I have been reading a lot these days about Western Philosophy. The most recent book that I have been reading is from one of the great Philosophers of the twentieth century, Karl Popper – “All Life is Problem Solving”. This is a collection of Popper’s writings. One of the great teachings from Popper is the concept of “falsification”, which means that as a scientist one should always try to disprove his theory rather than trying to confirm it. A classic example is the case of black swans (not Nicholas Taleb’s black swan) – if one were to say that all swans are white based on the empirical evidence of his observations of only white swans, he is looking to only confirm his theory. He is not actively trying to disprove his theory. When a black swan is discovered, his theory now breaks down. Loosely put, falsification should lead to attempts to disprove or challenge one’s theory. The more survival of attempts to falsify the theory, the more “reliable” the theory becomes. An extreme example is if I claim that I have the psychic ability to have my coin turn heads on all tosses. I can toss a thousand times and show one thousand heads. However, if I refuse to look at both sides of the coins to see if it is a two-headed coin, I am not looking to reject my claim. I am only looking for evidence that supports my claim.
My post today is not about falsification, but about Karl Popper’s advice on observation. Taiichi Ohno, the father of Toyota Production System, was said to have drawn a chalk circle on the factory floor and asked a supervisor or manager to stand inside the circle and observe an operation on the floor. The task he had was to find as much waste as possible by observing the operation. This has come to be termed as “Ohno’s circle” in the Lean world.
When I came across a section in the book, All Life is Problem Solving, where Popper also talked about observation as part of his three step scientific methodology, I was very interested. His three step model is as follows;
- Attempted solutions
In Popper’s words, the first step, “problem” arises when some kind of disturbance takes place – a disturbance either of innate expectations or of expectations that have been discovered or learnt through trial and error. The second stage in our model consists of “attempted solutions” – that is, attempts to solve the problem. The third stage in our model is the “elimination” of unsuccessful solutions.
Popper had strong words about observation;
The old theory of science taught, and still teaches, that the starting point for science is our sense perception or sensory observation. This sounds at first thoroughly reasonable and persuasive, but it is fundamentally wrong. One can easily show this by stating the thesis: without a problem, no observation. If I asked you: ‘Please, observe!’, then linguistic usage would require you to answer by asking me: ‘Yes, but what? What am I supposed to observe?’ In other words, you ask me to set you a problem that can be solved through your observation; and if I do not give you a problem but only an object, that is already something but it is by no means enough. For instance, if I say to you: ‘Please look at your watch’, you will still not know what I actually want to have observed. But things are different once I set you the most trivial problem. Perhaps you will not be interested in the problem, but at least you will know what you are supposed to find out through your perception or observation.
The standards on the production floor are an important aspect for observation. They tell us what the sequence of operations is, what the takt time is, and what the standard work-in-process should be. Another important aspect to look out for is muri or overburden. If an operator is doing an operation where he is required to lift heavy loads or if he has to reach out to grab something, then it is an opportunity to improve the work. Popper’s advice brings into mind that when we are out on floor and observing, we need to know what we should be looking for.
I will finish off with another great twentieth century philosopher Bertrand Russell’s somber Turkey story, that I have paraphrased;
There was once a turkey that lived on a farm, and the turkey was scientifically oriented. He observed that the farmer gives him food everyday at 7:00 AM. Since he was a smart turkey, he knew that he needed to take a lot of data points. He is observations were made on cloudy days, rainy days, sunny days, weekdays, weekends and all kind of days. Months go by, and by now the turkey feels that he has enough data now and feels confident that tomorrow the farmer is going to feed him at 7:00 AM. However, the next day was Christmas Eve and the turkey was not fed but instead had his throat cut.
Always keep on learning…
In case you missed it, my last post was The Effectiveness of Automation: