Three Manufacturing Efficiency Measurement Mistakes That Are Probably Costing You Money

An efficiency measure, normally OEE, is one of the first KPIs most manufacturing organizations aim to put in place. Unfortunately, common implementation mistakes can lead to major missed opportunities. There are three mistakes that are so frequent, that they are the rule rather than the exception. Common mistake 1: Using The Wrong Bottleneck Speed...

An efficiency measure, normally OEE, is one of the first KPIs most manufacturing organizations aim to put in place. Unfortunately, common implementation mistakes can lead to major missed opportunities.

There are three mistakes that are so frequent, that they are the rule rather than the exception.

Common mistake 1:  Using The Wrong Bottleneck Speed

The problem

Efficiency losses fall into three loss categories:

  • Downtime (Availability in OEE)
  • Speed losses (Performance rate in OEE)
  • Lost time producing out of specification product (Quality rate in OEE)

When you are standing next to a working production process, there is one loss that is much more visible than the other two. That loss is downtime.

Downtime is highly visible and very offensive to operators and managers alike (it is also hard work).

Often there are chronic process problems that cause downtime and these problems often become dramatically worse as speed increases. The common reaction to “hard to solve” process problems causing downtime is to reduce the production speed of the troublesome machines. When you do this, downtime levels drop and everyone feels a lot more comfortable.

After a period of running like this the reduced speed often becomes the “standard”.  This reduced speed will be written into machine settings and people can quickly forget that it’s even possible to run at a higher speed. It is not uncommon to see the bottleneck process running at just 50% of the rated speed for that machine.

Having been in this situation many times, I know from experience that when you challenge this kind of “slow running” the response you normally get is “Yeah, we can run at that speed but everything we produce is junk!”.

Solution

No one is suggesting that you just turn up speed and produce poor quality product. But, if you are confident that the process has the potential to run at a higher speed than it is currently running at, you should calculate your OEE based on the maximum bottleneck speed you know to be possible, even if it is plagued with problems and waste product at that speed. Doing this will highlight that there is scope for problem solving to eliminate the underlying issues. If the lost efficiency due to slow-running is then flagged as one of your top losses, the team should focus on solving those speed-related problems to root cause.

If you don’t know what the maximum potential speed of your processes are, then you need to run a “speed trial”.

This is a carefully supervised exercise where you gradually increase the bottleneck process speed to see at what point the process tops-out. Speed trials often cause understandable anxiety, particularly amongst the maintenance team, but if they are managed properly, they can be a valuable tool for showing true process potential. I found it not uncommon to discover the maximum bottleneck speed is considerably higher than the manufacturer’s stated figures.

Common Mistake 2: Ignoring Micro-Stoppages

The problem

Measuring downtime using manual data-capture sheets is a common way to capture downtime data. It’s a good approach, as the operator can add notes, comments and observations which would be completely missed out by an automated data-capture system. The problem with manual data-capture is that operators are often extremely busy people and they will miss out “micro-stoppages”.  Micro-stoppages are tiny periods, often only a few seconds, of downtime that may occur hundreds or even thousands of times a day. They are often a persistent feature in the process and happen so frequently that the operator barely even notices them. Micro-stoppages are almost never written down but can cripple performance of the production process.

Solution

Fortunately micro stoppages are normally pretty easy to quantify once you are aware of their existence. The best solution is to stand by the production process for as long as necessary, normally a couple of hours will do it, and record an average time for a selection of micro-stoppages, also keeping a tally of the total number of micro-stoppages over the period.

If your micro stoppages are sporadic, you have a little bit more of a challenge on your hands. The best solution is usually to work closely with the process operators. The operators need to be on-board with why micro-stoppages are an issue. Once the operators are bought-in they are normally happy to keep a manual tally over a fixed period of time, to give some solid data. 

Common Mistake 3: Hiding Our True ‘Available Run-Time’

The problem

Because KPIs are viewed by many as a ‘way to keep score’, teams often want them to be as high as possible.

If you are in the fortunate position of being able to sell all of your production output (production constrained), it’s important that the production process runs for the maximum number of hours available. A really common way in which opportunities are hidden is to exclude a number of hours in the day for ‘non-production purposes’. It’s not uncommon to see stoppages due to…

  • coffee breaks
  • lunch breaks
  • cleaning
  • machine maintenance
  • changeovers

… excluded from the ‘Available hours’ in the efficiency calculation. Why are they excluded? The normal reason is because “It’s not fair to include them”.

The truth is, if you need a machine to be running for as many hours as possible, every hour that you are not running represents a potential opportunity. It may not be possible currently to run through cleaning, coffee breaks or changeovers, but are we really saying that there is no way we could ever improve our operating practices or shorten changeovers to just a few seconds?

By taking those stoppages out of our OEE calculation we are taking the spotlight off those opportunities and saying “there’s nothing we can do to improve these”.

Solution

Include all hours in your OEE calculation, then break those ‘Unavoidable’ hours of downtime down by reason. If we are spending 12 hours a week on changeovers, perhaps it’s time we used SMED and worked on reducing those changeover times, rather than just excluding them.

Why go through this pain?

All of the solutions suggested here will make your OEE figures worse in the short-term, but will uncover opportunities that may have been previously hidden. Why do this? Ask yourself this question ‘Which team is most likely to improve, one that thinks it is 98% efficient or one that knows it is 50% efficient?”.

More Questions? Here Are Some Additional OEE and Process KPI Resources

Plain English guide to OEE 
OEE Cheat Sheet 
Process KPIs, the pain-free way

About The Author - Bernie Smith

Bernie has helped his clients deliver surprising levels of improvement across a wide range of industries over the past 20 years. His mission is to help clients with a repeatable, practical and jargon-free method for generating insightful and clear KPIs and management reports. He understands that most people don’t get excited by KPIs, but believes it’s a curable condition.

Source: blog.envisionable.com