When we started the Lean Learning Center, our explicit objective was to reshape the practice of lean to be principles-based and not tools-based. I tried influencing many of the top voices of the time, but a fascination with the magic of value stream mapping and the excitement of kaizen workshop events drowned out the message.

A different view exists today: Principles and behaviors matter, often more than the tools. However, teaching and actions directly focused on building a lean culture are still quite rare, as the draw to short-term results and tangible tool application is appealing. I have never seen an organization fail because they didn't have the right improvement tools. I have seen many organizations fail because they didn't have the right behaviors.

See Also: Operations Management Strategy & Best Practices

My early experiences really helped shape my thinking. I was managing material handling at Harley-Davison's main plant, where they had installed one of the first large-scale pull systems in the United States. All of the company's bikes were made in a very vertically integrated plant, including everything from laser cutting to chrome and zinc plating. The plant still seemed to treat material handling, and its system, as a secondary system, perhaps evidenced by the fact that they had given such an inexperienced person as myself so much influence on it.

The system was failing, and in order to understand it better, I turned off my radio and spent two weeks doing nothing but observing. I shadowed every material handler and understood where he went, what he did, how he did it, and what problems he encountered along the way. I challenge people to find four hours just to observe, so I am simultaneously astonished and gratified that I was able to commit two weeks to doing just that.

I observed that the system design was not the problem. The behaviors in using the system were the root cause of the failures. Assemblers would empty a bin and draw the kanban card when they saw the material handler coming in order to "help," but this sent false signals of demand back to the supplying operations. Machine operators would load "just a few more" parts than requested while they were set up, pushing extra parts out but delaying the supply of others. The aftermarket group would go straight to the assembly line and just draw whatever number of parts they were using, creating shortages the system wasn't designed to handle. And material handlers would grab an extra bin of parts if they had room on their cart just so they didn't have to make an extra trip.

The system design was fine, but it was only truly understood by a few in the plant. People were only taught the procedures of the system, as was I, during the system introduction. People knew what following the procedure meant, and when encouraged or forced, their compliance with the system increased. But because they weren't taught the thinking behind the system, any deviation from those procedures seemed trivial to them. But the consequences were often dramatic.

What I didn't understand at the time was that a pull system is in many ways fragile by design. Small problems and deviations result in problems, albeit small ones. This is by design, making small problems visible so that they can be fixed.

We failed to achieve this insight at the time, although we admittedly had bigger known problems that required whatever problem-solving skills we had available.The positive side of this is that we focused on behaviors around the system, and while we never achieved anything near perfection, a small increase in compliance had an outsized impact on the results. Failures were more isolated instead of compounding, and with only a couple of process tweaks and a lot of focus on behaviors, we began to get the system stable.
 

To read more of Jamie Flinchbaugh's articles on continuous improvement, visit www.iw.com/author/jamie-flinchbaugh.

Lean fails, above all else, for a lack of the right behaviors. In my next column, I will outline an approach for tactics to change behaviors.