3 Maintenance Myths that are Eating Your Cake

Who said, “Let them eat cake?” Marie Antoinette, many people will assuredly reply.

This idea was first espoused in 1766, when Jean Jacques Rousseau wrote of an incident he recalled from some 25 years earlier, in which an unnamed “great princess” was told that the country people had no bread. “Then let them eat cake,” she replied. When Rousseau wrote of this, Marie Antoinette was an 11-year-old child in Austria (the French Revolution would not begin for another 23 years!) The myth that she spoke these infamous words was probably spread by revolutionary propagandists, to illustrate her cold indifference to the plight of the French people. Yet the myth persists over 250 years later, a piece of common wisdom that resists evidence to the contrary.

It’s no surprise that myths are common in politics, but they invade every field of thought, including business, even when they have been shown to be false. They spread at lightning speed in these days of social media, 24-hour television news, and instant global communications. Some alternative facts, like the cake enthusiast Marie Antoinette, are ingrained from years of well-intentioned repetition.

In an operations setting, maintenance managers often succumb to this weakness for hearing a good yarn.

Think about the routine act of budgeting for plant maintenance. It is a mix of what you did last year, what the volume expectations are to be, and the big end-of-life moves you expect. But there is a misconception in most executive suites that maintenance is just a cost of doing business; the equivalent of a corporate ball and chain.

Maintenance should be a business to the business itself. In other words, it should always be providing value to the organization. If the maintenance practices are not adding value, they should be eliminated.

Heresy, you say? In cement, concrete and aggregates, maintenance activities consistently comprise over 9% of the business’s overall costs, according to 2012-2016 benchmarking data provided by the Portland Cement Association.  Maintenance is the highest individual component cost, exceeding spend in direct labor costs, salaries, purchased materials, and depreciation.  Does your company increase headcount or invest in capital unless there is multiple payback of the outlay?  Maintenance should be no different.

The first step in taking back control is recognizing there’s a problem. Below are three maintenance myths that are costing cement and aggregates companies millions every year.

 

Myth 1: A stich in time always saves nine.

In most maintenance handbooks, a low frequency of failure is touted as a key performance indicator of a well-oiled maintenance program.  From this paradigm a belief has manifested than if some preventive maintenance is good, more is better- a better way to flush money down the toilet. What maintenance departments and some accountants typically don’t consider is the overall cost of an equipment’s failure to the business. This includes the repair costs, lost sales (assuming a sold-out situation) and costs of customer defections. Because this data is difficult to obtain it’s easier to give equal attention to all equipment despite the very different overall failure costs.  Take a look at what your plants are preventively maintaining: odds are that 70% of those tasks are a waste of time.

 

Myth 2:  Operate to Failure is not a maintenance strategy.

On the surface, most intuitively don’t look at machine failure as a success. But that’s fertile soil for another myth.  It’s critical to analyze each plant item to ensure that you get the right maintenance policy. Ask questions like: what is the cost to the business if this piece of equipment fails? If it is small, then consider “Operate to Failure” as the strategy.  Most businesses always talk about “Planned” vs “Unplanned” maintenance and think that this means they must have planned maintenance for every item – which is just plain wrong.  Running certain pieces of equipment to failure is a legitimate strategy.

Myth 3: High OEE is a measure of smart maintenance

Overall Equipment Effectiveness (OEE) seeks to obtain a combined measure of a plant’s availability, performance and quality and track how close the plant is running to its “ideal.”  There are many ways to game the OEE calculations and this metric can become very political, especially when its being used to compare several operations around the globe. And when evaluating how maintenance dollars are spent, it’s a dangerous metric to use. There is little correlation between machines that are always available and business profitability. Expenses for improving (or maintaining) OEE should face scrutiny in the form of questions to the impact on cash flow, profitability and payback.  Managing to a target OEE number doesn’t take into account the degree of the cost of failure.  Be wary of frequent maintenance tasks on equipment where failure is not catastrophic.

 

When times are good, practices are usually far less scrutinized, and this is where some of the most powerful myths and misconceptions can be turned into standards. But don’t despair if you can identify with high maintenance costs; with some clear thinking you can have your cake and eat it, too.  Just like Marie Antoinette.

Are We Too Biased for Our Computer Overlords?

The starting NFL quarterback has had a great partial week of practice but, by the end of Wednesday, the coaches decided to tweak the next day’s drills in response to something they see in a sensory download report. The tackle sensors implanted in the pads that log direction, energy, muscle responses and other bio metrics indicate an abnormal change in recovery performance in the trapezium muscles after a vicious hit last Sunday. At precisely with 8 minutes and 31 seconds remaining in the 3rd quarter. The coaches will hold him out of plays that involve passing later in practice and the trainers will modify the physical therapy routine.  As it turns out, the QB plays deep into the next game and doesn’t miss any games for the season.  A successful intervention.

Algorithms that log data like this in sports science have mushroomed to two-thirds of all NFL teams and more than half of the NBA since 2012. Getting to this level of analytics not only serves the players well but also the fans, sponsors and teams.  Using data technology to spot potential opportunities and risk areas will continue to evolve and impact the way teams practice, game plan and substitute.  Those teams that don’t will be at a decided disadvantage.

North American sports are just catching up to the rest of the world in this respect.

Likewise, analytics is the working its way into how we market, operate and distribute as well.  Like the wave of Lean Manufacturing or the wave of outsourcing, U.S. business is now in a productivity wave- headlined by Big Data.  One of the more staggering statistics I have heard just appeared in a recent McKinsey & Co. publication. It stated one-half of all the world’s data was just created in the past 10 months.  Put another way, of all the data that mankind has EVER produced…half of it has come this year!

However, like North American sports, most businesses are having problems adopting to this. One of the issues is that despite paying to create lots of data, businesses are actually capturing very small amounts. Secondly, what is being captured is really not being used.

Does this sound familiar? It should, because we are all wired this way.

When I worked for a large consulting firm, we were trained to recognize the most commonly occurring human biases that influence our judgements and those of our clients. A few of the most well-known biases we exhibit include:

  • Confirmation bias, which refers to a type of selective thinking whereby one tends to notice and to look for what confirms one’s beliefs, and to ignore, or undervalue the relevance of evidence that contradicts one’s beliefs. For example, if you believe that your favorite football team wins big games when wearing orange pants, you will take notice of games played in orange pants, but be inattentive to the uniforms when momentous victories occur with other combinations. (This is sadly a personal example.)
  •  Mere exposure effect is the tendency for people to develop a preference for things merely because they are familiar with them. People will frequently select an alternative simply because they have seen it before, not because it’s the best answer.  It’s because of this bias we are forced to ensure the overused business axiom: “Think outside the box.”
  •  Outcome bias refers to the tendency to judge a decision by its eventual outcome, instead of judging it based on the quality of the decision at the time it was made. For example, a basketball player takes an ill-advised shot with plenty of time left on the clock and holding on to a narrow lead. The shot was a poor choice, regardless of whether it actually falls or not. This decision should be viewed negatively in both cases, but in reality he generally only gets yelled at if something bad happens.
  •  Actor-observer bias refers to a tendency to attribute our own poor outcomes to external causes that can’t be controlled, while attributing other people’s misfortunes to personal reasons. For example, imagine that you are getting ready to take a standardized test to get into graduate school. You fail to observe your own study behaviors (or lack thereof) leading up to the exam, but focus on situational variables that affected your performance on the test (i.e. the room was hot and stuffy, your pencil kept breaking, the student next to you kept making grunting noises, etc.) So when you get your results back and realize you are definitely not getting in to any grad schools, you blame those external distractions for your poor performance instead of acknowledging your own poor study habits. Of course, one of your buddies also did poorly, but you immediately consider how he often skipped prep class, never practiced, and never took notes. Never mind the same conditions.
  •  Illusory correlation refers to the concept of relating two variables even when they are not related. For example, drownings in lakes have been shown to increase when sales of ice cream go up. Therefore, the government earnestly tries to outlaw any and all ice cream products.

A review of the complete list of 105 (search Google for “cognitive biases”) leaves little doubt that in many complex decisions we see what we want to see. Relying on output from a decision support tool like a linear program for situational answers goes against our instincts.

However, there is a clear path to riding this new wave. A large part of the success rides on our ability to override our natural tendencies.

Go beyond gut instinct for doing business. Nobel Award winner Daniel Kahneman, in his terrific book titled Thinking, Fast and Slow, describes our two modes of thinking; what he calls “System 1” which is automatic, instant, intuitive and involuntary, relying on our perceptions of our knowledge and experience; and “System 2” which is more structured, controlled, analytical and effortful.  Because System 1 is automatic and requires little or no effort, we have a natural bias towards its use.  This leads us to sincerely believe a rational choice has been made, when in fact it was not!  We are often unaware of the powerful influence System 1 thinking has on our decision making and its potential for leading an individual or an organization astray.

Decisions have to be analytics oriented. Advanced analytics is the latest term for a concept that has been in use since WWII. To the computer scientist, it is called artificial intelligence; to economists it is modeling; mathematicians call it game theory; business people may use “optimization”. Whatever the term, it is a decision making process that employs mathematics, algorithms and software not only to sort and organize data, but to use that data to make recommendations faster and better than we can.

Some businesses have recently experienced the advantages of sidestepping these biases and using advanced analytics to support their business decisions. For example, a large international manufacturer operating at 100% shipping capacity was wrangling with decisions on which distribution hubs to close or expand as well as which plants should ideally supply them to maximize profits in the current environment.  After optimizing the network, the company could clearly see the benefits of switching supply sources in 6 scenarios and doing so with less purchased rail cars in the process. In addition, potentially new distribution locations were added to the mix and the overall network profitability was re-quantified.  From a cost savings perspective, the results amounted to 4% in bottom line improvement, not to mention to the role optimization played in evaluating their next acquisition.

According to one executive:

“We have been using [the linear program model] a lot, to get better information as to what we can do and if we’ll get any transportation synergies out of the deal. I have to tell you that it’s been really useful. I don’t know what we’d be doing without it…probably just looking at each other and coming out with clever answers (or maybe not that clever) just to get out of the problem of not knowing how to answer to a question.”

Overcoming our natural tendencies in complex decision making is not an easy step to take, especially when it involves relying on analytics that are faster, smarter and bias free.  Analytics has infiltrated our sports and is the latest in business operations. Is this the end of us and the beginning of the Age of the Machine?  Probably not quite yet. But just in case, to steal a line from the Simpson’s:

I, for one, intend to welcome our new robot overlords.