Sixty-four percent of all industrial equipment failures are recorded as operator error and this number is a lie that we tell to keep the machines in the room. It is a very tidy number and it makes the managers feel good because they can fix a person with a slide deck but they cannot fix a bad weld or a poorly placed sensor without stopping the whole world.

I sat in a room once with a man named Yusuf and we were looking at a screen that showed a very expensive piece of hardware and the trainer was a man with a very white shirt and a very loud voice. He got to slide fourteen and he told us that we must always point the connector toward the floor or the data would be lost and the unit might fail.

Yusuf raised his hand and he asked why a digital signal cared about the floor and the trainer said that it just worked better that way. He did not talk about the fact that the seal was weak and he did not talk about how water would pool in the rim if it faced the sky. He just gave us a rule to follow and he called it a skill and Yusuf wrote it down and he underlined it twice and he looked like a man who had just been told the earth was flat but he had to believe it to keep his job.

The Translation of Failure

This is how we move the blame from the metal to the human and we do it with a smile and a certificate of completion. If the machine breaks now it is because Yusuf forgot slide fourteen and it is not because the machine was built to fail in a wet room. This is a very common trick in the world of validation and process control and it is a trick that costs a lot of money and a lot of sleep.

I know about these things because I have spent my life looking at the guts of things that are supposed to work but do not. My friend Marie C.M. is a piano tuner and she spends her days dealing with the tension of two hundred and thirty strings and she told me once that a bad piano is just a box of excuses.

“You can try to tune it and you can try to wedge the pins so they do not slip but you are just learning how to live with a corpse.”

– Marie C.M., Piano Tuner

She has a way of looking at a pin block and knowing if the wood is honest and most of the time the wood is tired and the owner just wants her to make it work for one more concert.

We do this in the pharmaceutical plants and we do this in the food factories and we call it training. We spend four hours in a plastic chair and we learn that the software will crash if we click the save button before the green light flashes and we learn that the battery will die if we leave it in the cold.

We do not ask why the software is slow or why the battery is weak and we just learn to count to three and we learn to keep the loggers in our pockets until the last second. We take pride in these workarounds and we think we are experts because we know the secrets of the machine but we are really just high-priced babysitters for a design flaw.

It is like the time I liked a photo of my ex from three years ago and it was a mistake of the thumb and the screen but the app was built to make that mistake easy. I felt the shame of it for a week and I blamed my own hand but the button was right where my soul was weakest and the machine knew what it was doing.

The Case for the Rubber Wheel

There was a man named Alphonse Chapanis in the nineteen forties and he was looking at why pilots kept crashing the B-17 bomber when they were trying to land. The brass said the pilots were tired and they said the pilots were poorly trained and they wanted to give them more classes and more manuals.

Chapanis looked at the stickpit and he saw that the switch for the landing gear and the switch for the flaps were identical and they were right next to each other. A pilot would reach down to slow the plane and he would accidentally pull the gear up and the plane would belly land on the runway.

Chapanis did not give them a slide deck and he did not give them a lecture on being more careful. He put a small rubber wheel on the gear switch and a flat wedge on the flap switch so the pilot could feel the difference without looking. The crashes stopped immediately because he fixed the machine instead of trying to rewire the man.

We are taught to be the rubber wheel and we are taught to be the logic that the software lacks and we are taught to be the seal that the gasket cannot hold. This is a heavy way to live and it leads to a lot of burned out engineers who feel like they are failing when they are actually just fighting a ghost in the hardware.

When you are working in a thermal process and you are trying to prove that a batch of medicine is safe you do not have time to be a workaround. You need data that is real and you need a tool that does not have a slide fourteen. This is the reason I have such a strong opinion about how we build things and why I think the Swiss approach to measurement is the only one that makes sense.

They look at the steam and they look at the pressure and they look at the 134 degrees of a sterilization cycle and they build a housing that can take it. They use 316L stainless steel and they use glass-to-metal seals that are tested with helium and they do this so that you do not have to worry about which way the connector is pointing. The instrument is built to fit the process and the process does not have to be changed to fit the instrument.

I think about Yusuf a lot and I wonder if he ever had a batch fail because he held the logger the wrong way or if he just got tired of the lies and went to work in a different field. There is a specific kind of exhaustion that comes from knowing that the tool in your hand is waiting for you to make a mistake and it is a weight that we should not have to carry.

We should be able to trust that the seal will hold and we should be able to trust that the data will be there when the door opens and the steam clears. When we spend our time learning how to tolerate a flaw we are making that flaw permanent and we are telling the manufacturers that we are willing to pay for their mistakes with our own stress.

Marie C.M. told me once that she stopped working on a certain brand of piano because the pins were so loose that she felt like a fraud for taking the money. She said that a tool that cannot hold its own tension is not a tool and it is just a piece of furniture that makes noise.

Beyond the Hidden Tax

I think we need to have that same fire in our bellies when we look at the instruments in our plants and we need to stop accepting the workarounds as a form of expertise. If the trainer tells you that you have to hold the unit upside down and whistle a tune to make it work then you should stand up and walk out of the room.

You should find a tool that was built by people who know that a wet room is wet and a hot room is hot and a human being has enough to do without being a gap filler for a bad design. We are entering a time where we have more data than ever but we have less trust in the things that collect it.

We see the numbers on the screen and we hope they are right but in the back of our minds we remember slide fourteen and we wonder if we tilted the unit just a little bit too far to the left. That doubt is the hidden cost of bad design and it is a tax on our confidence that we pay every single day.

It is better to have a tool that is simple and rugged and honest than to have a complex system that requires a priest to keep it running. We should look for the stainless steel and we should look for the hermetic seals and we should look for the people who do not have a slide fourteen in their deck.

I still have that blue ink on my thumb from the day with Yusuf and I think about him every time I see a connector that looks a little bit too thin. I hope he found a machine that he can trust and I hope he stopped underlining the rules that do not make sense. We are all just trying to do a good job and we are all just trying to get the data right and we deserve tools that help us do that instead of tools that try to trip us up.

The next time someone tells you that a failure was caused by operator error you should look at the machine and you should look at the switches and you should see if they look like the ones that Chapanis found in the bomber. Most of the time the error started long before the operator ever walked into the room and the only way to fix it is to change the way we build things from the very first day.

We need more rubber wheels and fewer slide decks and we need more companies that care about the reality of the floor more than the profit of the workaround. If we do that we might find that the world is a lot more reliable than the statistics say it is and we might find that we have a lot more time to focus on the work that actually matters.

It is a long road to get there and it starts with a single person asking why the connector has to point down and not stopping until they get an answer that is based on physics and not on a lie. I will probably never go back to that training room and I will probably never see that trainer again but I will always remember the sound of Yusuf’s pen on the paper.

It was the sound of a man accepting a burden that was not his to carry and it is a sound that I hear in every factory and every lab in the world. We can do better than that and we must do better than that if we want to build a future that is not held together by grease and hope and a half-day course on how to keep the water out of a leaky box. It is time to demand more from the metal and less from the people who have to use it every day to keep the rest of us safe and healthy and fed.