Day 3 - Solve the Performance Issue
The team started at 8 AM. As usual, my co-leader and I arrived early
to be sure that our meeting area was still set up correctly. Once the
team arrived, I began the session with a brief review of the purpose
of the event, our approach, and what we had accomplished thus far. I
asked the team for any thoughts or comments about our progress to date
before we looked at the day's agenda. There were none, so I previewed
the day's agenda with the team and got its feedback.
The priority focus for Day 3 was Task D3. Solve the Performance Issue (Exhibit
15). In this task, the team conceives and selects the best ways to achieve
the Kaizen event's goals and elevate the performance of the target work
process. The team had four main activities to complete: (1) generate
improvement ideas, (2) trim improvement ideas, (3) conduct an experiment,
and (4) select improvement ideas. The team would only conduct an experiment
if it needed to test whether an idea would be effective in eliminating
Generate Improvement Ideas
There are two main techniques a team uses to generate improvement ideas:
brainstorming and "ask an expert." In brainstorming, team members voice
whatever ideas come to mind as ways to eliminate a form of waste. In
ask an expert, the team uses the input of other employees in the work
process; this input is collected in response to the pre-event flyer and
in the course of the personal interviews done during the walk through.
When available, the team also uses the improvements stored in best practice
databases and improvements already in use at other work sites. Our pre-event
preparation indicated that neither of these latter resources were available
in this case.
The team brainstormed improvement ideas to each work process separately,
since each work process had a separate set of goals and process observations.
We began with cylinder preparation. In our Kaizen method, we first build
brainstorming sheets as a focusing aid. Each sheet is a flipchart page
on which a separate goal is recorded. Below the goal, the team lists
examples of the waste targeted by the goal. These examples are taken
from the Process Analysis Sheet. The examples listed on each brainstorming
sheet are the most resource-consuming instances of the waste the team
observed during its process observation. Exhibit 16 presents the brainstorming
sheets for the cylinder preparation goals.
We split building the sheets among the team members so it took only
10 minutes to prepare them. We next proceeded to generate improvement
ideas. We worked on one goal at a time, with a time limit of 10 minutes
for brainstorming solutions. The team read over the examples as a way
to focus brainstorming. I stood at the easel and recorded the ideas as
team members offered them. In brainstorming, any idea is accepted and
listed. Our rules are simple: (1) one speaker at a time, (2) say whatever
comes into your head, (3) no one judges any ideas during brainstorming,
and (4) keep generating ideas until time is up. After the team finished
brainstorming ideas for all the goals, we returned to each goal and looked
at the input other employees had provided about how to improve the work
process. We added those ideas to our lists if they were not already included.
Once the team finished generating improvement ideas for the cylinder
preparation work process, we built brainstorming sheets for the blending
work process (Exhibit 17) and repeated the brainstorming procedure.
Trim Improvement Ideas
After generating improvement ideas for each goal, the team's task was
to review each idea and explore whether it was worth pursuing further.
This meant that, for a given idea, the sense of the team was that doing
the improvement would not be prohibitively costly, would be effective
in advancing the achievement of the goal, and would not harm safety.
If an idea satisfied these concerns, it was retained. If it did not,
it was dropped from further consideration. If the team was uncertain
about the idea, we did an experiment to test the question about which
we were unsure. The key to this trimming process is having the team thoroughly
explore each idea with the person who offered it so that everyone understands
just how it would work. Frequently, team members have very different
notions of what an improvement idea as recorded on the list really means.
The use of WWO skills is critical in drawing out the meaning of each
idea. This benefits the team later when it proceeds to execute the improvements.
After discussing each idea for improving the cylinder preparation process,
the team voted whether each was worth pursuing. All the ideas passed
the "worth pursuing" test. The team next explored the improvement ideas
it generated for the blending work process. Here, three ideas raised
concerns; the remaining ideas passed the team's "worth pursuing" test.
The first idea raising concern was moving the label scraping operation
from the blending work process to the cylinder preparation process. Nathan
was not happy with the notion of adding work to the cylinder preparation
process. He understood that it made sense in terms of the type of work
it was, but reminded us that he currently did not have the time to get
the process done as it stood. He asked, "How am I going to get it done
by adding still another operation to it?" The team took the concern seriously.
Without Nathan's sign-off, this improvement would not be made. Through
discussion, the team worked out with Nathan what the likely time savings
would be on the cylinder preparation process when we made all the proposed
improvements. We then struck a deal with him. Reggie, speaking for the
team, offered: "If we don't get down to this cycle time for cylinder
prep, then we will put the label removal and painting operation back
into the blending work process." Nathan agreed with the offer. Reggie
added that he would show Nathan a technique the fill operators used to
remove labels that seemed to work well.
The next idea that raised concerns involved eliminating the vacuuming,
purging, and venting operations in the blending work process. Vincent
summed up the feeling of those who supported the idea: "We are just duplicating
what has already been done in cylinder preparation. We never see any
benefits from it." But other team members felt that eliminating this
operation was a risky step and needed to be tested before it could be
pursued. The team decided to do an experiment on this idea.
The last idea raising concern involved reducing the rolling time for
filled cylinders from 10 minutes to 5. The rationale was that the added
time was unneeded. Reggie said: "We had to do this [reducing rolling
time] a few times last year to get products out to customers who were
screaming for them because they were so delayed. Our backs were to the
wall. We did it, and we had no problems. Maybe we can do it regularly?" On
the other hand, some team members worried whether we could reach the
mixture levels of the blended gases that we promised our customers by
rolling cylinders for just five minutes. Again, the team felt an experiment
Conduct an Experiment
An experiment tests the effectiveness of an improvement idea. It provides
the information needed to rate the value of an idea in eliminating waste
so that the team can make the most intelligent selection of improvement
ideas. We structure an experiment by using a template that guides the
team in defining the question it is exploring, the method it will use
to develop information about the question, and the test it will apply
to judge what the answer to the question is. This experiment protocol
is also useful when the team must submit a formal request for a change
in a work standard. We add to the protocol the findings of the experiment
and attach it to the submission as evidence supporting the request.
The team developed and then executed protocols for each experiment.
The protocol for the rolling experiment is presented in Exhibit 18, along
with the results found. The reduced rolling time proved effective in
ensuring the proper mixing of the blended gases. Eliminating the added
vacuuming of the cylinder in the blending process also proved effective,
as it did not result in added impurities in the cylinder. The experiments
took an hour and a half to complete. Clarice, Reggie, and Vincent did
the experiments, while the remainder of the team moved to the next task.
18. Experiment: Testing the Effects of Reduced Rolling Time
whether reducing the current standard rolling time cylinders from
10 minutes to 5 will produce an adequate mixing of gases.
||Clarice T., Reggie
B., and Vincent L.
||If we compare the mix
achieved after rolling a sample of three cylinders of the most complicated
nonflammable blend for 5 minutes with the mix realized after rolling
for 10 minutes, then we will observe no difference.
||Three B-size cylinders
each with a blend of 10% N2/He. This blend of gases is
the most difficult for which to achieve a proper mix.
||Analyze gas to determine
||If we find that the
mix of the N2/He gases analyzes at 10% for each cylinder
after 5 minutes of rolling, then we will conclude that rolling for
5 minutes is effective.
||A 10% N2/He
mix was blended in three B-size cylinders. The N2 (the
heavier component) was added first; the He was added second. Three
cylinders were rolled on a cylinder rolling device for 5 minutes.
The cylinders were analyzed, and the level of mix achieved was compared
to the target level (10%).
||The mix in each cylinder
analyzed at 10% after 5 minutes of rolling, the same result achieved
by rolling for 10 minutes.
||Five minutes is sufficient
to allow mixing of gas components when the heavier gas is added first
and the lighter gas second.
||No additional learning.
Select Improvement Ideas
The team members not involved in conducting the experiments reviewed
the list of ideas for improving the cylinder preparation work process
to determine their priority for execution in the event. We used a voting
method to rank the ideas, placing at the top the ideas that were most
effective, least costly, and most "doable" within the confines of the
event. The team then performed the same voting activity for the list
of ideas for improving the nonflammable blending work process and included
the two ideas under study simply because removing them, should they fail,
would be easier than adding them. The team then decided which of the
ideas on each listbeginning with the highest priority ideasit
could implement in the event. Any remaining ideas were placed on a follow-up
action list to be pursued after the event. One idea placed on the follow-up
action list involved installing a switch in the blending booth for the
oxygen pump. This would not have been costly (estimated at $200) but
would have greatly reduced travel/transport, as the oxygen pump was some
100 feet from the blending booth and had to be accessed repeatedly. When
discussed by the team, there was enthusiastic support for the idea but
a firm conviction that it would take a great deal of conversation with
the Engineering Group which would have to approve the actionand
which was against such an idea. On this basis, we put the idea on our
follow-up action list.
The team listed the improvement ideas selected for execution in the
event on a separate flipchart page (Exhibit 19, next page). When Clarice,
Reggie, and Vincent finished the experiments, they returned and reported
the results to the team. Both experiments proved the effectiveness of
the improvements tested, so the list of ideas for implementation in the
event did not need to be adjusted.
Close Day 3
Once we finished listing the improvement ideas we would implement in
the event, we completed our close-of-day activities. I reviewed the day's
agenda and what we had achieved. Then we completed our plus-minus exercise.
The team again felt that one of the pluses for the day was its use of
WWO skills, which helped especially when it explored the different improvement
ideas brainstormed. The other big plus was the brainstorming exercise
because, team members said, they could get all their ideas out and no
one stopped them short. They had no minuses on Day 3. Before we finished
for the day, I alerted the team as to the focus for the remaining two
|Kaizen Desk Reference Standard
||Excerpt: Kaizen In Action