I am in the process of (finally!) reading Mike Rother’s great book Toyota Kata. The book has already gotten great reviews out there, and I am not going to contribute a lot more to that dialog except to say I think it is the first substantial addition to community knowledge about TPS since Steven Spear published his dissertation in 1999.
As I go, I am taking notes of thoughts that are provoked, and I will share a few of them as they come up.
Early in the book, Rother makes a clear distinction between managerial systems that try to cost justify each and every activity, and those which have their focus on an ideal future state.
He cites a number of examples of how the dialog in these companies is shaped by this vision, or lack of it. This is one:
Almost immediately the assembly manager responded and said, “We can’t do that,” and went on to explain why. “Our cable product is a component of an automobile safety system and because of that each time we change over to assembling a different cable we have to fill out lot traceability paperwork. We also have to take to the quality department the first new piece produced and delay production until the quality department gives us an approval. If we were to reduce the assembly lot size from five days to one day we would increase that paperwork and those production delays by a factor of five. Those extra none-value added activities would be waste and would increase our cost. We know that lean means eliminate waste, so reducing the lot size is not a good idea.”
The plant manager concurred, and therein lies a significant difference from Toyota. A Toyota plant manager would likely say something like this to the assembly manager: “You are correct that the extra paperwork and first-piece inspection requirements are obstacles to achieving smaller lot size. Thank you for pointing that out. However, the fact that we want to reduce lot sizes is not optional nor open for discussion, because it moves us closer to our vision of one-by-one flow. Rather than losing time discussing whether or not we should reduce the lot size, please turn your attention to those two obstacles standing in the way of our progress. Please go observe the current paperwork and inspection process and report back what you learn. After that I will ask you to make a proposal for how we can move to a one day lot size without increasing our cost.” [emphasis added]
To this hypothetical Toyota manager, the ideal state is achieving one-by-one flow, on demand, in sequence, as requested, without waste. He sees the drive toward this ideal state as the method to drive the next problems to the surface which, when solved, will remove a round of waste from the process.
There are a number of key points to take away from this.
Avoiding waste vs. removing waste. The assembly manager is confusing the avoidance of wasteful activity from removing wasteful activity from the process. The large lot sizes are avoiding wasteful activity, but it is still there, lurking under the surface as a barrier to further improvement.
One-by-one flow as a crucial goal. In this example, the concept of one-by-one (or one piece) flow worked exactly as it should. The discussion, or attempt, to implement the next step in that direction revealed the next barrier that must be addressed on the way. Without the imperative to reduce lot size toward that ideal, these reasons, justifications, barriers, excuses prevail and the wasteful activity remains. Yes, its effect can be mitigated by larger batches, but applying that logic, why not make the batches even larger?
It isn’t about what waste you can remove. It is about what wastes must be removed to achieve the next target condition. Having this sense of the ideal to define the direction of progress focuses the team on removing the next barrier to higher performance. This keeps things on a path toward higher system performance. Contrast this to the classic “drive by kaizen” approach where we go on waste safari and then look for opportunities where waste we can remove the most easily. This results in a patchwork of improvements that rarely find their way to the bottom line.
Cost justification would say leave the waste there. The cost justification for reducing lot sizes comes primarily from the reduction of inventory holding costs. Unfortunately, unless the inventory is insanely expensive (like jet engines), that alone rarely justifies the effort. But this thinking also stops any further opportunities for improvement in their tracks.
On the other hand, if the goal is to drive toward the ideal of one-by-one in a way that does not increase cost, then once this single-day lot size is achieved, they would be asking “What will it take to get to half a day?” Eventually they will be asking about every part every day. And then true mixed one-item-flow.
At each juncture an important thing happens. They must confront the next problem, and the next waste in the system. They must find a way to remove that wasteful activity without increasing cost. Eventually they will be making one cable each takt time, and following that, making the cables and installing them immediately.
To get there all of the inspections and traceability paperwork requirements will either be removed as no longer necessary or will be built in to the process itself. How will that be done? I don’t know. Nobody knows. That is the point. We can’t solve all of the problems ahead of time, nor can we solve them before they are discovered. The path to the ideal state is vague and unknowable when you begin. It is revealed step by step, as you take those steps.
The key is to have faith that, among all of the people in your operation, someone will find a way to crack that next problem.
This is how you harness people’s creativity.
In his classic JIT Implementation Handbook, Hirano says to force one piece flow to reveal the waste in the system. That was in 1988. Rother introduces the critical human element into the equation. One piece flow is not an abstract concept, it is a critical guide for management decision making.
Rother goes on to make other examples. In each case, the tool or technique – such as takt time or leveling – specifies how we want things to work. That provides a baseline for comparison so we can see the problems we need to work on next.
Here is another example. Ironically, while Rother uses it in his book, I have experienced it in real life. The factory is trying to introduce leveling. They set up a stock of finished goods. They set up a kanban leveling box, and they set up a fairly well thought out process of leveling the demand on the production stock, allowing the inventory to absorb the ups and downs. Their rule is that if the inventory hits a high or low limit, they will take action and assess what is happening.
Then the email arrives – “A big order arrived, and flushed out our inventory, we had to catch up, what do we do?”
After doing what is necessary to recover the system, this was a great opportunity to look at why orders arrive in big batches. In this case, I suspect it was an artifact of the company’s distribution network. The plant is at a critical crossroads.
Each of these examples leads me to the same thought.
The “Yes, but…” and the “Now what?” situations are going to happen. They prove the system is working to surface the next problem to be worked on. The reply that follows decides the fate of your improvement process.
You can reply the way the Toyota manager did in Rother’s example, and work hard to improve your process.
Or you can buy the Basic Story that is presented, and in effect agree that no further improvement is possible on this process, or that “lean doesn’t work for us because [put your “unique situation” here]”.
Which do you do?