A critical lesson from World War II to address today’s manufacturing challenges
What happened during World War 2 to reduce the number of Allied bombers shot down might seem far removed from today’s manufacturing issues like unexpected breakdowns, defective parts, or process deviations in production. But the lessons learned back then are just as relevant and valuable today.
Abraham Wald was a renowned mathematician in the Statistical Research Group (SRG), a classified program that assembled the might of American statisticians to the war effort. In Wallis’s words, the SRG was “the most extraordinary group of statisticians ever organised’.
Abrahams Problem Statement
Planes weren’t returning from missions. To keep planes safe from enemy fighters, you need to add armour. Too much armour makes the plane heavier, slowing it down and using more fuel. On the other hand, too little armour leaves the plane vulnerable. The key was finding just the right amount to keep the plane protected without sacrificing performance.
Data Collection
The military came to the SRG with some data they thought might be useful. When American planes came back from engagements over Europe, they were covered in bullet holes. But the damage wasn’t uniformly distributed across the aircraft. There were more bullet holes in the fuselage, not so many in the engines.
The military reviewed the data and suggested placing armour on the places of the bomber that has the most bullet holes and reducing armour in the areas with the least number of holes. But exactly how much more armour belonged on those parts of the plane? That was the answer they came to Wald for. It wasn’t the answer they got.
Wald’s Insight
Wald suggested that the armour shouldn’t go where the bullet holes are, but where they aren’t—on the engines. His simple yet brilliant question was, “Where are the missing holes?” If the damage were spread evenly across the plane, there should have been holes in the engine casing too. Wald figured it out—the missing bullet holes were on the planes that didn’t make it back. Planes returning with minimal engine damage likely survived because engine hits were often fatal. Meanwhile, planes with heavily damaged fuselages still made it home, proving that those areas could take a hit and keep flying.
By carefully examining the data—and noticing what was missing but right in front of them—Wald’s recommendations were quickly put into action, ultimately saving many lives.
Jumping to Conclusions
In this example, the officers made an unconscious assumption: that the planes that returned were a random sample of all the planes. If that were true, they could draw conclusions about the bullet hole distribution on all planes just by looking at the surviving ones. But once you realize this assumption, it’s clear it’s completely off base—there’s no reason to believe that planes have an equal chance of surviving regardless of where they’re hit.
Biases in Problem-Solving
For Wald, the real issue was something called survivorship bias, where conclusions are drawn only from those that made it through. We see similar biases in the workplace when tackling problems. During the Effective Problem Solver course in November, you’ll get to know different types of biases and become better equipped to address root causes effectively.
If you want to increase your teams ability to get to root cause and create problem solvers, then you are invited to two days of in-depth training with our Effective Problem Solver course, 6th – 7th November 2024, Ansty Hall, Coventry.