Dr. Filiep Samyn
Senior Operations Analyst
View Profile

“Are you using the right metrics?”

“The goal of the business is not to show-off its efficiencies.”

The Goal (Goldratt & Cox, 2004)

The Overall Equipment Effectiveness (OEE) metric has enjoyed increasing popularity because of its perceived simplicity.  Indeed, it allows decision makers to understand how effective the production equipment is used to create sellable product.  However, the use of a single number to quantify complex processes is not without danger. 

OEE Calculation Variations

OEE results from multiplying three metrics:  availability, performance and quality.  There is consensus about the three components but views diverge as how to define and calculate each of them.  It is IE’s experience that clients use their own definition of OEE, sometimes with the intent to inflate OEE numbers.

  • Availability is the ratio of actual operating time versus planned operating time.  The actual operating time is the planned operating time minus setup time, scheduled downtime, and unscheduled downtime.
  • Performance, also called operating level, is the ratio of actual throughput versus rated throughput.  Operating level uses hours in the calculation and considers available hours and lost time due to running below rated speed. 
  • Quality level is the ratio of good products versus total products which includes good product, scrap and rework, based on production over a given period. 

The planned operating time (aka production time) is typically defined as calendar time minus market or demand losses.  This means that we only consider equipment time needed to meet market demand.  It is important to note that some operations are not market constrained, all product can be sold (example gold production).  How to come up with the needed production time is often open for interpretation. 

We need to keep in mind that:


It is easy to see that planned operating time will affect the availability level; hence, the analyst can influence the calculation.  There are variations on the calculation of availability. For instance, King (2009) stated “setup time or changeover time should not include time getting properties back within specifications after changeover” (p. 115). IE analysts would argue with this interpretation because changeover time is defined as the time lapse from last good product produced to first good product produced.

Three Critical Elements of OEE Performance Evaluation

  1. Observed Consistent Performance

Design throughput is what the manufacturer claimed the equipment is capable of.  Based on IE’s observations, this number can deviate positively or negatively from what is actually and consistently possible.  It is therefore necessary to validate ‘possible throughput’, i.e., what the equipment is capable of producing consistently.  Failing to quantify true equipment capability may lead to two problems: 

  • If underestimated, decision makers may invest in unneeded new equipment
  • If overestimated, equipment may be pushed beyond its limits and increase maintenance cost as a non-linear function of throughput 

Only observed, consistent performance while stressing the equipment reveals the true possible performance, i.e., the technical limit.  IE is often engaged by clients to derive accurate equipment technical limits.

2. Material Grade

Both grades will be sold but the benefit is higher for A-grade.  This is not really a quality loss as typically defined (it is not bad product and it does not require rework) but rather an opportunity loss.

To clarify this further, let us take an example where the equipment manufacturer has specified a capability of 100 tons/hr. of A-grade material.  For simplicity we assume that manufacturer ‘sticker plate’ capacity equals the technical limit (which, in reality, is often not the case).  The current output is 80 tons/hour of A-grade material and 20 tons/hour of the easier to produce B-grade material.  Without appreciating the subtlety of material mix in assessing capability, one may be tempted to state that ‘sticker plate’ capacity has been reached when in fact, it has not!

3. OEE Comparisons Require “Similarity”

Improving OEE toward a target requires analyzing the composing factors over a sufficiently large time span.  These factors include similar equipment and similar materials used to produce similar finished goods. Similarity is mandatory to make meaningful comparisons and helps decision makers identify improvement opportunities.

Additionally, decision makers who compare OEE between identical operations without considering demand, equipment capabilities, and other differences, may set unrealistic improvement targets.  For example, assume two identical manufacturing setups but with vastly different demand and required operating time. If the operation with the lower operating time increases its availability by investing in preventative maintenance, unplanned downtime will be reduced.  And because unplanned downtime has a more detrimental impact on total downtime than planned downtime, availability (and OEE) for the setup with lower demand will likely be higher. 

Avoid Pitfalls of OEE-Based Improvements

Pursuing OEE-based improvements need to make business sense.  Before embarking on any OEE improvement initiatives, an objective analysis of bottom line impacts is recommended. The following are three examples of how OEE-based improvement may actually generate negative results.

Example #1.  To ensure an investment will drive bottom line improvements, system constraints need to be taken into consideration.  In one of our recent client engagements, IE witnessed a maintenance manager investing in costly measures to increase availability.  While these measures increased availability, their ability to increase throughput was constrained by a fixed ore supply from the mine.  And although there was an uptick in OEE, profitability declined due to higher maintenance costs! “The goal of the business is to make money” (Goldratt & Cox, 2004) or put differently, decisions need to make business sense and cannot focus on a single metric.  If increased availability comes at a great cost without selling more product then the net ROI is negative.

Example #2.  Similarly, a myopic focus on OEE improvement often ignores schedule attainment, i.e. the need to produce in accordance with customer demand.  Increasing throughput by running the equipment closer to its technical limit (everything else remaining the same) will produce more product, however, without market demand this improved OEE will result in overproduction, one of the seven forms of waste.  As an example, assume that a piece of equipment can produce A-grade material at an OEE of 50% and B-grade material at an OEE of 75%. The OEE difference is the result of differences in quality losses, speed losses, downtime, or a combination thereof. The overall OEE then becomes:

It is obvious that producing more B-grade material increases overall OEE; however, with a higher demand for A-grade material pursuing higher OEE will result in excessive (stored) B-grade material and not meeting customer demand for A-grade material.  Although OEE has increased, schedule attainment has suffered! Again, focusing on a single metric can lead to reduced profitability and market share loss.

Example #3.  Asset utilization provides additional insight that OEE was not designed to do.  Let’s assume that current daily demand results in planned operating time of four hours for three identical pieces of equipment.  If you chose to use only one piece of equipment for 12 hours, the availability will remain the same (assuming downtime remains unchanged).  

As a simple illustration we use the above figures of 12 and four hours, and three hours downtime for 12 operating hours.  Performance and quality levels are estimated to be 90%. Using one piece of equipment we obtain:

Using three identical pieces of equipment with equal load we obtain:

As expected, the OEE for both situations are identical (60.75%) and provides no indication if we should make a change.  On the other hand, let’s take into consideration asset utilization, which is defined as:

If we assume that capacity is consistently 100 tons/hour for maximum daily output is 2,400 tons.  Using one piece of equipment we have nine hours of production resulting in: 

The losses that we observe are due to availability losses and market demand losses.  Using the same calculation for the three-equipment setup results in 12.5% asset utilization.

The low asset utilization for this last situation should incite the manager to decommission, either temporarily or permanently, two pieces of equipment.  And because of the one-piece setup, there is enough cushion to cope with incremental increases in demand. 

Use OEE to Benefit Your Business

Before embarking on OEE-based improvement opportunities resulting from a comparative study or an industry benchmark, the different components of OEE need to be understood fully. It is imperative to consider the following before heading down the decision path. 

Visit to learn more about Implementation Engineers’ capabilities and methodology.

Implementation Engineers is a business-optimization firm that specializes in accelerated performance and improved financial gains while providing immediate results. We differentiate our services by being solely focused on implementation. For this reason, our services go Beyond Consulting SM.

References:  Bicheno, J., & Holweg, M. (2009). The Lean Toolbox (4th ed.). Buckingham, England: PICSIE Books, Goldratt, E.M., & Cox, J (2004), The Goal (3rd ed.). Great Barrington, M A: North River Press, King, P.L. (2009), Lean for the Process Industries, New York, NY: CRC Press, Suzuki, T. (1994). TPM In Process Industries. Portland, OR: Productivity Press

Subscribe To Newsletter