Hexagon is frequently at the fore of innovation. Like many in its industry, part of the company’s initiative to succeed is rooted in the constant push to automate tedious processes that take up valuable time, eat up resources and personnel, and delay production and delivery of products. Tools like ESPRIT EDGE software and ProPlanAI not only decrease hours—or even weeks—of delay, but also lead to more accurate data and higher quality products.

In an interview with Ryan Pembroke, product manager of the production software department at Hexagon Manufacturing Intelligence, Quality Digest delves into the ways these AI tools can make production faster, more accurate, and easier for manufacturers. Pembroke provides insights into the Hexagon AI co-pilots and how machine tools and machine learning benefit everyone involved, from those on the shop floor to the programmers working to develop increasingly accessible, user-friendly software.

Quality Digest: Many manufacturers are primarily looking for ways to streamline production processes, cut costs, and retain—or improve, if possible—data accuracy and product longevity. What effect does AI have on these goals?

Ryan Pembroke: AI can help reduce bottlenecks that exist within a production process by making users more productive. As companies have started to automate the production process on the shop floor, the time it takes to get a product to the shop floor has become more of a bottleneck. This means that machines are sitting idle more frequently as they wait for the next job. AI has the ability to reduce the time it takes to move products to production on the shop floor and keep machines running.

QD: What are some examples of innovation in Hexagon AI co-pilots and machine tools as of this year, and what improvements and new additions are expected to be available in 2025?

RP: Programming automation for 2.5-axis operations on milling machines and the subject-matter chat assistant are the first capabilities coming this year. Later in 2025, programming automation will be extended to turning operations on lathes and mill-turn machines, along with added chat interactions with our software, such as ESPRIT EDGE.

A prior example of innovation would be the AI links engine inside of ESPRIT EDGE. Generating machine-aware AI links includes kinematics-specific strategies and recovery options, as well as links tailored to the unique requirements of mill-turn and multichannel machines. This leads to reduced cycle times and increased user confidence when running sophisticated CNC machines with optimized tool travel and automatic resolution of link errors.

QD: What notable outcomes are early adopters, such as RODIN Machining, reporting, and how do these translate to widespread adoption and distribution of AI tools?

RP: The biggest advantage of using machine learning is the minimal effort it takes to get up and running with a trained model that’s specific to the user’s organization. Companies like RODIN Machining are able to provide data and get up and running with ProPlanAI in less than a day. This makes AI more accessible to customers, even those who don’t have prior knowledge of AI. Traditional rules-based systems often require weeks to months and specialized knowledge to get to the same level of automation that ProPlanAI offers.

After the initial training, ProPlanAI continues to learn as the user provides more data. All the user has to do is upload the information, and ProPlanAI takes care of the rest. 

QD: When it comes to efficiency, how much time is saved with tools that automate the setup and production processes? Can you provide some specific data or examples of this with certain tools and company findings?

RP: Programming a single milling operation can take minutes. There are tens to hundreds of variables and parameters that a user must consider. AI automation tools are able to analyze everything in a matter of seconds to produce quality processes. This allows the user to focus their knowledge and valuable time on more complicated tasks than just typing in numbers. This has been shown to take the programming of parts from hours to minutes. 

QD: What are some ways manufacturers from small companies, ones that are fairly straightforward in their operation setups and might not yet incorporate much automation, can prepare for a continued increase of AI tools and AI co-pilots?

RP: Companies looking to leverage AI tools in the future should start by looking at ways that they can digitize and organize their data. This can be as simple as adding material information into the CAM program or standardizing a naming convention for tooling and using a single tool database. By doing this, tools like ProPlanAI can perform more efficiently because there are more data to learn from.

QD: ProPlanAI was reported to reduce machine tool programming time by 75%. That’s incredible! How does it do this?

RP: Manually programming operations takes minutes. ProPlanAI is able to analyze the feature that needs to be machined and return a prediction in seconds. Within the same time it would take a programmer to manually create a handful of operations, the programmer can now create all the operations to program a part. The programmer can now focus more on verification and optimization of the program.

Learn more about Hexagon and its incorporation of AI tools here.