This week we cover: Rationalizing production, TPS Engineering, Cognitive load research, Single piece flow. Our weekly improvements,  and a discussion of developing software vs buying off the shelf.  
Continuous improvement is not unique to lean. It is something we all do, but lean and TPS do bring a systemized and scientific approach to improvements that is very helpful. TPS is not about reducing waste. It is about solving your problems in a rational and scientific manner. So what is meant by rationalizing your production? Don't seek local optimum but instead rationalize your entire process to and make sure your decisions are inline with your goals.  Ask why 5 times to get to the root cause and solve the problem at the source. 
Toyota seems to focus on the human element.  
"We are what we repeatedly do. Excellence, then, is not an act, but a habit." - Aristotle  
"Practice does not make perfect, practice makes permanent. Perfect practice makes perfect."  
TPS Engineering achieves better results with half the engineering hours in half the time. Their teams, unlike most companies, are large at first then get smaller over time. The savings they seem to be leveraging is reducing rework. 90% planning, and 10% execution. This is hard because it shows slow progress at the beginning but has very little rework. Devin relates this to his process of designing fixtures. How do we think about this approach when compared with the concept of "fail-fast, fail cheap"?  Do we fail fast and cheap on a whole project or can we break things down to de-risk a crux of a process, it can avoid a lot of rework.  
Mental work load is something that is barely talked about in an explicit way in a lot of writing on TPS. There is some very interesting research on this topic. Uriel spoke with Kenneth Kotovski who spoke about some of his research on the impact of "problem isomorphs". They saw a 15x increase in time of solving a structurally identical problem. There are many examples of difficult mental tasks that could be eliminated or greatly simplified. There are so many examples of us overusing our brains which pulls them off of much more useful tasks.  
Single piece flow. What is it? Devin and Uriel give an overview of single piece flow. The idea is that a single piece of the thing you're producing moves from one process to the next with no waiting between and not in batches. It's very hard to achieve. Eliyahu Goldratt mentioned that Ford used a spatial constraint, Toyota uses inventory constraint, to achieve flow. He then proposed using the release of stock to limit overproduction.  How do dishwashers impact the kitchen. This is a good example how batches have far reaching effects. More storage space for more dishes means traveling further to move around the kitchen because you need so many dishes because of the batch size of a dishwasher.  Single piece flow as a north star to move toward. It has far reaching impacts from quality, space, and highlighting new issues to solve.  Devin and Uriel go over some of their small improvements.  Devin shares some of the challenges he's having with quoting new jobs and the software available. There are some very expensive solutions that MAY do what he needs, but there are also some cheaper options including just building tools in house. The question is which direction to go.