A Way to Safer Airplane Designs. Blog 5

 

My Boeing Days

 Blog 5 of 10

Photo: 747, JAL, Near Mt. Fuji.

 

Starting at Boeing with a New Assignment

Soon, all newcomers were assigned tasks. My job was to conduct the Failure Modes and Effects Analysis (FMEA) for the autopilot system on the 747-300. The 747-flight control system is triple redundant, meaning that if one computer fails, another takes over, and if the second fails, the third will take over. Conducting the FMEA involved a lot of repetitive calculations.

Seeking Efficiency: The Quest for Better Computing

By nature, I am lazy when it comes to repetitive tasks. I always try to find ways to use automation or computers to perform repetitive tasks quickly, allowing me to focus on tasks that require thinking. There were two HP9830 computers in the group. After repeated requests, one was assigned to me for four hours a day. The computer was so slow that one calculation used to take about an hour. It had a small cash register-type printer, and the result was printed on that paper. There was a one-line display that was hardly visible and readable. Soon, I got frustrated with it too. There were no other computers to be seen anywhere in the building where I was sitting.

Discovering the Computer Room

One day, my friend Ed and I were coming back from lunch. Someone had mentioned a computer site in another building, so we decided to check it out. When we went there, we saw a big computer room kept in a special temperature-controlled environment. The whole thing looked like a set from a Star Wars movie. We talked with the site coordinator to ask for permission to work there. He asked us to get a form filled out by our manager. We immediately completed all the formalities and started working there. Now, my program was running in 30 minutes instead of 2 to 3 days!

The Critical Autopilot Project

I was working on a very critical part of the autopilot program. In those days, the 747 autopilot landing was not perfected. On simulators, the airplane sometimes deviated from the runway and landed on grass. It was very important to fix the design as soon as possible and get certification from the FAA for the new autopilot systems. According to FAA rules, the failure rate should be below 10^-9, i.e., one catastrophic failure in a billion landings. The old design did not meet the criteria.

Discovering a Mistake: The Importance of Integrity

I started coming up with the failure rates under various conditions. These results were presented in multiple meetings, and long discussions were held on design improvements. During one such meeting, Jack, the second-level manager, made a side remark that for a specific condition, the failure rates seemed to have improved considerably. I heard that, and it made me feel uneasy. If it is too good to be true, is it true?

Every other week, Jim repeated his message loud and clear: "When you design an airplane, remember your spouse, your children, your parents, your family, your friends, and many others who are going to fly on that airplane that you are designing. Their safety is in your hands." The message was ingrained in us. I had to go back and check my numbers.

Confronting the Error: A Test of Character

After the meeting, I rushed back to my desk. It was lunchtime, but first things first—I had to check the numbers. I did, and I found a mistake. I did not know where it was happening, but I was sure that the numbers presented in the meeting were wrong. Without thinking anymore, I rushed to my lead engineer's desk.

Pete was having his lunch. "Pete," I said. "Yes?" he replied. "I have made a mistake in my numbers," I told him. Pete was neither alarmed nor curious to find the mistake. Instead, he said, "So what do you want me to do?" "Would you please inform Jack that the numbers are wrong?" I requested. I was following the chain of command and not trying to bypass anyone. Pete laughed. He said, "Do you know how Jack shouts and screams at people?" I interrupted him and said, "But Pete, the numbers are wrong. I can't keep others in the dark. I have to inform Jack." "Ok," murmured Pete. "You do what you want. I'm not coming."

A Culture of Openness and Forgiveness

Jim's words had made a strong impression on me. I could not push my mistakes under the rug. I didn't think twice. I rushed to Jack's office. Jack was eating a sandwich with his left hand and working with his right hand. "May I come in?" I asked. Jack nodded. I went in. "The numbers that you saw today for the failure rates are wrong. I have made a mistake," I said. Jack's response was unexpected. "So, when can you give me the new numbers?" I told him that in two hours he would have the numbers. Later, I found that in my program I had swapped the labels for graphs.

I was very relieved by the way Jack responded. His response ensured that if I made mistakes in the future and informed the management, I would be forgiven. This is the best way to ensure that the safety of the airplanes is prioritized. To err is human; to forgive ensures openness. Openness, in turn, leads to better design and airplane safety.

Years Later: Another Test of Management's Support

About seven years later, one of our team members, Andy, was assigned a task that required an IBM PC. In those days, PCs were not easily available. Our group did not have the budget to purchase them; however, a neighboring group did. When Andy wanted to use their PC, he had to go through permissions from the bosses. Ed Schroeder was our boss. It was agreed that Andy would use their PC after hours. Thus, Andy started to work from 5 pm to 9 pm in the other group.

One day, Andy came to me. He was very scared. He told me that by accident he had pressed the Ctrl-Alt-Del keys, and the entire disk was wiped out. He was waiting to meet Ed to tell him what had happened. After telling Ed about the mishap, instead of saying anything, Ed started laughing very hard. Andy was all confused. Ed told Andy, "If they don't have a backup disk, then it's their problem. I'll talk with them."

Prioritizing Safety Through Supportive Management

This is the type of management we had—very supportive, forgiving, encouraging openness, and taking care of issues. The idea of an airplane safety issue not coming to light at an early stage was unthinkable in this environment. Safety was ingrained in our minds by Jim through his biweekly speech.