Equipment Efficiency Loss Analysis: The Distinction Between OEE and TEEP

That day, the boss stared at the output report, his brow deeply furrowed

I remember one time when an old Diffusion Furnace on the production line suddenly saw its output drop to an unbearable level. During the meeting, the boss stared at that week's report, his brow deeply furrowed. He directly called me out, "Ah Liang, why is this machine's OEE so bad? Is it a human operation issue, or is the equipment itself faulty? You need to find out!" At that moment, a thought popped into my head: Boss, you're right to talk about OEE, but regarding this machine's 'true potential,' aren't you forgetting something?

Where is the problem? OEE or TEEP?

Frankly, when the boss asked me about OEE, he wanted to know, "How efficient is this machine during its scheduled production time?" This is intuitive: it compares the good products you actually produced against the total quantity you 'planned' to produce.

The calculation for OEE (Overall Equipment Effectiveness) is well-known: it's Availability * Performance * Quality.

• Availability: How long was this machine actually running during its 'scheduled time'? Was it under repair for too long? Did material changes take too long?
• Performance: Did it operate at full speed when running? Did vibration or wear lead to a slowdown in production speed?
• Quality: Were the products produced all good? Was there a lot of rework or scrap?

But frankly, often when we only look at OEE, we overlook a deeper question: How much could this machine 'originally' produce? In other words, if this machine operated non-stop, 24 hours a day, all year round, what would its maximum capacity be? This is what TEEP (Total Effective Equipment Performance) tells you.

TEEP is essentially OEE * Loading Rate (Loading Time / Calendar Time).

In simpler terms, TEEP considers, "From the moment this machine was purchased, how much could it 'theoretically' produce?" It accounts for all time, including unscheduled periods, holidays, and even Chinese New Year breaks.

So the point is: if you only look at OEE, you might think, "Wow! My machine's OEE is 80%, that's great!" But if you only schedule this machine for three days a month, its true potential is far from being realized. In such cases, TEEP provides a more realistic answer.

How to actually do it? Is the equipment itself bad, or is there a problem with how it's used?

Let's take an example. Assume you have a CVD machine:

1. Machine Rated Speed: 10 wafers per hour.
2. Total Annual Time (Calendar Time): 365 days * 24 hours = 8760 hours.
3. Your Scheduled Production Time (Loading Time): You usually only schedule two shifts from Monday to Friday, no shifts on weekends, totaling 5 days * 2 shifts * 8 hours/shift * 52 weeks = 4160 hours.
4. Actual Running Time: Due to machine maintenance, material changes, etc., downtime accounts for 20%, so actual running time is 4160 hours * (1 - 0.2) = 3328 hours.
5. Actual Good Product Output: Assume an average of only 9 wafers per hour, and the quality rate is 95%.

Now let's calculate:

• OEE = (Actual Good Product Output) / (Scheduled Time * Rated Speed)

* Availability = Actual Running Time / Scheduled Time = 3328 / 4160 = 80%

* Performance = (Actual hourly output) / (Rated Speed) = 9 / 10 = 90%

* Quality = 0.95

* OEE = 80% * 90% * 95% = 68.4% (Similar to the result above, difference due to rounding)

• TEEP = (Actual Good Product Output) / (Total Annual Time * Rated Speed)

* Or TEEP = OEE * (Scheduled Time / Total Annual Time) = 68.4% * (4160 / 8760) = 68.4% * 47.5% = 32.5%

Do you see it now? If you only tell your boss that OEE is 68.3%, he might think there's still a lot of room for improvement. But if you simultaneously tell him that TEEP is only 32.4%, this implies a crucial piece of information: This machine is not being fully utilized! Most of the time it's idle. In this case, the problem might not be inefficient equipment, but rather issues with your scheduling strategy or capacity planning. Is the equipment itself "bad" to the point it can't run at full capacity, or are we simply "not letting it run"? OEE tells you the former, TEEP tells you the latter.

The most common pitfall: pretty numbers, but not maximized profit

The most common pitfall I've encountered is people only pursuing high OEE. To make the OEE figures look good, scheduled operating time is often deliberately shortened. For example, even if you know there are no weekend shifts, you might keep the Loading Time very conservative to prevent Availability from being pulled down. The result is that the OEE figures might look impressive, but in reality, the machine has plenty of time available for production, yet it sits idle due to lack of scheduling.

Honestly, this is like an athlete who runs very fast (high OEE performance) and doesn't fall during a race (high OEE quality), but because they only participate in two competitions a year (low TEEP loading rate), they simply can't earn enough prize money. Looking at OEE is like only seeing performance in a single game; looking at TEEP allows you to understand if they are maximizing their athletic career.

One thing you can do today

Go back and look at your machine's OEE report, then ask yourself: "If this machine ran 24/7, what would its TEEP be?"