Automating the manufacture and assembly of aerospace and defense components is no simple task. Parts are often complex, with a high/low mix of components that range from robust to micro in size, and this diverse range doesn’t necessarily lend itself to automation.
Myriad skillsets are needed to automate these tasks—and to lead the teams that bring them to fruition. They can be of the hard engineering variety: coding, robotics, artificial intelligence/machine learning, metrology. Often, they require certain “soft” skills, such as team building and leadership, and personality traits, such as tenacity, curiosity and creative thinking.
Two leading engineering executives, Nicole Williams at The Boeing Co. and Marie-Christine Caron at GE Aviation, oversee automation efforts at their respective companies—and possess these skills in abundance.
The women, whom Smart Manufacturing magazine this year named as two of the “20 women making their mark in robotics & automation,” spoke recently in a related webinar (https://bit.ly/Robotics2paths)—detailing how they got their start, the roles they play, the issues and challenges they face and what the future holds for the next generation of female engineers.
These women rose to their corporate positions through different life, academic and career choices.
For Williams, a life of math and science seemed a foregone conclusion. Her homelife was like an engineering playground: Her father and uncle were electrical engineers, and her aunt was a mechanical engineer. Around the house were electronics just waiting to be disassembled to see how they worked. Flat surfaces were home for coding magazines that allowed Williams to practice her budding coding skills, and prototypes of ornaments and musical cards that her father brought home from his job as a development engineer at Hallmark. (A childhood favorite was an ornament that displayed a 3D snow-covered holiday scene with a small train chugging through a tunnel.)
“Ever since I was very young, I’ve been interested in taking things apart, computers, programming, and problem solving,” Williams said. “I’ve always liked math. It is consistent and dependable. It is not arbitrary or whimsical or changing.”
Her mechanical engineer aunt taught her that mechanical engineers can work on anything from designing commercial products to medical implants to robotics and nuclear facilities.
“I really liked the variety of projects that I could support. I loved learning new things, using my skills to solve different types of problems, both inside and outside work,” she said.
Her interest in robotics began at the University of Missouri–Rolla (now Missouri S&T) where she worked with a SCARA (Selective Compliance Articulated Robot Arm) configuration robot to sort rectangular wood blocks from circular blocks and pick and place them as needed.
This academic project helped hone her programming skills—skills she said helped her get hired at The Boeing Co. in 1999.
Initially, Williams supported a development robot that was employed to handle a variety of parts. The gantry-style robot featured a large end effector that held spools of carbon fiber, epoxy material. The machine used B+ programming, and many of her early assignments involved creating drawings for this unit.
While working on these assignments, she absorbed other skills.
“I learned about the patent application process and how difficult it can be and how long the process can take,” she said. “I soon became involved in programming with a [robotic simulation software] product called IGRIP, creating robot programming tools and simulation tools for robot programs.”
She became adept at using a Boeing invention called RAC, or robot assembly self-control, that relied on a goal-based autonomy that over time would allow the company to improve its overall equipment effectiveness (OEE), first-time quality, and usability for both maintenance and mechanics.
“In order to make automation as flexible as possible in an aerospace environment, we rely heavily on this goal-based control,” she said. “Rather than writing an explicit script of actions for the robot to complete or execute like in traditional robot programming, we give it a set of goals and rules on how to complete these goals.”
In one project, a workcell featured localization using machine vision, the RAC goal-based supervisory controls and robot accuracy through kinematic calibrations.
“We have specific kinematic calibration that we use for our robots. And then a complete use of OOP without any touchup,” she said, referring to object-oriented programming. “That’s something I think is pretty rare, to be able to take a program right off the NC programmer’s computer and go run it out in the shop without having to do a lot of dry running or adjustment.”
It became Williams’ role to assemble all the models into IGRIP, which is programmed using a graphical simulation language and command line interpreter. ”Each individual program and robotic system had to be simulated to identify potential issues prior to production,” she said.
One issue that presented itself was drilling holes into C-17 pylons.
“At the time, I was the only one with the software and capability to bring all of the pieces together into the work cell, including the tooling, the part, the robot, and the end effector. We identified portions of the tooling that were blocking areas that the robot needed to access to drill [the pylons]. In the end, we ended up having to cut away part of the tooling to allow access for the robot end effector.”
This project taught her an important lesson, and that was to bring all of the stakeholders together earlier in the process and to simulate multiple conditions and scenarios prior to construction. In fact, many of the simulations Williams created have been used in meetings with machine tool suppliers, helping all stakeholders visualize concerns.
“This often affected the machine design and modifications,” she said. “Soon, I started to travel to train NC programmers on how to use the tools we had developed, and how to use the simulations in a production aircraft environment.”
The idea of using automation and robotics as a tool that enhanced workforce flexibility and worker ease-of-use is a concept that would follow Williams throughout her career. It is a mindset that Caron also believes in, and which she has employed as she built her career.
Like Williams, Caron found science and math to be driving forces. But it was her prowess on the tennis court that quite literally served her well when it came time to study for a career in engineering.
“I reached out to many of the universities in the U.S. that had a tennis program, and told them, ‘Hey, I live in Quebec. I play tennis, and I want to study engineering,’” she said.
Before earning her tennis scholarship to the University of Massachusetts at Amherst, she sent letters to 50 schools across the U.S.
With scholarship in hand, she jumped into a new life, in a new country, with a new culture, and learned to balance academics and athletics.
Those challenges “got me to believe in myself and understand the fact that, even if you don’t know what to expect, you can still have fun and succeed,” Caron said. “It grew my resilience skills and adaptation skills, and opened my mind to others and really get to know and really understand how I can be successful and how I fit in a team.”
After graduation, Caron moved back to Canada and took a job at IBM, working on microelectronics. “When you speak microelectronics, you speak automation because everything is so small and [assembly] so rapid that everything is automated. That was my first introduction to true automation and really got me to love the link between technology, logistics, and quality—to make the best product in the most efficient way.”
Caron climbed the IBM ladder, and eventually was promoted to managing engineering teams. “I went from being a technical person to leader, but I was always very linked to technology, trying to make the team be very successful.”
After 13 years, she joined GE Aviation’s facility in Bromont, Quebec. This career jump would take her from a “super precise, mini microelectronics team to super robust, but, at the same time, very complex aviation world.”
The Bromont location makes engine components for Boeing and Airbus aircraft, and is home to the company’s Global Robotics, Automation and Instrumentation R&D Center that develops advanced robotic processes and software applications.
Not too long after she joined GE, there was an opening in the Global Research Center, which is known as the GRC, and she “jumped on that opportunity.”
The job at the GRC allowed her to further explore what could be done with automation and robotics from an engineering perspective.
A nine-month assignment in the Czech Republic helped her hone additional skills. There, Caron was not just an automation or robotic expert, she was a project manager.
“I was lucky to be part of the engine development program in Prague,” she said. “I learned to really be open minded on culture and understand their manufacturing processes. I had a lot of [my own] answers, but they weren’t fitting necessarily with the way they were seeing things. I learned I needed to understand the constraints, understand the environment and propose the right evolution for each site.”
While automation can bring productive gains, Caron learned that potential users might not always accept the help so readily.
The mindset of some is that if they need more capacity, they will just shoot an extra body to it. The core of this concern is jobs.
“People were like, ‘it’s going to take our jobs.’ No, it won’t; it will secure your job. Because you will do more parts, more accurately and we will still use your brain” for other jobs, she said.
One of her more challenging assignments looked at a manual project that called for seals to be inserted into small assemblies.
In a two-inch by two-inch (50.8×50.8-mm) piece, for example, the assembly worker might have to insert 40 seals with tweezers. “It was very tedious work for the operators to do and was taking them forever. So how do you automate this?”
For a human, it is an easy, albeit laborious, task: Simply pick and place the seals.
Humans can determine if the seal is seated correctly, or at least within spec, and even if the seal is present.
But there are “a lot of things that your brain does that are quite difficult to put into a system,” said Caron.
To automate this assembly process, machine vision was installed and AI/machine learning integrated into a workcell.
“Automation robotics is not only robots, it’s everything around it,” she said. “How do you see it [seal insertion], how do you localize it, how do you know where you’re at in 3D space, and how do you reliably perform a task every time?”
To help accomplish this, five different cameras were installed within the work cell, each of them having a specific application.
By analyzing the images, AI can determine a good from bad seal. “If you were to say ‘good’ or ‘no good’ without the AI, the system would say, ‘It’s different, so it’s no good.’ But AI brings you that extra capacity to say, ‘I don’t need it to be black and white; it can be gray’,” Caron said.
From Williams’ days in college where she used imaging technology to sort wood blocks, to her graduate research that used image data to train a series of neural networks, she has extensively utilized imaging and AI techniques to solve automation challenges.
One project, which would earn Williams and her team a Boeing Silver Phantom award, analyzed drill patterns on aerospace components for commercial aircraft.
The control scheme used Boeing’s RAC concept and incorporated part-scanning, as well as data regarding tooling and part features, to determine accurate final drilling positions.
“In aircraft-building, edge distance is an important datum,” she said. “If you drill a hole too close to the edge of a rib or spar, you risk premature failure of that part.”
While Williams’ range of projects is diverse, she said that one of the most effective projects that she worked on was also one of the simplest: tracking worker’s tools.
Previously, to ensure that tools are accounted for at the end of the day, a technician would take a piece of foam and draw the shape of a tool on the material. They would then take a Dremel and hand carve out the tool’s shape.
The automated solution used a mobile cart in which the workers lay out their tools and a machine vision system would capture an image of the toolbox, translate it into a binary image, then translate it in an Excel file and finally the file would be sent to a laser cutter to cut the foam.
“We had to figure out a lot of vision system challenges. Some tools are particularly shiny. Some workers have made ‘custom adjustments’ to their tools, such as adding taping,” she said. “That’s why we couldn’t use an off-the-shelf system.
“It was a pretty simple project, that went together very quickly, but it saved our operators who were able to go from maybe finishing eight or 10 [tool] drawers a day to being able to complete multiple full toolboxes in a day,” Williams said. “It was a really dramatic throughput increase for the team.”
In the long run, the goal is to make better, more cost-effective parts. And the talents that Williams and Caron bring to their positions are making that happen in GE and Boeing facilities around the world.
And, while the women are still somewhat of an anomaly in manufacturing engineering these two women have proven that there are no “men only” jobs.
Williams and Caron have worked with and for and have led male colleagues—and they have risen to the top of their professions.
Each did so through hard work, and a willingness to learn and try new things.
For the next generation of female engineers and manufacturing company executives, the pair urges those considering these professions to follow the things that interest and inspire them.
“Think about what connects you, what drives you, whether it be a particular technology, a particular skill set, or particular group of folks you know,” Williams said. “And then continue your education. Take one class, take a couple of classes. Just dip your toe in the water and see what interests you, and what really drives you.”
Caron agreed: “Go with your heart. If you like math, don’t worry. That profession is evolving and there’s so many branches you can take. Don’t be afraid to try it. Trust yourself, be yourself, and do what you love.
“You grow from every single step of your work experience and your life experience,” Caron added. “Make the most out of them, and take the small nuggets and put them together into what you want to become and how you want to be. And really, for me, this automation journey is really an example of that.”
Connect With Us