's time to benchmark." That's a rule of thumb from Bob Schmalbach, chairman of the Foundation for Industrial Maintenance Excellence and a retired plant engineering consultant for Dupont.

By Richard L. Dunn, Editor, Plant Engineering magazine March 1, 2001

When you start feeling good about yourself, it’s time to benchmark.” That’s a rule of thumb from Bob Schmalbach, chairman of the Foundation for Industrial Maintenance Excellence and a retired plant engineering consultant for Dupont.

Schmalbach is a proponent of the benchmarking process originated by Xerox. The process looks at two things: first, metrics and performance measures, and second, work practices. “It’s a good process,” he says, “because if you only look at metrics, then the only thing you can understand is that there are better people out there than you are. But you don’t have any insight as to why they’re there – just that they have better numbers. But, if you just look at work practices, all you get is an understanding that there’s more than one way to skin a cat. What you really want to do is to find those guys with better metrics than you’ve got and to compare how they process work, and how you process work, and what the differences are. The way they process work is probably what makes the difference in the metrics.”

Most discussions about benchmarking revolve around working with other companies. But that isn’t really where it should start. Internal analysis – internal benchmarking – is the place to start, according to Rick Herold, director of manufacturing engineering for Harmon-Motive, Inc.

“I think when you’re first starting out, internal benchmarking is great,” Herold says. “You have to set foundations and core values. Then you goal set and see if you’re making any progress. Once you attain a certain level, you realize that your internal benchmarking is not good enough. Then you have to go outside and benchmark yourself against others.”

Tom Williams, manager, North America Plant Engineering for 3M Co., tends to agree. “It’s important to understand first what you’re about, why you’re here, and what’s important to the business. What’s important in one plant isn’t necessarily critical in another plant. So once you understand what’s important, and you’ve established some internal goals and metrics, maybe that’s the time to go external and see what your competitors and other businesses are doing.

“Maybe it’ll spark a new idea,” Williams continues. “Not being afraid to go out and look at someone who’s different from you is very important in benchmarking. They may bring something to light that you never even thought of. It’s going to be a benefit even if you say, ‘didn’t learn anything.’ Or maybe you did. Maybe you learned that what you’re doing is pretty good anyway.”

John Blumenshine, vice president, facilities, S&C Electric, offers an additional caveat. “Internal benchmarking is dangerous and can be destructive to the improvement process,” he says, “because you’re measuring yourself against past performance, and that constant drive to lower that number may be the exact opposite of what you need to do. Until you cross the threshold to outside benchmark, you may never know that. You have to get to the point where you want to know where you stand against the best.”

Many plants find it is helpful to hire a consultant to facilitate benchmarking with other plants, but it may not be necessary.

“I personally don’t think you need to get a third party to help,” says Ed Mayer, director of plant engineering for Syngenta. “It doesn’t have to be so complicated that you have to spend a lot of money to do it. We participated in the North American Maintenance Excellence Award, and we immediately got access to a number of knowledgeable people. There are more than enough ideas you can get just by making phone calls to people. And people are more than willing to share their insights.”

Even so, third parties do have a role to play. If the information you want belongs to a competitor, for example, you may need a third party to serve as an intermediary.

There’s also the question of proper procedure and training. Schmalbach points out that there are ethical and legal issues that need to be understood.

“If you go about it unethically, your partner is going to shut you off,” he says. “If you go about it illegally, you could end up in jail. There are things you cannot do legally, and there are things you wouldn’t do ethically. And you need to understand them.”

Another area that requires understanding is the measures that are being benchmarked. As Don Asmus, plant maintenance manager for Buckeye Florida, cautions, “The thing about benchmarking that has got me in the past is that even something like replacement value means something different in each organization. You can get caught sometimes. If you adopt a benchmark process, you need to stick with it and not intermingle other ones without knowing the right questions to ask in order to boil it down to the right numbers.”

One particularly troublesome area is manpower levels. Upper management, it seems, is always looking for a magic number when it comes to the number of maintenance employees in their plants. But such a number remains elusive. As Blumenshine advises, “I don’t think you can benchmark head count. There are too many variables: the cost of labor, the burden cost of having an employee on the payroll, the outsource cost that you can buy a service for, the union/nonunion considerations, etc. When you benchmark headcount, you run the danger of not comparing apples-to-apples.”

And as Schmalbach cautions on a broader scale, “There is no one measure of maintenance excellence. You’ve got to look at numbers in sets before you can get any real insight as to what’s going on. Usually, high reliability is good and low cost is good – but not necessarily, if you’re working on the wrong stuff. You can drive maintenance costs to nothing just by stopping maintenance. So when I look at cost numbers, I also look at schedule compliance numbers, at PM compliance numbers, and for indications that there’s a reason for the low costs – and that is that people are processing work in the right way.”

15 useful maintenance benchmarks/The following 15 ratios are common in industrial maintenance benchmarking. Values shown were established by the board of directors of the North American Maintenance Excellence Award as representative of superior maintenance operations. They are not intended as targets to be met or guidelines to be followed in maintenance evaluation.

BenchmarkProcess industriesDiscrete manufacturingComments
Maintenance craftspeople per first-line supervisor 15 16 Indicator of how well first-line supervision is leveraged
Maintenance craftspeople per maintenance planner 28 25 Measure of the emphasis placed on maintenance planning and scheduling
Estimated plant replacement value ($MM) per maintenance craftsperson $4.9 $4.5 Amount of assets maintained by each maintenance employee; indicator of maintenance labor effectiveness
Annual training days per maintenance craftsperson 9.3 9 Indicator of management commitment to employee development and functional excellence
Annual training days per maintenance supervisor 7.4 5 Indicator of management commitment to employee development and functional excellence
Total maintenance cost as percent of estimated plant replacement value 2.2% 2.5% Indicator of maintenance cost-effectiveness; high values may indicate potential for cost reductions; low values may indicate milking of assets
Maintenance labor cost as percent of total maintenance cost 24.5% 52% Indicator of overall maintenance manpower planning effectiveness; influenced by numerous factors
Outside contractor cost as a percent of total maintenance cost 20.3% 15% Indicator of utilization of outside contractors
Unplanned (emergency) manhours as a percent of total maintenance manhours worked 7% 18% Indicator of maintenance planning and control levels
PM/PdM manhours as a percent of total maintenance manhours worked 27% 39% Indicator of commitment to proactive maintenance and assurance of equipment reliability
Planned repairs schedule compliance 87% 87% Measure of performance in completing work as scheduled
Preventive maintenance schedule compliance 90% 90% Measure of performance in completing preventive maintenance tasks as scheduled
Work order coverage as percent of manhours worked 100% 98% Measure of maintenance documentation effectiveness. Affects ability to evaluate other factors
Storeroom annual inventory turnover 1.52 1.53 Indicator of storeroom management effectiveness and control; reflective of materials management policies
Storeroom investment as percent of plant estimated replacement value 1% 1% General measure of stores inventory management