OEE vs. OPE – Toward a Holistic Understanding of Your Machine Data

Some manufacturing metrics have stood the test of time. OEE is one of them.

Recently, however, many of the manufacturers we work with have begun to shift their attention from Overall Equipment Effectiveness (OEE) to Overall Process Effectiveness (OPE). If OEE is designed to present a picture of machine performance, OPE broadens the perspective to account for what’s happening around the machine.

Crucially, it accounts for one element that OEE misses: the human.

In this article I’ll define OEE and OPE. I’ll also explain why understanding machine performance (OEE) ultimately means understanding human performance.

Download our Complete Guide to OEE to Improve the Way You Measure & Analyze Equipment Data →

Overall Equipment Effectiveness (OEE) is a measure of how well a machine performs relative to its capacity during scheduled runs. OEE is calculated by multiplying scores for Availability, Performance, and Quality. These scores are each represented as a percentage, and defined as follows.

Availability – The amount of machine uptime during a scheduled period (availability = uptime / scheduled time)

Quality – The number of quality units produced (no rework or defects) divided by the total number of units produced during a scheduled run. (quality = good units / total units)

Performance – A measure of the machine’s throughput relative to its designed capacity

The equation to calculate OEE looks like this:

OEE = Availability * Performance * Quality

In the digital era, measuring OEE is easier than ever. Machine protocols make it easy to extract machine performance data, edge devices multiply the kinds of information engineers can generate from their machines, and self-serve analytics platforms have made this new data easier to interpret. Digital apps can even automate the OEE calculation process.

For more about OEE, check out our Complete Guide to Overall Equipment Effectiveness (OEE).

Your call.

TEEP, or the Total Effective Equipment Performance, is the balance between fully productive time and all time.

Where as OEE accounts for what happens during scheduled production runs, TEEP puts production against a 24/7, 365 sense of time–all of the time your machines could potentially be productive.

TEEP is useful for identifying potential increases in production, as well as for benchmarking your current performance.

Importantly: OPE is not TEEP. They’re totally different metrics.

So which should you measure?

While OEE is important to understand, the Industrial Internet of Things can create blindspots. In many factories, the most common contributors to poor machine efficiency are not reducible to the machines themselves. They’re problems with machine usage.

In other words, if you’re only measuring OEE, you might not be isolating–or even measuring–the factors contributing to low effectiveness.

OPE accounts for the full range of variables and steps that impact a manufacturing process. It’s an end-to-end account of the value stream.

Crucially, OPE integrates machine data with an account of what happens around machines. It’s a representation of human action as well as machine performance.

It helps to look at an example.

Acme Tables make kitchen tables. Tables are made across four stations. The first cuts the lumber to specification. In the second, workers assemble the lumber into tabletops. At the third, the tabletops go through a processing station that applies finish and sealant. And at the final station, a worker attaches legs.

In Acme’s factory, throughput starts to decrease. An engineer notices buffers increasing before and after finish is applied to an assembled table top. Her first suspicion is that one of the machines isn’t performing optimally.

When she checks the machine data, however, it hasn’t changed. Work-in-progress is moving through the machines as before, but with fewer completed tables at the end.

So what’s the cause?

To move beyond OEE, she asks the following questions. Are parts moving between stations as quickly as possible? Are changeovers being performed efficiently and correctly? Are the right tools available when needed? Is there a particular step up or downstream that’s slowing production? Are the humans in between machine processes working efficiently? Are quality tests happening as quickly as prescribed?

These are just a few of the reasons why production could have slowed, and they’re all factors that can’t be accounted for by OEE. They’re all issues that can’t be reduced to machine availability, quality, or performance.

So in order to answer these questions, you need to be collecting data at each step of production. The best way to accurately balance lines and optimize the flow of work through the value stream is to understand what’s happening at every stage. True process visibility requires looking at the human and machine factors in an operation.