Patrick Mosher recalls his supervisor Pete at Corning Glass Works replacing a $100,000 robotic arm with a $1.37 cam to fix a quality problem on Corelle plates. The lesson: deep analysis often reveals simple solutions that over-engineering misses.
Chapters
Last century, I was a young chemical engineering co-op student. I got to do my first professional gig as a chemical engineer after just one year of freshman engineering curriculum under my belt. Freshman calculus, freshman chemistry, freshman English. Hardly qualified to do any real engineering work. Nonetheless, I traveled from my hometown of Cincinnati to Corning, New York, headquarters of Corning Glass Works. I was reminded over and over that I was Corning's first co-op student, a talent experiment for this chemical engineering department, which serviced all of Corning's locations and diverse products. Like, like bulbs, micro-sized medical products, lab benchmarks, Corel, Pyrex, resistors and capacitors for computers, Stubben fine crystal, optical wave fiber, and so on and so on. I quickly got enthralled with the unique qualities of glass. Did you know that glass is an amorphous solid? It exists in neither solid nor liquid state. It's always somewhere in between. Well, Pete was my supervisor on that first project. That summer, I learned as much about life as I did about glass and chemical engineering from Pete. He had a dream to live on a property with minimal inputs from the world. He showed me plans for the simple cabin and how he streamlined all the inputs down to just salt and kerosene. Clearly, Pete loved to streamline solutions to their simplest form. His solutions required deep understanding of the problem, creative calculations, and a sharp mind that cuts through the confusion and ambiguity to see things from a particular point of view, always striving for that simplicity. Pete taught me about simple, elegant solutions. In the orientation week, he introduced me to simple, elegant solutions on a plant tour where Corning manufactured Corelle. Corelle is a line of dishes that is nearly unbreakable. So get ready for a little geekdom here. What makes Corelle unbreakable? Well, the plates are laminated with 3 layers of thermally bound glass. The 3-layer combination of different glass treatments applied at different temperatures provides a surface tension that makes it nearly impervious to impact. You can bang it around quite vigorously and it just doesn't break. But when it does go, wow, it shatters into millions of pieces. Well, anyway, as Pete was taking me through the plate-making plant, we stopped between conveyor belts. The first conveyor belt applied one glass layer to the plates. As the plates went to the next operation, they needed to cool slightly and move into the next stage of heat treatment at a different speed than the first conveyor. This meant that the single conveyor belt needed to be divided into two conveyor belts. So Pete explained that the solution provided by the first engineering team was a robotic arm with a suction cup that picks up each plate from the conveyor belt number 1, turns left, and drops it onto conveyor belt number 2. You can see it here. So it picks up the plate, moves it over to conveyor belt 2, and then swings around in time to pick up the next plate and move it over to conveyor belt number 1. 3 and swings around, and so it just keeps swinging back and forth. The robotic arm swings back and forth, dividing the plates from conveyor belt number 1 to conveyor belts number 2 and 3. The problem was that the plates developed a faint ring where the suction cup was interrupting the cooling process. A team of engineers was deployed to investigate and solve. They analyzed the temperature of the glass coming out of conveyor belt number 1, They looked at the minimum suction required to pick up and move the plates. They looked at minimizing the time of suction on each plate. They looked at the, the lengths of the conveyor belts and the cooling time inside and outside the ovens. And then they asked Pete to analyze the problem. He glanced at the operations, the robotic arm. He reviewed the calculations and provided his solution. Instead of a $100,000 robotic arm with lots of moving parts and software. Why not install a small cam that pivoted right and left? When a plate hits it, it moves. So when a plate comes down, it hits it, drops the plate here, but the cam swings, the next plate comes down, and it moves the plate to the other conveyor belt. One to the other. This was a $1.37 solution. Every time I tell the story, I get a little teary-eyed as I think about the beautiful, simple elegance of this solution. And this is one of the most important lessons I learned as a chemical engineer. Resist the urge to over-engineer your solution. Look for simple, elegant solutions. Whether you're solving an engineering problem or seeking steering committee alignment, look for the simple, elegant solution. Believe it's there. Always. Simple, elegant solutions require deep analysis. They require creative thinking. And you'll know when you have it when you present it and everyone is silent for a moment, gobsmacked with the simplicity. When faced with my most complex, ambiguous challenges, I envision Pete tipping back in his chair across from me in our adjoining cubicles He gets that smile of wisdom on his face, a glint in his eyes, and the simple, elegant solution emerges. Rest in peace, Pete. Learn deeply and continuously.
