A Single Machine Data Point is More Insightful than You Think
Everyone talks about IoT, IIoT, Big Data, and “AI for manufacturing”. Regardless of the fact that these are buzzwords (hey, even you digital twin), it means lots of machine data, right? That data all has to go somewhere. But, where?
One of the most common questions we get at Mingo is,
“Will our internet connection be able to handle all of the data we send to the cloud?”
(The answer is yes, you are equipped to handle a large amount of data, and you will see why).
There are so many misconceptions about machine data and how much you need to do something meaningful. Most companies today are not automatically collecting any data and they are using Excel and paper to collect data manually. That leaves room for lots of improvement.
The reality is you can do a lot with a single machine data point and a tool designed to help you contextualize it, like Mingo. It doesn’t have to be complicated or expensive. A single data point can go a long way for manufacturers. A single data point can get you the insight needed to increase efficiency on the floor.
A Common Scenario In Manufacturing
Let’s set up a scenario.
A company wants to reduce downtime, and count cycles on some older equipment that does not have a PLC. They retrofit these old machines with wireless sensors to count each time the machine cycles. This single data point, a cycle for the machine, is sent to a contextual manufacturing analytics system each time it occurs.
Using this one data point we can track cycles on the machine; which can give us a part count. So far, with this one data point, we know how many cycles the machine performed and how much product is produced.
Automatic Downtime Detection
Let’s take this one step further.
Inside of manufacturing analytics tools, like Mingo, you can configure what we call a shortstop.
This is a setting that allows manufacturers to automatically detect downtime if it does not see a signal for a certain percentage of the cycle time.
For example, if the cycle time of the part is 15 seconds per part and Mingo does not see a signal from the machine for 30 seconds, it will mark it as down.
This is a good start, now we can track cycles, part counts, and availability of the machine all from a single machine data point. A huge gain in the efficiency sense.
Prompt for Machine Downtime Reasons
But, we can do more.
When Mingo detects a shortstop or downtime, it can prompt the operator to enter a downtime reason code.
They can do this with a simple touchscreen interface on a tablet or computer mounted next to the machine.
With this data, we have something really powerful.
We can track downtime with reason codes, which can help improve utilization.
We can track cycles which will give us part counts and help us understand throughput and capacity.
And we can do all of this with a single machine data point. Again, a single data point is more powerful and insightful than most manufacturers realize.
Calculate OEE
Let’s talk about one more example. Say a manufacturer wants to figure out a complete OEE calculation. Using the scenario from above, a manufacturer has already tracked availability and downtime, with scrap reason codes inputted by operators on the floor.
Companies whose operators also do quality checks on the line once an hour will enter reason codes for scrap.
With that one manual data point, you’re able to calculate quality (using the availability and performance numbers we calculated from above) which can be used to complete an OEE calculation, a huge win for you.
(That said, a single data point can be manually entered or automatically collected, but with the right software, the insights can transform the floor.)
Examples of a Single Machine Data Point in Use
For food process manufacturers or beverage manufacturers, it would be common to add battery-operated photo eye sensors to count the packages and/or bottles as they’re going by. A single data point (the number of items) can tell that manufacturer if they’ve produced on schedule or not.
As a metal stamper, counting strokes on the press with a discrete wireless sensor would be the recommended path. A single data point gives the manufacturer insight into the productivity of the press.
In both cases, operators use a tablet to record scrap and reason codes to better track the data, giving insight into when a machine is down or production is not on track.
The overall consensus is that a single data point is more beneficial than most manufacturers realize.
A Single Machine Data Point is a Great Place to Start
If you’re not using machine data this way, starting with a single machine data point and a single goal is a right place to begin. That gets you on track to start small, think big, move fast, the central strategy for achieving success with manufacturing analytics.
Often, manufacturers will get jaded by all the people in the space talking about AI and IIoT (we’ve got a guide to help you with IIoT, though) and feel like it is really hard to get started with any of that with where they’re currently at. It can be confusing, but it doesn’t have to be.
That’s why you start with something simple. See how it works, adapt your organization to looking at data this way, and understand how you get ROI out of these types of projects. Manufacturing analytics can help improve efficiency and productivity on the factory floor, but you have to start somewhere – what better way than with a single data point?
