Replying to LO22148 --
Hi Eugene,
I agree with you on all points:
> It is necessary to think about the whole system and for
> them to believe that best quality and least cost occur at the same time
> when the selected system works exactly right.
The whole system approach was in our case illustrated using the system
archetypes: that the cost of not preventing or solving a quality problem
comes back to you in one form or another. The problem in understanding the
working of the system is in looking through the instances of the problem.
Most people have learned to "solve" a problem by shifting the burden,
reducing your own pain, getting rid of the attention and have developed
for every instance of a problem a reaction. With instance of a problem i
mean that for me "material shortage" is the problem, while for most
planners, buyers, production workers, quality engineers etc. the shortage
of part number X differs from the problem of wrong colour of part number Y
(in my view: a shortage problem), which is rather a different problem than
retooling for using component Z in stead of A, because A was unavailable.
People become very effective in solving all these "small", local problems,
not noticing that the reappear and reappear.
Once there we had a shortage of a simple red light emitting diode (led).
I rembered that it had been a problem about a year ago. We looked into our
files: it was within two week correct. That same led was short a year ago.
We looked further and noticed that aagin two years ago the same led was
out of stock. Now, the normal reaction is just to blame the material
planner, who blames purchasing, who blames the supplier, who .... ok
problem gone as soon as they ship the component. We looked into the cause
of the shortage and found out that this led had very long "legs": the
light had to be up front, but the contacts were deep in the module. And
there was only one factory in the world who could supply us. And this
factory only wanted to make a large batch for us, so we ordered more or
less our years' supply. Ordering a years' supply hides the problem for at
least 11 months, so problem solved. For off course, we had a design
problem: the design had to changed to accomodate for a standard component.
> Another paradigm that must be changed is that long runs are efficient and
> adopt new indicators of performance. Historically we have used worker or
> machine efficiency. Two better measures of system performance are tow
> process results quality or elapsed time in process.
Machine or worker efficiency is only a relevant measure when your direct
costs are about 80% of your total costs. In most industrial firms, this
was in the nineteentwenties.
On the other hand, i returned to some "classical" measures, like delivery
performance, trends in stock levels and pipelines, inventory turns and
return on investment (but defined at the lowest level, for a production
cell, a production line).
Once i asked for a weekly summary report on goods ordered, raw material in
store, Work in Process, goods in stock, and goods in transit. The IT
department told me that that report had been canceled a few years earlier,
because there were now more modern costing and efficiency reports. I knew
that, they were blocking my view. I was able to prove that the costs a
production manager could influence was about 5 cents a day.
Kind regards,
Jan Lelie
--Drs J.C. Lelie CPIM (Jan) LOGISENS - Sparring Partner in Logistical Development Mind@Work - est. 1998 - Group Decision Process Support Tel.: (+ 31) (0)70 3243475 or car: (+ 31)(0)65 4685114 http://www.mindatwork.nl and/or taoSystems: + 31 (0)30 6377973 - Mindatwork@taoNet.nl
Learning-org -- Hosted by Rick Karash <rkarash@karash.com> Public Dialog on Learning Organizations -- <http://www.learning-org.com>