Corso Systems CEO Alex Marcy’s stomach lurched like he was in free fall. He was visiting a new city, and had just blurted out what he assumed to be the most obvious piece of information in the world - only to receive a look of complete bewilderment from the group.

They were trying to choose where to go for dinner. Someone mentioned they were really craving a good bowl of soup and wondered what the soup of the day was at one of the restaurants on the list.

“It’s Thursday. The soup of the day is going to be broccoli cheese.”

Keep in mind, no one in the group had ever been to this restaurant before questions were in order.

“No one here as been to this restaurant, how could you possibly know that will be the soup of the day?” someone in the group asked him.

Alex triumphantly replied, “The soup of the day schedule: Tuesday is chicken noodle, Friday is clam chowder, Thursday is usually something with broccoli…”

a woman eating a bowl of soup, is it one that Alex assumed would be the soup of the day? What will the data say?

He trailed off unable to name the rest of the days of the week, and quickly turned to his phone for internet-based proof that he wasn’t completely off his rocker, only to find no sources to back him up.

Tracing the logic of this concept (something he had held true for 28 years of his life at this point) it all made sense. When Alex was six, his Dad had had surgery, and his Uncle said “Awesome! It’s Tuesday, we’ll go to Lyon’s Restaurant and get chicken noodle, the best soup for when you are recovering.”

Friday was often clam chowder, something Alex was keen to steal from his Dad whenever he ordered it, and beyond that soup wasn’t something that his family usually ordered. Alex’s parents were raised by children of the Great Depression who would extend leftovers throughout the week ending in some form of soup by the weekend. Once they were free of their parents’ menu choices, they scoffed at eating soup given their childhood associations with stretching every penny to its fullest.

Until now, Alex’s assumption about a national soup schedule had never been tested. It was simply a belief that had held up to the extremely minor level of scrutiny he had given it his entire life. But now, he had never felt more deflated or embarrassed.

They ended up going to that restaurant for dinner and no, the soup of the day was not broccoli cheddar, it was cream of mushroom. At least it was a delicious bowl of soup for the person craving it.

When You Assume, You Make…

Unlike our other recent posts, the story above has nothing to do with manufacturing specifically. Although the concept of using unproven assumptions to make decisions is a daily issue at many manufacturing companies.

Manufacturing operations are built on countless assumptions, from production schedules and quality standards to equipment maintenance intervals and workforce productivity. While these assumptions often develop from experience and industry best practices, relying on them without validation can lead to inefficiencies, unexpected downtime, and missed opportunities for improvement. Data-driven decision making offers a powerful alternative, allowing facility managers to test their assumptions against reality and make more informed choices.

two people walking in a warehouse, one carrying a scanner and the other with a clipboard

Assumptions like these may not have a big impact on your overall operations, they could also be hugely detrimental. Luckily using tools like Ignition to collect data from your equipment and track values over time can help you reduce the amount of assumptions you use to run your business. This is also a great application for using a Unified Namespace (UNS) to help add context to your production data from all across your business. Using a UNS can help you spot where your assumptions are causing you problem and give you additional inputs you can use to address the root cause of problems head on.

Using Data to Test Assumptions

The key in making your data work for you is to identify which assumptions have the greatest impact on your operations and systematically test them with the data you have available. This exercise might even inspire you to collect new data from your equipment to help make more informed decisions.

For example, you might assume that a particular machine requires maintenance every 500 operating hours, but analyzing actual performance data could reveal that extending intervals to 600 hours can maintain reliability while reducing maintenance costs. Similarly, production scheduling assumptions about bottlenecks, changeover times, or optimal batch sizes can be validated or corrected using historical production data, real-time monitoring systems, and performance metrics.

A series of three large robot arms on a production line

The goal isn't to eliminate all assumptions. That's neither practical nor necessary. The goal is to ensure that your most critical decisions are grounded in evidence rather than habit. By regularly questioning and testing assumptions with data, you create a culture of continuous improvement leading to significant gains in efficiency, quality, and profitability. And this all happens while reducing the risks that come with operating on outdated or incorrect beliefs about how your facility actually performs.

Transforming Assumptions Into a Hypothesis

As nearly everyone learned in school, the scientific method is an important learning tool. Rather than making assumptions about your operations and running with them until something breaks, you can use data to help make a hypothesis about how things are operating and test it with a series of experiments.

Start by collecting baseline data on critical processes and outcomes, then design simple experiments to test your most important assumptions. This might involve adjusting parameters within safe ranges and measuring the results, or analyzing patterns in existing data to identify discrepancies between expected and actual performance. Quality control data, equipment sensors, production logs, and even employee feedback can all provide valuable insights that challenge or confirm operational assumptions.

Taking a more scientific approach to understanding the reality of your processes to confirm or adjust your beliefs with data will help you make better informed decisions (along with the receipts to back them up). If your assumptions turn out to be correct, that’s great and you can continue operating with the same framework if it is working for you. If they turn out to be wrong, you will better understand why they are wrong, and learn how to make better informed decisions for your process with new information.

Wrapping Up

With tools like Ignition available to every manufacturing company, having access to usable data is an easily solved problem. Even if you currently have an entirely manual process or use machines without PLCs, you can integrate machine monitoring solutions with a low cost groov RIO from Opto 22.

Retriving data from other sources like your quality control management systems, ERP systems, and even employee feedback is a relatively easy task in Ignition and will provide very important context about your production and real-time process data.

Once you are collecting data, you can utilize your choice of trending and reporting tools in Ignition along with other tools like Excel to help you make sense of all of it. Now, you will have everything you need to start understanding and challenging your assumptions to ultimately make your manufacturing operation run as smoothly as possible!