How Machine Learning Works with Existing Operations Data
Having data and metrics is great. But there comes a point when so much data is collected that the user has no clue what to make of it. Data overload can be a serious problem. Here’s a great example to show why.
While visiting a client site recently, we asked the VP of operations how their current MES solution was working and if it was providing what they needed. He told us that, while it was much better than the old days of putting everything on paper, many of his plant managers and supervisors spend six to seven hours a day interpreting data.
When a production issue occurred, they could easily see the problem. Reports and dashboards were everywhere. Giant TVs hung with pride on the shop floor. But understanding why the issue occurred was not easily seen. All their reports and dashboards also didn’t tell them how the problem could be prevented next time.
As a result, the VP said his team would have to spend countless hours diving into the data to figure out the root cause to any problem. This pulled them away from the shop floor where he wanted them to spend more time, making sure quality product was produced.
The VP also pointed out that shop floor operators were now ignoring alarms because they didn’t always trust the data or found it lacking context.
Transforming production data into operational directions
With these issues in mind, his request was simple: Can a new solution run the plant?
He wasn’t advocating for the removal of all human interaction, but rather challenging us to figure out how we could leverage all the data captured by the MES to provide his team with recommendations based on concise, actionable data? They wanted a dashboard that would show what they need to do to complete their job, while not forcing them to become data analysts.
Machine leaning (ML) is a method by which machines improve their performance over time using data generated during production. ML algorithms detect patterns and trends in data, allowing them to make predictions or decisions based on this information.
This is where machine learning (ML) comes to the aid of manufacturers. ML is a subset of AI (artificial intelligence) and refers to the ability of machines to learn from data without being explicitly programmed. It is a method by which machines improve their performance over time using data generated during production. Machine learning algorithms detect patterns and trends in data, allowing them to make predictions or decisions based on this information.
ML is a popular topic at the moment and many companies have started implementing programs centered around it. The big question is, how many companies are ready?
How machine learning works and the role of MES in it
To answer that, it’s important to understand that ML first works with pre-existing data. If you feed the system bad data, you are not going to like the results. After all, the system is only going to be as accurate as the data it’s given.
When our clients discuss their desire to implement ML solutions, it’s critical to first establish that their current systems provide accurate, real time data from the production process. Operations, performance and quality data are foundational elements regardless of how they are collected.
I still believe strongly that MES is relevant today. In fact, it’s more relevant than ever. While the focus is shifting rightfully to next-gen AI and ML solutions, we can’t lose sight of the day-to-day operations of a manufacturing plant and how to harness its data. MES fits well into the overall software architecture as a key component of any advanced manufacturing data strategy. It’s critical that any digital manufacturing strategy defines how data will be collected from the shop floor equipment, systems and humans. Skipping this step will potentially negatively impact any other initiative.
This is why, without MES, it’s impossible to successfully adopt AI and ML solutions.
With so much focus on data initiatives, many companies are jumping in headfirst to apply AI and ML. But it’s important to remember that these programs take time and require a clear strategy and roadmap. That’s why Actemium takes a consultative approach to help our clients define the challenges facing them and solve them. We also focus on creating the foundation to help correctly implement new technologies in the future. Our digital transformation team is made up of experts in automation and controls, MES and DataOps. We believe bringing an experienced cross functional team together provides our clients not only value but de-risks their initiatives.
Dan Purcell is senior account manager at Actemium Avanceon LLC, a certified member of the Control System Integrators Association (CSIA). For more information about Avanceon, visit its profile on the CSIA Industrial Automation Exchange.
Leaders relevant to this article: