Adopt a Pace for Project Work
July 18th, 2005 by Hal The Project Reformer's e-Tip of the Week |
| 043: Adopt a Pace for Project Work |
|
Toyota uses the term mura — inconsistency or unevenness — to identify one of the contributors to defective work. Unevenness gets in the way of completing what we set out to do. In a physical sense, variability in drill speed would result in poor quality of a drilled hole. Variability in the ingredients of a batch of batter would result in bad cookies. Six Sigma is Low variability in the project setting is harder to get our heads around. The desired outcome — always getting the intended result — is less a function of the materiel process than human processes. There are two keys to high project capability. First, what we ask people to do is already in a condition for completing it. Second, people manage the promises they make. Pacing project tasks is one way to eliminate one source of variability. This has similar effect as pacing a production line. Performers always know what is coming at them next and when they have to deliver. Establish a drum beat for doing work by designing tasks so they all take roughly the same duration to complete. For instance, establish a schedule for completing design details on an everyday basis rather than batches of different sizes. This will improve the reliability of completions which aids follow-on performers depending on that work as input to their work. The overall effect is to counteract the compound effects of dependence and variability.
|
©2005 Hal Macomber | RPM | e-Tip Archive | PDFs | Submit Tip |
I intend to publish 9 more Project e-Tips before ending this series. What have I left out? What help can you offer? Please send along your proposals.
Related Posts
- Pace Arrivals to Increase Throughput The everyday understanding of lean is do something without waste. Another aspect is to reduce variability. Doing s...
- Ask Two Great Questions to Minimize Misunderstanding In my last posting I wrote about misunderstanding and a few things we can do to avoid it. One of the best practical a...
- Toyota is Inspirational! I had two great plant tours this week. As part of the Lean Project Leadership (Shusa) Program that I comduct with Greg ...
- Adopt Agile Development I couldn't let this one pass. The folk at Trizle (that's the name) bring a wonderful writing style with their strong op...
- The Key to Improve Construction Safety Construction safety has to improve. Fast. But it hasn't. Not in over 15 years. About 1,200 people have been dying ...
July 19th, 2005 at 9:08 am
Six Sigma is the western response to attacking unevenness.
Mmmm. Six Sigma is *a* western response to attacking unevenness. possibly, but probably not.
Lean IS a western response. it flows from colt rifle via ford’s model t line, shewhart control charts, scientific method coupled with an understanding of variation, epistemology, systems and psychology. it includes the customer and the suppliers. it is inclusive rather than elitist.
OK so all these western elements were first put together in the tps.
It seems to me that Six Sigma improves everything - including the production of muda. it is a-systemic; it looks at every step in a pocess as if it produced value just like every other step. there is no value stream analysis, no deployment flowcharting. it is elitist and supports a command and control approach that does not sit well with collaboration.
That apart - must have been an aberation Hal - that appart I agree with everying you say. the faster you go the better it gets.
July 19th, 2005 at 9:34 am
Alan,
I don’t take exception to your criticisms of Six Sigma as a stand alone initiative. ‘Lean Six Sigma’ is becoming the accepted approach in the US for exactly why you criticize Six Sigma. And I’ll go one step further.
In our work with clients on their projects, we start with an assessment of the constraints to throughput to bring focus to the improving actions. This generally results in very fast paybacks on our efforts.
I’ll make the edit to the tip.
Hal
July 20th, 2005 at 9:21 am
Hal
Thanks for your response. YEs lean Six Sigma is better, so I’m surpridsed that in addition to changing *the* to *a*, you didn’t also add *lean*.
perhaps becuase there are other approaches than lean that can make a difference — on Monday Rip Stauffer posted the following on the Deming Electronic Network in a thread titled “No Conflict Between Deming’s Ideas and Six Sigma”.
After reading Rip’s post I wonder how widely Lean Six Sigma is understood compared with the original.
Six Sigma is, unfortunately, often practiced by hundreds
of hacks out there whose only exposure to Quality was through Belt training
conducted at the point of a gun by another hack whose only exposure to
Quality was through Belt training conducted at the point of a gun, whose
only…etc. (Rule 4 of the Funnel is alive and well among many of these
folks!). I have been in serious discussions with some of these people who
have told me things such as: “Hidden factory is all the mistakes that the
line workers make and hide from management,”"The best measure of center for
a negatively-skewed data set is the 75th percentile,”and “Systems thinking?
What’s that? Why would anyone need that?” among other things. This Rule 4
has been exacerbated by the fact that Welch popularized Six Sigma as a
cost-cutting strategy. Now you can pick up any five books written about Six
Sigma and find that it’s about cutting costs, that it ignores the customer,
that it’s about quality, that it’s customer-focused…you name it, it’s
probably been written. Then you have ASQ trying to define a standard for Six
Sigma (thus leading many to believe that such a standard exists).
That doesn’t mean that Six Sigma itself is bad. The “Six Sigma metric
(process sigma)” is, in my view, fundamentally flawed (although I have found
DPMO to sometimes be useful in estimating overall defect levels by
aggregating the continuous and discrete measures together–as long as you
understand that it is just an estimate, and you base it on knowlede of
stable processes). The idea of putting together a project that focuses on
working to uncover key elements of the common cause system, so you can
improve performance in a process that is stable but not performing where it
should, is a sound one, I think. Having an improvement-focused project life
cycle (DMAIC) or methodology to help guide a team through that complex task
is OK with me, and we’ve been pretty successful with it. Trying to use the
project methodology by itself as your whole quality system won’t work, and
it certainly doesn’t seem to leave Six Sigma leaders with anything like
statistical thinking (witness Welch and his minions who adopted “Fire the
bottom 10 percent every year”).
Six Sigma can be practiced by Deming advocates. I think most people in the
DEN would recognize that I am pretty consistent and enthusiastic in
attempting to learn and expand the Deming Philosophy. So much so, in fact,
that I had some misgivings when I left the Navy a few years ago and went to
work for another long-time friend and follower of Dr. Deming’s, Lou Schultz,
as he was trying to put together his Six Sigma practice. Lou always looked
at Six Sigma as a vehicle for carrying Quality principles and the philosophy
to organizations. I think he was right…I’ve been able to show the Red Bead
and the Funnel and talk about SoPK with a lot of executives now under the
umbrella of Six Sigma training and coaching. I’ve had many people in
organizations, people who had never heard of Dr. Deming, get very interested
in the Deming Philosophy and Quality because of video tapes we showed and
discussions we led in Six Sigma classes.
Best regards to all,
Rip Stauffer
Woodside Quality Solutions LLC
612-916-0197
[Alan writes: for those unfamilar with Deming’s Funnel experiment (let alone rule 4) see WE Deming (1986) Out of the Crisis MIT Press pp 327-8, go to http://www.managementwisdom.com/funexvol9.html, look through the slides starting at http://www.shsu.edu/~mgt_ves/mgt481/lesson5/sld007.htm
In the funnel experiment the intention is to get a marble on to a target by dropping it through a funnel. The four rules are:
Rule 1. Leave the funnel alone
Rule 2. Move the funnel from wherever it is at in an equal but opposite direction from where the marble landed in relation to the target
Rule 3. Move the funnel back to its rest position before moving it in an equal but opposite direction from where the marble landed in relation to the target
Rule 4. Move the funnel over the last position of where the marble landed
What the experiment shows is that adjusting a stable process to try to compensate for variation creates a worse output than doing nothing at all.]
July 20th, 2005 at 10:22 am
The reason I didn’t add “lean” to the Six Sigma is because it is not the usual approach. As Alan points out in his first comment the usual practice of Six Sigma is a-systemic and without regard to what is producing value for the customer or throughput for the firm. Practice significantly lags theory.