Real Lean Transformation

When Great Lean Tools Fail

You may have all seen the spaghetti diagram that took months to complete, yet provided very little in the way of results. Or, you’ve seen the standard work that was methodically created by external assessors using measuring tapes, stop watches and pages and pages of process steps; the type of work which never really lives up to your expectations. What is wrong here? Why do so many implementations of Lean Tools fail to live up to the expectations placed upon them, or have any sustainable results? 

The answer is probably some combination of bad tool, misapplication of a good tool, or my personal favorite (and the one that happens most often); good tool, detailed application, wrong time.

It is true that many Lean tools (i.e. Heijunka, 5S, Fishbone Diagram, Spaghetti Diagram, etc.) can be applied in a variety of ways and industries, but many Lean practitioners fail to realize that improper timing of implementation can cause the tool to fail. Let me illustrate this with a story.

While working on a lean implementation in a lab, I came across an area that was being assessed for its potential shift to a work cell.  The current state of the area was that of confusion with individuals running to and fro, from one side of the lab to the other in the accomplishment of their testing tasks. Those conducting the assessment stood patiently for hours, thumb on stopwatch to obtain the most accurate representation of what work was actually being done. 

The total amount of different work that was being done was in the region of 60 hours of work each week. This work was done several different times per week, and so there were plenty of opportunities for those measuring to collect the data. The measurers were also aware that only taking a measurement a single time would not provide a very accurate measurement. As a result, each activity was measured 3-4 times to improve accuracy. This amounted to 150-200 hours of work, measuring the current state.

Unfortunately, the lab did not have standard processes for their work. Without a stable and standard process, the same activities or same order of activities was not being measured. In real terms, the results of the measuring activity were much less useful than anticipated. In this case, we had also had the analysts provide estimates of how long the work would take, which turned out to be less than 5% different, overall, from the actual measured times. 

What can we learn from this? There is a time and place for each tool. We shouldn’t suppose that just because we used a certain tool successfully (perhaps to great success) in a specific instance, it will work just as well in other situations.   The underlying principle, for this particular example, was understanding the current state. There are many ways to do this and just as many different levels of detail that can be obtained. Implementing detailed time measurements on a process which is long, unstable, and not standardized, will probably not get the desired results. However, implementing the same tool months, or even years, down the road can be very valuable when a standard process is in place. Also, when a standard process is stable, it will not require as many measurements to be considered accurate.

This blog was written by Preston Chandler, Consultant at BSM-USA Inc. For further information on implementing Lean and the use of Lean Tools send an e-mail to Preston Chandler.