Measuring What Matters: Metrology, On-the-Job Training, and Staying Fulfilled in Manufacturing

Open on YouTube ↗
Overview

This episode of Manufacturing Happy Hour brought together three people who approach measurement from very different places. Steve Ilmrud runs North American operations for Hexagon Manufacturing Intelligence, Jeff Van Horn owns and serves as senior project manager at Industrial Technology Solutions Inc., and Randy Smith is a CNC programmer and machinist in aerospace and defense. Host Chris started with a subject the show had never directly addressed, metrology, and used it to reach a broader question: how do measurement, proactive training, and access to modern tools build competent people, and why do these three still find the work satisfying after decades in it?

21 min read

What Metrology Is and Isn't

Steve's first point was that metrology is not meteorology. He admitted he didn't know the difference for many years, and joked that if he could predict the weather he'd probably have another job. His basic definition is that metrology is the science of measuring things. The branch he, Jeff, and Randy work in is industrial metrology, which he described as making sure the things they produce are accurate and reliable.

Jeff framed it in terms of data. He quoted a saying he likes, "In God we trust. Everybody else has to bring good data," and said the message he repeats to his crews is "verify, verify, verify." To show how hard the basics can be to pin down, he described a meeting at U.S. Steel the day before the recording. It was scheduled for 45 minutes and he left seven hours later. About an hour of it went to defining what metrology is and how it differs from what surveyors do.

On-Machine Probing on the Shop Floor

Randy gave the machinist's view. He said the simplest use of spindle probes is making setups easier, and that current probing software is easy enough for someone who has never used a probe. His example was stock with saw-cut lengths that vary quite a bit. A probing cycle can locate the stock and find its centerline, so that when the outside is milled the same amount comes off each side. Later operations can use probes to pick up datums or specific features on the part. In Randy's view this means fixtures and tooling don't have to be extremely accurate: "I can let that float a little," because the probe will pick up the feature every time. He called these the low-hanging fruit, things a shop can start doing now with very little effort and training.

He then described using probes in automated pallet systems running overnight. If a critical feature measures undersize, the machine can probe it, compensate the tool, and recut it. If the feature is oversize, Randy said there is arguably no reason to keep going. The system should reject the pallet, set a flag, notify the operator that something needs attention, and load the next job so the machine doesn't sit idle. He said most automation and pallet systems have a setting that allows this. His conclusion was that real-time inspection and feedback is critical regardless of what parts you make, and saves a lot of time and aggravation.

Measuring Change: Steel, Heat, and the Las Vegas Sphere

Chris asked Jeff how people misuse metrology compared with how it should be used. Jeff described good practice more than misuse. Metrology, in his description, is the study of numbers for decision-making: looking at small deltas, looking for reliability, and building confidence that machines are installed correctly and steel is fabricated and erected in the right place. His firm tracks fabrication jobs to learn what heat, cold, and weld procedures do to the steel. He described metrology as measuring "the changes or deltas in a process."

When Chris asked what happens if those factors aren't measured, Jeff used the Las Vegas Sphere as the example. His team fabricated the trellis, the outer structure that holds the exterior screen, in a shop to a tight tolerance. The pieces were then trucked through the desert, exposed to vibration and large temperature swings. Jeff said he could be "burning up in 90 degrees and looking at snow caps." At the site, a steel erector wanted to know who had done the metrology back at the shop, because the steel was an inch off. Jeff's team then started monitoring what transport, heat, and cold did to the steel. He said steel has stress-relieving properties, and how you treat it and handle it "is everything that you need to know."

One practical result: on large job sites, pieces are often flown into place early in the morning before sunrise. Jeff attributed that practice to the measurements they took, and said you'd be "crazy as heck" to fly something in during the afternoon sun, because in his view expansion is much worse than contraction. Chris summarized the point as using measurement to anticipate material changes and plan for them. Jeff agreed and added that the goal is to do everything possible to help the field erectors.

Metrology Is Everywhere: Airplanes and Offline-to-Inline Inspection

Steve started his career in jet engines. He argued that almost everything made in a factory, from the coffee pot to the car to the cell phone, is measured somewhere during production, even though people rarely think about it. His main example was aircraft. An airplane's fuselage might be made in France, its engines in the U.S., and its tail in Germany. Each factory uses metrology to control the dimensions and geometry of its own parts, and that data is also shared with the final assembler so everything fits when the pieces meet. "You can't grow or shrink these parts," he said. Being an inch off at final assembly is a serious problem.

Steve also described how inspection itself has changed. Traditionally a finished part went to a CMM lab, where "the guy in the white lab coat" declared it good or bad. He said the industry has moved from that offline approach to near-line and then inline inspection, where parts are checked while they are being made. That lets manufacturers monitor the process and avoid producing scrap or rework in the first place, which he considered a much more efficient way to run an operation.

NASCAR: Optimizing Parts That Are Supposed to Be Identical

Asked how metrology helps manufacturers make the best possible products, Steve talked about Hexagon's work with NASCAR teams. Teams used to build their own chassis and body panels within the rules, so a car's quality varied a lot from team to team. To create parity, NASCAR now has one company supply those components to every team.

In theory every team now gets identical parts. Steve said some teams measure every part they receive anyway, because the parts aren't perfect and some are "big and small and twisted and curved." They use that data to decide how to build the best possible car each week from what they have. Steve said the teams doing this are winning a lot of races.

Speed as a Result: Rapid Alignment at U.S. Steel

Chris brought up something a professor once told Jeff: people only care about results. For Jeff's clients, the main result is how quickly a job gets done. Jeff went back to the U.S. Steel visit, where much of the discussion was about getting results faster. He estimated that about 80% of the people there had never seen a Leica tracker, and they were impressed when they saw the AT500 and ATS600.

The client asked Jeff's team to check a roll's alignment on a cold mill line that was about to be commissioned and to see how quickly they could establish control. Jeff said the client first noticed the startup time. According to Jeff, competitors' equipment takes 45 minutes to an hour to warm up. His team set control with the AT500, which Jeff said reaches 500 feet in each direction, and used the ATS600 to scan the roll. They had all the data in about 35 minutes. Jeff said the client told them the competitor usually takes about four hours. He attributed this to process, and added that there's "no substitute for knowledge either."

Reliability, the Vital Few, and Driving Rework Toward Zero

On reliability, Jeff said the key is understanding the process and focusing on "the vital few items versus the trivial many." He described his team as people with backgrounds in mathematics, statistics, and engineering. In his experience, the problems hurting a client the most, usually rework, are fairly easy to identify at the start. Current equipment also lets them do work such as crane evaluations without shutting operations down. He said he's tired of managers telling him a shutdown costs a million dollars an hour: "I didn't put them in that position. We're trying to help them get out of it." Good statistics and data going into a project, he said, is what supports reliability.

Asked how he takes rework to zero, Jeff said the answer is training everyone, including millwrights, steel erectors, fabricators, and fitters, to use metrology and to know when to call it in. You can't wait until parts are finished to inspect them, especially in fabrication shops. Anything not caught during the process ends up as rework. He also noted that some tape measures in the field are out of tolerance. People laugh when he says rework can go to zero, he said, but his company has a 40-year track record that he says proves it. He called it a mindset and a paradigm shift.

Building Inspection Into the Program, Not Adding It Afterward

Randy connected this to his own tools, specifically Hexagon's ESPRIT CAM software. It includes an add-on for programming probing cycles inside the system. Randy said probing routines used to be hand-coded because they weren't well supported. In his opinion Hexagon has done "an outstanding job" of bringing this into the software, so probing moves can be simulated, checked for collisions and clearance, and built into the program from the start.

His larger point was about mindset. Probing should go into the program at the beginning, not as an afterthought. When a programmer sees tight tolerances or key features, such as a feature that locates the part for the next operation or a tight-tolerance hole, getting it wrong means either rework, which he called painful and costly, or scrapping the part and losing all the time and material already spent on it. Randy argued that manufacturers need to allow time in the process for these checks and think ahead about what could go wrong: delivery delays, lost goodwill with customers, lost cash. Until shops accept that inspection has value at the front end as well as the back end, he said, they'll keep reacting to problems instead of preventing them.

Modern Tools as a People Strategy

Steve had said before the episode that attracting and retaining people depends on flexibility, training, and access to modern tools. He focused on the tools. Because he runs Hexagon's factories in the Americas, he said, whenever Hexagon acquires a company with new hardware or software, he's on the phone asking how to get it into his factories.

Hexagon has a machine shop that makes its high-value machined parts. When Steve arrived a couple of years earlier, he found that some CNC programs were 20 years old and hadn't been reviewed in a long time. Using ESPRIT, the shop built a digital twin of its actual process and simulated different cutter paths and machining approaches offline, without disrupting production. In his words, they made their mistakes "in a virtual world where you don't crash machines, you don't crash parts, you don't hurt anybody." He said this greatly improved productivity by modernizing the programs, and also gave employees a technology they can use at Hexagon or wherever they work later.

He also mentioned NCSIMUL, a Hexagon simulation tool with a workstation trainer that can simulate different controllers. Employees can virtually run CNC machines or CMMs for safe offline training.

Steve believes the products themselves help keep people engaged. He said Hexagon's CMMs typically measure to about a micron. For scale, he said a sheet of paper is about 100 microns thick, so some of their CMMs measure within one micron across their full measuring volume, about 1% of a sheet of paper. He added that some ultra-high-accuracy machines reach one-third of a micron. It isn't a jet engine "spewing fire out the back end," he said, but he thinks the engineering is impressive, and that exposure to this kind of technology is part of what draws people in and keeps them.

Learning to Listen: Jeff's Madison Square Garden Story

Chris asked Jeff how he helps his teams understand what tools are available, referring to a Madison Square Garden story. Jeff first wasn't sure which story Chris meant and briefly mentioned a different project before returning to the question. His answer was an admission: one of his biggest personal challenges is listening. The Madison Square Garden project, he said, became very successful and accurate largely because of the ATS600, and he had been against using it. A couple of demonstrations changed his mind.

He took a broader lesson from it: "we're all students for the rest of our lives." He said that in his 60s he's probably one of the hardest-working students in his organization. On that project, he said, his team learned from steel erectors, engineers, and site surveyors, and his advice was to never close your ears to ideas.

Chris mentioned another of Jeff's lines, that you can learn and train on the job and still beat the old production numbers. Jeff agreed. People thought "an old man" couldn't beat them, he said, and his team didn't just win, they "killed them." What matters, in his view, is working smart rather than hard. He added that metrology is a control. If you stop using it, "you just go back to ground zero again," which is why training has to be continuous. Otherwise rework comes back and costs go up.

Randy's Worst-Case Thinking as a Training Method

Randy believes you should prepare for problems that haven't happened yet. He said one of his "superpowers" is imagining the worst case: someone puts the wrong tool in the spindle, loads the part slightly shifted, or transposes a number. He described these as honest mistakes. Someone reads one thing, grabs another, and nothing in the process catches it.

When he trains people, he uses these situations as scenarios. Suppose a dimension wasn't hit, or the roughing tool wore out too soon, or the finishing tool didn't perform. A check could go in at that point to measure and update, or to set a flag that says "come look at me." He points out when a missed dimension would cascade through later operations, and he doesn't consider these scenarios unrealistic. Some come from things he has seen, others from things he can imagine happening.

His approach is to lead people toward answers instead of handing them over, even though as subject-matter experts he and his colleagues often already know the answer. He wants people to think critically, which he believes builds confidence. He finds it rewarding when someone takes a small piece of what he taught, recognizes they've done something similar before, and applies it somewhere new. He also suggested that a simple question on the floor can become an impromptu 10- or 15-minute training session.

Magic Wands, Pain Points, and Eating the Elephant

In a follow-up, Chris asked how Randy gets people to ask the right questions and be willing to learn on the job. Randy said he tries to encourage curiosity. He asks people what they think the shop can't do, or what they would do with a magic wand, and he asks about pain points: what would make their day much better if it were solved. Not every idea is possible, he said, but digging in often shows the solution is easier than expected, sometimes because inspection probes or software can already handle it and people didn't know.

He said large projects can look overwhelming, including to him, so he uses the "how do you eat an elephant" approach and breaks them into small pieces. When no one is sure something can be done, his answer is to try it: take measurements, inspect, record the data. He said he and others have often found the job wasn't as hard as they thought, and that it mostly took confidence and tenacity. He also mentioned side benefits. Researching a system variable, a machine function, or a software feature adds another tool that can be applied across the shop.

The Paradigm Shift Depends on the Organization

Chris returned to Jeff's paradigm shift and asked whether more people are now willing to change. Jeff said it depends on the company's "aptitude." He is blunt about it: if someone isn't willing to change how they think and work, he's "really not interested in working with you." His method is to get buy-in from everyone, then produce quick wins to show it works. His example was a Kaizen event that involves all the different groups. Once everyone buys in and sees a success, he said, the mindset changes remarkably fast.

Steve's Three Ps: People, Parts, Process

Chris asked Steve to summarize manufacturing operations in terms of people, parts, and process. Steve said the framework came from a hard point in his career, about 10 years earlier. His company launched a product before it was technically ready and before the supply chain was set up. He called it one of the most challenging periods of his career, when everything that could go wrong did. Trying to get a handle on it, he reduced the problem to three Ps, and he said most challenges since then can be approached the same way.

People come first and matter most. That means the right people with the right attitude, education, and willingness to learn, in an environment where they enjoy the work, understand the goals, and communicate well. Parts means supply chain, which Steve said is difficult right now whether sourcing is local or foreign. He said there's no magic pill. Suppliers come and go and have their own problems, so companies need flexibility and options, and his specific advice was to have a good first source and a good second source. Process means turning people and parts into something dependable, repeatable, and consistent, which is where metrology comes in. You have to measure what you're doing and use good metrics to see whether you're on track or trending out of control. He said this includes work instructions, process control, fixtures, and jigs. Together, he said, the three Ps work as a guidepost for whatever a manufacturing environment throws at you.

What Keeps Them Going

Chris noted that measurement isn't usually thought of as exciting and asked each guest what has kept them satisfied.

Randy, a third-generation machinist, pointed to how varied manufacturing is: micro-machining, aerospace, composites, medical, and all kinds of unusual fabrication. He cited the old joke that if you put ten programmers in a room with one part, you'll get ten different ways to make it, many of them probably correct. He said he doesn't have "an artistic bone" in his body, but turning raw stock into a finished part that many people have worked on is his creative outlet, and he finds many of those parts beautiful. If someone loses interest in one sector, he said, there is always another area to learn that can "recapture that imagination." Programming has always been fun for him, and the constant changes in software and methods keep him on his toes. He said he especially values having some leeway: being given a problem with some guardrails and trusted to solve it. He wants others to be encouraged the same way, because he believes solving problems makes people feel valuable and confident enough to take on harder projects.

Jeff, whose sons are mechanical engineers and whose wife is the company president, said simply that he likes to see people and companies succeed, because "success begets success." He said what has kept his company successful and "never bored" is that people it trained have moved into larger roles at other companies and keep calling. He credited a mentor who told him long ago: "We invested a lot of time in you. You make sure that you invest a lot of time in the people that come behind you."

Closing: Uncertainty as an Opportunity

Steve gave the final reflection. He noted that the recording happened on a day when tariffs were the top news, and that there is a lot of fear and uncertainty in U.S. manufacturing. He believes it should be seen as an opportunity. Many companies are trying to do more in the U.S., and he said it has been difficult, because the country has arguably lost some of the edge it once had as the leading manufacturer in the world.

His advice was to buckle down and invest in technology, tooling, and people, including cross-training and mentoring, to build the industry's future workforce. He believes that with those investments U.S. manufacturing can be at the top of its game again. He urged listeners not to be afraid and "close up shop and go home," but to invest, learn, train, and grow.