Shingijutsu Gemba Kaizen Seminar – Kaizen Key Points

The last post was a bit of a narrative, and I think it is appropriate to call out a few key points and express them succinctly.

  • The theoretical stuff all emphasizes “initial process stability” as a requirement for progress. Ohno said “without standards there can be no kaizen.” Mark says – “Without parts there can be no assembly.” Everyone knows this, then goes ahead and tries to implement standard work in the assembly areas. Bzzzzzzzt. Wrong answer. You simply can’t if they are blocked from performing to the standard.
  • Key Point: Stability is often implemented from the outside in. If you have unreliable suppliers, or an unreliable or inconsistent conveyance process, you are going to have to stock the inventory you think is necessary (DO THE MATH!) to buffer you supplier’s issues from assembly operations.
  • Once we started even going through the limited motions of establishing stability in three operations, it was clear that there was available cycle time for the waster strider (mizusmashi).
  • Key Point: Adding a water strider (if done correctly) does not add cycle time etc, to the supported process. There simply isn’t any reason NOT to do it.
  • Once again the standard work combination sheet proved to be one of the most valuable, least understood, and least used tools in the kit.

One of the features, if you can call it that, of attending these seminars is that you go home with a nice stopwatch. In this instance, the stopwatches have a lap-time recording feature. You can leave the watch running, click the left button, and it records the elapsed time for each sub-event.

Intuitively this is useful for cycle time studies, however I found a few disadvantages from the old way of recording the continuously running time on the sheet and breaking down the individual times later.

  1. If the work cycle never varies you can happily click away at each observation point and you will get great cycle times. But if there is variation in the work cycle itself, you are hosed. The stopwatch only records times for “LAP-002; LAP-003” etc. Taking times on a sheet of paper allows you to record what was actually happening even if it varied from the work sequence you originally wrote down.
  2. Press the wrong button at the wrong time and you lose it all – the stopwatch clears. Actually I think the lap times remain in memory, but still, I am not comfortable with this. Make a mistake on a paper time observation sheet and you can recover. Even if you accidentally stop and clear the watch, it is easy to recover by re-starting the total elapsed time baseline.
  3. Maybe I am old school here, and I admit that I have lots of practice, but by the time they were finished explaining how to get the lap times out of the stopwatch and write them down, I had already calculated all of the component task times, total cycle times, and was finished.

The bottom line with the stopwatch bit is that I believe they would better serve their clients’ education needs by teaching the running-time method. Anyone who gets a lapping stopwatch can figure out how to do it the other way. The running time approach does require practice to get down. But I think it is more robust. YMMV, that is just my opinion.

4 Replies to “Shingijutsu Gemba Kaizen Seminar – Kaizen Key Points”

  1. Great Article i’ve recently been on a Shingijutsu event my self, only that they came to my plant… i woould’ve preferred to travel to Japan.

    on the 4th Key Point you mention a water strider… is this the same thing as a water spider? was it a translation mistake or is it a different kind of thing?

    and on the 5th Key point you HAVE to remeber people what Ohno said about standards “Where there is no standard ther can be no Kaizen”

  2. Alberto –
    Thank you for the comment.
    As for water strider/spider – the Japanese is ‘mizusumashi’ which, AFAIK, is a diving beetle. However, at some point, the little bug got changed to a “water strider” also known as a pond skimmer – a little insect that scoots around on the surface on four legs, supported by surface tension. “Water spider” came into being when interpreters heard “strider” but said “spider”

    Honestly, the name doesn’t matter that much. The concept is that assemblers should not have to leave their assembly work to retrieve parts – and ideally, they should not even have to change parts orientation.

    “Where there is no standard there can be no kaizen” – absolutely true. I think many people insert the word “work” after “standard” and change the meaning.

    The point of the quote, as I currently understand it, is that improvement originates when there is a gap between “what should be happening” and “what is happening.” If there is no standard, then there is no way to know “what should be happening” because it is not defined.

    Thus, if I am looking at an operation, and I can’t tell what should be happening, the very first step is to clarify. Initial 5S and visual controls are often the tools put to use here.

    It is *not* necessary to first establish a standard for every little detail. Rather, standardization grows with time as you apply the PDCA cycle. If the first question – what should be happening? – is answered “Not sure” then the first step is “clarify” which is to apply ‘P’ As you go around the improvement cycle, you will always come back around to that question.


  3. Hi Mark,
    I know this is an old post, but I have a question. I don’t completely understand the “running-time” method vs the lap time method you are describing. In the past when I was charged with completing time studies, I would observe the process and write down the steps that were included in a complete cycle. The second phase was actually timing the sub-steps using a stopwatch that had the lap ability. I believe my method is exactly what your team was being taught. How does it differ from the running-time method? Thanks for the insight!

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.