This particular questions asks why it is so hard to cooperate across functions to solve problems by using the scientific method. Not the exact wording but close enough for short discussion. Problem solving via any method (scientific method or otherwise) is not all that simple when you stop and problem solve the process of problem solving. At least I have never found that to be the case. Solving actual production or engineering problems is far simpler and more straightforward. An actual answer on this topic could take dozens of pages and examples but I will at least outline some short points and attempt to provoke some thinking.
For starters it is now highly fashionable to throw around the “scientific method” phrase when discussing Lean or TPS. The first TPS handbook in 1973 made this comparison clear and its limits if you read closely between the lines (unfortunately it is in Japanese). Taiichi Ohno remarked multiple times in his books that often all he had was his power of keen observation and the scientific way of thinking at his disposal. Multiple Toyota based TPS presentations from over two decades ago by a mentor named Iaso Kato (retired director of training at Toyota) compared problem solving and kaizen to the scientific method and how they are similar and yet different. You see it more and more today in various locations. So it is not a recent or totally inaccurate representation. However it also has its limits.
For starters keep in mind that even the vaunted scientific method is just a method. I get a chuckle when people take pot shots at method based approaches involving tools or rules etc. for lean but then are comfortable advocating the scientific method as the next great saving technique for lean implementation. A piece of the puzzle perhaps but so too is standardized work for that matter or flow of product or pulls systems, etc. etc. Methods alone will never be enough to accomplish all of the Toyota Production System…not even the scientific method will be sufficient as it has no directional compass or specific advice on how to do things i.e. the Toyota Way, etc. Try to reconcile respect for people and team work in the Toyota sense versus what goes on in most scientific endeavors for example.
Also I think it is an insult to the real scientific method to imply that is what problem solving or even TPS is in all honesty. The real scientific method is far more difficult and time consuming if you consider it in its highest form. Let’s use a generic instance of the scientific method below.
- Define a question
- Gather information and resources (observe)
- Form an explanatory hypothesis
- Test the hypothesis by performing an experiment and collecting data in a reproducible manner
- Analyze the data
- Interpret the data and draw conclusions that serve as a starting point for new hypothesis
- Publish results
- Retest (frequently done by other scientists)
There are many subtle differences between the scientific method and problem solving in lean however let’s just consider the last two steps of publishing results and re-testing by others for validation. Publishing in the scientific world involves a lengthy document that has to survive this incredibly difficult process called peer review and then be accepted for publishing in a journal. In the best cases all the data for the experiment, code, test methods, equipment and other necessary items are made clear as well. In this way the test can be re-tested and validated by others. That is the objective beauty and rigor of the scientific method and that process can take years! (Think of the climate, solar system, physics, and other realms). We don’t have that kind of time in business or we’d go out of business! We have to solve problems much faster and the bar for re-validation by others is much lower for the most part.
That extended detour now complete let’s try and get back to the question of why it is hard for organizations to problem solve across boundaries. There are multiple reasons why that I will just touch upon. First there is the mistaken notion that having knowledge of methods is enough to solve problems and it is not in many cases. Toyota is a great problem solving company in engine production for example because of the towering pockets of expertise in the organization about casting, forging, machining, testing, etc. etc. etc. When you are an expert in something AND you know how to problem solving things move along more quickly. Otherwise you are going around in circles hoping to find an insight by mere trial and error. Scientist have an advanced degree in a specific subject matter and general expertise in the ways of the scientific method. The same must be true in companies for lean to work. Don’t fall for the “methodista” view that all you need to know are the right questions and some general knowledge of problem solving or kaizen. Reality is far more complicated.
Additionally companies struggle because they often fail to make distinctions between various types of problems and type of approach needed. For example some problems are best worked upon by individuals while some are best handled by groups. Some problems are easy (Just do it!), some are medium (need some structured PDCA routine) and some are really really hard (time to brush up on those DOE techniques!). Using the same approach in all cases is a mistake that Six-Sigma companies made early on. It is like driving the Ferrari to the grocery store when a Ford Focus will do just fine. Similarly some problems reside very much in a single process (vertical) while others tend to flow across the company (horizontal). Different resources are needed in each case to address the problem. Also some times decisions or policies just need to be made or training needs to take place. This is not really problem solving but can make a huge impact when done correctly, etc. Even problem solving has many nuances such as decision based risk analysis, gap from standard problems, difficulty in maintaining standards, improving upon standards, etc. etc. I won’t even mention the different types of problems faced in product development versus HR, or production versus IT. Different techniques are employed in each case on a needs basis. Only a general framework (e.g. DMAIC or PDCA, etc.) at best will be the same.
The difficulty I suspect that organizations will face and struggle with is the big tendency to have a “one size fits all approach on this topic”. Follow my thinking rules. Fill out this format to guide your thinking. Always assemble a team, etc. etc. My retired Toyota friends like to point out that it is usually the fool who knows one way of doing things and insists that you follow that one way. There is even a Japanese proverb about that point. Conversely it is the expert who knows many ways and the strengths and weaknesses of each one.
On a different but tangential vein if the scientific method was the holy grail of TPS then our national laboratory system would be the most efficient place on the planet and PhD’s from the lab would be exported as our greatest lean leaders. Unfortunately that is not the case by any stretch of the imagination. Methods alone do not guarentee success. Not every company will enjoy the same success level and not every scientist will win a Nobel Prize. There is still this reality of skill and execution involved. Problem solving in an organization of any size is also a difficult exercise because people are involved with different learning patterns, behaviors, expectations, attitudes and other factors. Problem solving, Kaizen, and other improvement flavors derive from from the scientific method. Of that there is no doubt. However that “method” will not be easy or enough to produce the results that Toyota achieves in business. There are other factors at work including strategy, leadership, execution, people development and many many other topics as well. Still I urge everyone person and organization to work on problem solving skills. It is a no regrets move like exercise and proper diet.