OEE Made Easy

Jun 23

Fifteen years ago, most manufacturing companies had come around to the idea of using SCADA systems to run their processes. Most of these early SCADA systems had built in trending tools with functional data collection systems, but weren’t doing more than recording data to throw on a graph. The ActiveFactory product in the Wonderware portfolio was the first data analysis tool that many companies experienced. ActiveFactory tied into historical data with some advanced trending, calculation, and reporting tools. It was great if you had a Wonderware system and were using their historian, but proved difficult to use if you were on any other SCADA system. The folks who originally built ActiveFactory spun off to build a product called Incuity, which was basically a more advanced version of ActiveFactory that could connect to anything. Rockwell bought Incuity after Wonderware released it, and it became FactoryTalk VantagePoint which is still used today—even though it is not very far removed from what Incuity looked like in 2008.

Digitalization is Continuous Improvement by Another Name

Why is this history lesson about data analysis software from fifteen years ago is relevant to OEE?

It is relevant because we are experiencing the same revolution right now. Instead of process historians, SCADA systems, and dashboard tools, companies are trying to figure out how they can leverage OEE to benefit their bottom line.

OEE is hardly a new concept. Manufacturing companies have been using it for years to boil down the performance of complex processes into a single metric. OEE is a Key Performance Indicator (KPI) companies can easily improve in many cases once they understand where that KPI is coming from.

Hurdles to implementing OEE:

Understanding how a single number can benefit your bottom line
Data acquisition
Calculating metrics
Analyzing the data
Profit???

Using KPIs to Optimize Operations

The first hurdle—understanding how a single number can transform your organization—will be described in detail over the rest of this series. The rest of this post will give you a high level overview to set up thought processes for the solutions to the rest of the list.

We’ll start with an example from our popular MES Essentials Webinar video which is also available at the end of this post.

OEE Calculation Breakdown Example — Here we see the three values we use to generate our OEE KPI. Availability, Performance and Quality.

Availability

Availability is your overall uptime vs. your planned production time. If your process was running 100% of the time it was supposed to be running, then your availability would be 100%.

In reality, you will likely have various downtime events. A machine might be have a mechanical problem, or starved for new material, or even blocked by a downstream machine that is jammed up. Each of these events will be logged by your process historian, and the OEE Engine used by your SCADA and MES system will sum these up to give you an overall downtime amount.

You can then examine the overall downtime by reason code to understand which downtime events are causing the most issues. Once you know what’s causing the problem, you can prioritize fixing the problem.

To understand downtime, you will need to have logic in your PLC or SCADA system to classify when and why a machine is down. This subject will be covered in more detail in a future post.

Performance

Performance is a little more complex than Availability, but is simply “how many units did I produce in a given time period vs. how many could I produce in the same period under ideal conditions”.

Performance is more complex because it will be affected by downtime, and can also be affected by machine slowdowns, operator input, quality of materials, or any number of other factors. Performance can also be difficult to measure in processes that are not based on individual units. For example it is easy to understand “We made 100 widgets this hour, and we could make 120 under ideal conditions”. Applying this concept to “we produced 10,000 lbs of sausage today and we could ideally produce 15,000 lbs” can be a little trickier depending on how the OEE Calculation Engine is configured.

Performance is also harder to quantify with improvements because you will need additional data to provide context: What were the machine setpoints? Did any downtime that cause slowdowns? Are there supply chain issues, or any number of other issues?

Understanding how to best determine performance and how to improve it will be covered in a future post.

Quality

Quality is deceptively easy to quantify. How many “good” units did we produce, vs. how many total units did we produce? It’s easy because most processes have some form of quality control, so getting the values is simple. Quality can get complex because many times, quality control validation only occurs sometime after production, requiring the data and OEE calculations to be updated at a later.

This usually means you will not have completely accurate real-time OEE data, so you will need to delay your reporting and analysis until after the Quality data is tabulated. This can be hours or even days after production in some cases.

Quality is also trickier than downtime to quantify, because you will need to understand why your products failed QA/QC so you can determine how to fix the issues.

While obtaining quality is usually the easiest of the three metrics, doing something with this information can be the most complex. Luckily, we will cover how to do this in a future post.

OEE

Finally, to calculate OEE you simply multiply Availability, Performance, and Quality.

Throw OEE on a TV in the plant, and watch your productivity and profits soar!*

*It isn’t usually that easy, which is why we are giving you the information you need to use OEE to actually optimize your operation!

Analyze Your Data and Do Something About It

The real value from OEE comes after you get the value itself. From there you can start to dig into why the value is what it is, and how to improve it. This is where data analysis tools come into play.

For OEE, data analysis tools usually are dashboards and reports. Common examples are reports like “Top 5 Downtime Reasons” which can let you know what is causing the most problems, or “Top 5 Quality Issues”.

Generating these reports is not the difficult part.

The hard part is when you start to fix the issues to actually improve OEE. This usually involves a culture shift at a company. What got you here will not get you to where you want to be, and change can be difficult for people to embrace. This is where a lot of OEE implementations fall flat.

Fortunately, Corso Systems is well versed in helping people use OEE to their advantage. We understand how to help shift your team’s mindset, and use information and data to back up your decisions.