CHICAGO – As manufacturers accelerate investments in automation and artificial intelligence (AI), industry experts are urging companies to assess their operational readiness before deploying new technologies across their factories. Energy requirements, physical AI testing and the identification of practical use cases are among the challenges manufacturers must consider before committing to large-scale implementation.
At the International Manufacturing Technology Show (IMTS) in Chicago, Sven Diedrich, head of digital transformation and business solutions at Pinaxis, highlighted four fundamental elements that determine whether automation initiatives can succeed: people, processes, data and systems.
Diedrich describes these interconnected components as the “four pillars” of automation. He said manufacturers frequently struggle because each element develops independently, leaving gaps that prevent technologies from delivering their expected benefits.
“Are your people, processes, data, and systems actually aligned, or do they improve on their own in silos?” he asked, describing the disconnects he regularly encounters across industrial organizations.
One common mistake is allowing the appeal of new technology to dictate automation strategy. Robotics, software applications and AI each offer specific advantages, but selecting tools before understanding the broader operational requirements can undermine the overall project.
“When companies start thinking about automation, it’s very, very easy to begin with the technology because that’s the shiny new object,” Diedrich said. “Everybody wants to get involved.”
Manufacturers might introduce robotics to increase throughput, software applications to improve operational visibility, or AI to strengthen predictive capabilities. Although each investment can create value, none provides a complete picture of the organization when implemented in isolation.
For Diedrich, the benefits depend on connecting these technologies so they can deliver the visibility, decision-making capabilities and operational value businesses expect.
Why disconnected automation projects struggle
Diedrich illustrated the problem with a hypothetical manufacturing company facing a capacity constraint. Engineering, IT and other departments propose an automated solution, only to discover later that the available data is inadequate and the proposed processes do not comply with established company procedures.
The result can be an expensive investment that fails to deliver the expected operational improvements.
“You can have a high-dollar investment in the latest and greatest technology and still create a mediocre automation environment because you miss the connected pieces,” he said.
An honest assessment of workforce capabilities, operational procedures, data quality and existing systems before implementation can help companies avoid unnecessary expenses, delays and complications.
1. People: Set out responsibilities and get the team ready
Companies often forget the human dimension of automation, focusing on equipment, software, business cases and return-on-investment calculations.
“Manufacturers have to understand who owns what process, who runs the automated solution, who provides the technical support and who deals with problems when they arise,” Diedrich said.
The people that are going to use and maintain the technology need to be involved in the planning process from the beginning, not after the equipment has been installed.
Skills development is just as important. Automation doesn’t just take away work, it changes what work employees do, and it adds new technical requirements.
“Automation doesn’t just cut down on work,” said Diedrich. “It brings different challenges, different skill sets and we need to make automation work and support it.”
Knowledge gaps and new responsibilities and workflows must be identified and addressed through collaboration between IT and operational technology (OT) teams. The manufacturers are making systems that are more and more connected and intelligent and it is critical to have better coordination between these departments.
Another consideration for companies is whether their current workforce has the capability to implement and sustain the technology. Installing a new application doesn’t automatically mean employees can take on its responsibilities without further preparation or resources.
Someone has to take care of the application, the root cause analysis of repeated exceptions, train new employees and improve the system continuously.
Those responsibilities stay, even when the process underneath is automated. Manufacturers must therefore consider the ongoing work to maintain performance and provide employees with the requisite time and expertise.”
2. Processes: Fix flaws before automating
When employees intervene manually, they mask operational weaknesses that automation can reveal.
Experienced operators can compensate for inefficient procedures or unexpected problems with knowledge and quick judgment. Conversely , machines tend to follow the instructions and rules they have been given and may not respond to changing circumstances with the same flexibility .
“You have an operator who might know how to fix certain issues really quick,” Diedrich said, noting that automation removes the human intervention that can mask shortcomings in current processes.
So when the task is automated, weaknesses could show up much earlier.
Manufacturers must take clear ownership of system availability, data management, continuous improvement and other critical operational responsibilities. If no one is held accountable, it can lead to different teams thinking that someone else will fix the problems.
If everybody owns just a little bit, something here, a little bit there, nobody owns anything, and we are setting ourselves up for failure and chaos,” said Diedrich. Hence clear rules and responsibilities are essential for the reliable operation.
Companies need to be careful not to simply move inefficient workflows straight into automated systems.
“Automation can make good process very efficient, but it can also make bad process very efficient as well,” he said.
Robots can create more waste reliably, software can digitize inefficient processes without uncovering the underlying problems. If you speed up or standardize a bad process, you do not eliminate the root cause of the inefficiency.
Therefore, instead of relying on technology to fix existing operational problems, manufacturers should first take a look at and streamline their workflows before automating them.
3. Data: Build automation on reliable information
Data quality is another key condition for successful automation. Manufacturers are able to collect huge quantities of data, but that data only becomes valuable when it is properly managed, understood and utilized.
“There needs to becredible, real-time data available and understood across the organization,” Diedrich said. The more companies lean on artificial intelligence for operational decisions, the more critical data quality becomes.
“The more we give decision-making to AI, the more important trustworthy data becomes, and the processes need to be rock solid to be combined with this data,” he said.
There are three types of information that manufacturers should especially be aware of:
Master data includes basic product and item information such as item numbers, dimensions, weights and construction materials. These records contain the basic information needed to consistently identify and manage products.
Transactional data captures operational events such as orders, receipts and shipments of inventory. Accurate transaction records allow companies to see the flow of goods and the status of their operations.
Process data records the real performance of manufacturing activities. This means cycle timings, quality signals, execution records, manual corrections, downtime and rework.
“Taken together, these data sets give a more complete picture of the industrial operation.” But their value depends on consistent management and connecting information across departments and systems.
Unreliable data leads to unreliable results, even with sophisticated automation and AI tools, and companies don’t get the improvements they expect.
4. Systems: Connect technology, departments and operators
The fourth pillar concerns the systems that support automated operations. Manufacturers need reliable communication channels between departments and effective connections between equipment and the people responsible for operating it.
Diedrich said IT and OT teams frequently remain separated within their respective organizational silos. When information is not shared effectively, employees on the factory floor may fail to receive the information they need at the right time, contributing to operational failures.
“We have different IT [and] OT, and everybody’s comfortable in their own silo,” he said. “But we don’t share the information, so somebody on the shop floor may not get the right information at the right time, and that causes fails.”
Beyond communication, companies must understand their existing application landscape and define the role each system plays. They also need to examine how information enters their systems, how it is processed and what measurable benefits they expect automation to produce.
These questions help ensure that technology investments support a coherent operational strategy rather than creating another isolated layer of software or equipment.
Scaling automation is not simply about copying a successful pilot.
After manufacturers have established the four pillars for a pilot project, the next step is to determine if the solution can be scaled for broader implementation.
Diedrich warned that scaling is not a matter of copying and pasting an implementation from one factory or country to another. Variations in workforce experience, operating practices and local conditions may mean the same solution doesn’t perform as expected.
Part of the success of a pilot project may be because experienced operators know how to informally work around the shortcomings. That expertise can obscure weaknesses that prove more costly when a solution is scaled up to an entire process or implemented across numerous facilities.
“A pilot can be successful with their work operators because they have the tribal knowledge to… find a way around the shortcomings with a normal workaround,” he said.
The problem comes when manufacturers try to replicate that success outside the original environment. What works in one department or facility may not translate directly to a larger operation with different personnel, procedures or requirements.
“This is where the true readiness of a company is shown, whether the operational foundation is strong or there are still holes that need to be plugged,” Diedrich said.
Therefore, the goal is not only to acquire advanced technology or to prove that a pilot is effective for manufacturers looking for sustainable automation. Success requires early integration of people, processes, data, and systems, clear accountability, identification and resolution of inefficiencies, and readiness of the entire organization to support the solution as it scales.
