5 Common OEE Mistakes

If you're measuring Overall Equipment Effectiveness (OEE) it is worth spending a few minutes to verify that you are avoiding the common mistakes that people make when building their OEE data foundation.

These are the 5 areas where we most frequently find problems that undermine accurate OEE data. Do any of these apply to you?

• Monitoring the wrong step (not the constraint)
• Using 'budget' Ideal Cycle Times (not the fastest cycle time)
• Over complicating stop reasons (too many events or reasons)
• Excluding certain stop times (removing 'inconvenient' events)
• Counting rejects incorrectly (including rework in good count)

Monitoring the Wrong Step

OEE should always be monitored at the constraint. The constraint is the equipment that is the slowest step in the process. It's not the machine that's running 'least-well'. Check that: 

• The constraint has been correctly identified.
• All OEE calculations are based on losses at the constraint. 

Interested in learning more? Read this article about improving the constraint. 

Using 'Budget' Ideal Cycle Times

The Ideal Cycle Time (ICT) is the theoretical minimum time to make a single piece at the constraint. It should NOT be a 'budget' or 'standard' time as these often include allocations for losses. Simply - if your OEE Performance is greater than 100%, your ICT is wrong. Check that:

• ICT is based on the constraint.
• ICT is the theoretical minimum time to make a piece.
• There is a unique ICT for each part (not an average time across all parts).

Over complicating Stop Reasons

There are two types of stop loss - unplanned (down time) and planned (typically setups). To capture accurate and useful stop information provide the operator with a small number of reason codes, so that they can focus on running the equipment. It is unrealistic to expect that an operator records 'root causes' for stops as this takes considerable training. Check that:

• The operator is capturing no more than 15 down events per shift (more than that indicates that the threshold between Small Stops and Down Events should probably be adjusted up).
• The operator has no more than 15 down reason codes.
• Setups are consistently measured (e.g., from last part to first part).

Interested in learning more? Read this article about down time tips. 

Excluding Certain Stop Times

OEE is intended to measure how well the equipment runs when it's scheduled to be running. If the equipment is scheduled to run, but for any reason it can't (e.g., not enough operators, no materials, equipment failure, changeover) then this is lost time. It can also be a best practice to include breaks and lunches as lost time, although many companies exclude these events as being beyond the control of the operations team. Check that:

• Shift start and end times are clearly defined and followed.
• The events that truly should be excluded from OEE are clearly defined and consistently followed.  

Counting Rejects Incorrectly

In OEE, a reject is any part with a defect at or after the constraint - even if that part is reworked. In other words, from the perspective of OEE, if a part is not good in the first pass of production it should be counted as a reject. It's important to remember that OEE Quality is a measure of lost time, not material yield. Check that:

• Reject counts include reworked parts.
• Reject counts are taken at and after the constraint.

For more guidance for how to handle rework with OEE review our OEE Implementation Decisions FAQ page.

More Resources

We have some great resources to help you implement and improve your OEE. Here are a few that are most relevant to this email:

• Step-by-Step OEE Implementation Guide
• OEE Frequently Asked Questions
• Three-Step Guide to OEE Improvement

Can We Help?

We are passionate about improvement and we love to learn new things. So if you have any questions, ideas or insights - we would love to hear from you.