Busting The Top Maintenance Myths

Ben Stevens

By: Carroll Mccormick

The worth of standard rules of thumb for maintenance practices can be overvalued when looked at from the perspective of a company's bottom line.

There is often a poor fit between maintenance departments' understanding of how they add value to their companies and how executives calculate the value of various maintenance activities on their corporate bottom lines. That's the view of Ben Stevens, president of OMDEC Inc., a Godfrey, Ontario-based company that supplies maintenance-optimizing solutions to an international client list.

"I almost never come across anyone who will put value in the equation when talking about maintenance," says Stevens. "Maintenance is a business, not a service. Its purpose is to add value to the organization. If not, we should change the rules of preventive maintenance."

A maintenance manager who can speak the language of executives can be a better advocate for his team; for example, for correcting the misconception in some boardrooms that maintenance is a cost centre and is the equivalent of a corporate boat anchor.

In fact, some maintenance practices do not add value to a company, Stevens advocates, and should be discontinued.

Here are his top five maintenance myths, through which run the common thread of money and value to the company.

Myth 1: Frequency of failure is a critical Key Performance Indicator.

All equipment is not equal in its value to a company. Maintenance departments frequently fail to consider what is called the future risk of failure -- the full cost of equipment failure to the business.

The full cost must include the breakdown repair, the cost of lost production and the cost of the company's loss of reputation when it can't deliver goods on schedule. Stevens finds that maintenance departments often have no easy access to the revenue loss figures that result from equipment failure, which can lead them to put equal resources into maintaining equipment with hugely different future risks of failure to the company. For example, a boiler may fail four times in a given period of time, with 16 failure hours. A primary circulation pump may fail only twice, with just six failure hours. Should the boiler get more PM? Adding a revenue loss of $500 per hour attributable to the boiler and a penalty cost (lost production and reputation costs) of $18,000 yields a total failure cost of $28,400. The pump, on the other hand, cuts revenue by $15,000 an hour and carries a penalty cost of $90,000, for a total failure cost of $158,000. "You must base your judgement on whether to do maintenance on the total failure cost, or not," says Stevens.

Myth 2: A Run-to-Failure approach is bad maintenance.

"If the cost of preventing the failure is higher than the cost of failure, then let it fail," says Stevens. "Most companies do not properly track the cost of failure." In the drive to evolve to preventive maintenance (PM), maintenance teams sometimes forget the value of run-to-failure. Keeping in mind that cost of failure includes breakdown repair, the cost of lost production and the loss of reputation (on a really bad day these three points could mean "lose your job, lose your customer, lose your company"). Stevens advises, "Only if the cost of the failure exceeds the cost of the PM should we not let it fail." The cost of PM can be obtained from the CMMS (computerized maintenance management system) work order; e. g., the cost of parts, labour, contractors, tools, etc. The duration of the failure can be recorded by the work order and the failure cost to the company needs to be calculated. With this data, maintenance is in a position to decide which approach, PM or run-to-failure, makes the best business case, says Stevens.

Myth 3: More PM is good maintenance.

The belief that 'PM is good' has in many cases morphed into 'more PM is better', in Stevens' experience. Simply put, PM is supposed to prevent the deterioration of the performance of a machine, but often the wear rate of a machine bears no resemblance to the frequency of PM done on it. PM must be linked to preventing failure. It is true that determining the link is not necessarily simple, since you don't want to run some equipment, say hoist chains, to failure just to see how long it takes. Too, regular PM, by its nature, prevents tracking run-to-failure time. That said, a fairly simple exercise can help fine-tune PM schedules, bite by manageable bite. "Take about 10 work orders a week, look at them carefully and ask whether they are doing what they should do: examine the list of tasks, list of tools, materials, sequencing of tasks, the reason why the tasks are being performed," Stevens suggests. Critique the PM tasks, but ask what the cost of failure is if a PM interval is stretched too far. Up to 70% of PM tasks are a waste of time and maintenance managers should be encouraged to challenge the issue of PM, he says.

Myth 4: The purpose of maintenance is to provide reliability, availability and maintainability.

The ultimate purpose of maintenance is to add value to the organization. Tasks that do not add value should not be done. Adding reliability, for example should not be done if its cost exceeds its value. Overall Equipment Effectiveness (OEE), for example, which seeks to obtain a combined measure of a plant's availability, performance and quality, which can be used to track how close a plant is running to the ideal, is becoming more prevalent in maintenance. Yet, says Stevens, "You need to consider the impact of maintenance (or its lack of impact) on cash flow, ROI (return on investment) and profitability. What is the cost of increasing OEE and the payback to the company of increasing OEE?" This is a huge area of discussion, but consider one example: What is the value of changing oil at a certain interval? It surely has value to the oil supplier, but is a longer machine life, for a machine that might require replacement for other reasons, worth the sum cost of those oil changes? Can the oil change interval, or any other replacement interval, for that matter, be lengthened beyond those that manufacturers recommend? "Many maintenance tasks are too frequent and therefore too costly," says Stevens. "But we need to be careful to balance our thinking where the cost of failure is catastrophic. The higher the cost of failure, the more conservative we should be; in other words, the more we should err on the side of premature replacement."

Myth 5: The more condition data we collect, the better our analysis will be and the better our maintenance will be.

"Collecting condition data is only useful if the ability to extract knowledge from and take action as a result of the data adds to our understanding and the prevention of degradation and failure. Most condition-based monitoring data is not useful and its collection should be discontinued," Stevens believes. OMDEC has done case studies in which plants are collecting 20, 30 or even 80 streams of data, but only a tenth of which are useful for predicting machine degradation or failure. What do the data streams have to do with the failure that maintenance is trying to avoid? What are the conditions that will help predict failure and how can they be measured?

Stevens illustrates the problem with another oil example: Is an oil analysis that measures iron, lead, chromium, selenium, carbon, coolant, etc., identifying the condition that is causing the contaminant, or which critical components are damaged, or the optimum time to adjust or change that component? Which of those contaminants actually have any impact on performance and degradation? Is it really the best business decision simply to change the oil? Too, it costs money to collect this data; e. g., more channels to collect more data streams cost more to buy. What does it cost to look at this data, assuming it is even being looked at? Is it obscuring data that maintenance really needs? And what is the point of the exercise, notes Stevens, if the data is being collected faster than it can be analyzed? Certainly these viewpoints are not typical of standard thinking in today's maintenance environment. Stevens' suggestions may require thoughtful consideration, yet if applied could result in significant maintenance cost and labour savings.

Montreal-based Carroll McCormick is the senior contributing editor for Machinery & Equipment MRO.