QUERY Models, Problems and Systems

[Ralf Lippold <ralf_lippold@we]

Posted by Ralf Lippold <ralf_lippold@web.de>

Kim,

you have made an interesting point that struck my thinking.
> Posted by ""Kim Warren"" <Kim@strategydynamics.com>
>
> Thanks Jack - I guess I'm questioning the assumption that we always build models
> *because* there is 'a problem' - surely people commonly want models [of whatever
> kind] to help understand and better-manage something,

The very nature of thinking about ""why things are like they are"" is
already a modelling process (first in our heads, John Sterman would call
that ""the mental model""). The next step -trying to get this into formal
modelling and simulation.

Perhaps we all should rethink what makes a problem a problem?

Shigeo Shingo's new book ""Kaizen and the Art of Creative Thinking"" gives
interesting answers on that. Even though not directly connected with
system dynamics it gives similar thoughts as Jay has given in the
initial system dynamics work.

Following the most interesting discussion I would like to add a few
thoughts that go through my mind while reading the posts for the last
days/ weeks:

1. As a systemdynamicist drawn to the field by observing complex
problems in an auto company that was facing problems coming back
persistantly over time I wondered what are the hidden root causes of
that specific system behavior. This is -on the other hand- strongly
connected to lean thinking, one of the pillars of Toyota's success over
the last decades (especially recently to be observed). Perhaps Toyota
is one of ""the SD companies"" in the world - who knows?

The stating what a problem really is, attracts engineers much more than
other people as they think very analytical and data driven. Clearly if
there is a stopping or slowness in the supply chain, the reason must be
that the throughput time is not fast enough. Easy to see, or?

Improving the throughput time makes to first problem go away BUT
strangely there occurs a new one -outside the engineers' vision- in the
outbound logistics point where way too many products come in now (as
everything is going very smoothly upstream). Problem No. 2! So these
guys are trying to handle their problem and manage -in the end- to
deliver the products faster to the customer (if there is still exceeding
demand on the product). And -as you might suspect- there will occur
problem No. 3 as the service department will get lots of requests as
there are much more products out in the market now and the chance that
customer come back with certain questions, request will have increased.

What do I want to make clear? There are still other underlying drivers
that lead to the effect that the people hit by the three different
problems will almost never talk together as they all perceive different
parts of the problem through there specific ""process glasses"" -shaped by
their ""departmental culture"". The question is, ""Can't they change the
glasses, so they could see a broader view of the problem case and solve
the problem in a sustainable way?"".

Doing a survey/ practitioners discussion myself on that kind of question
in the lean thinking area
(https://www.xing.com/app/forum?op=showa ... ;offset=40) I
have learned that people often don't challenge their own assumptions on
what the problem is. So there must be another problem behind that
thinking -and that is ever so difficult to break open!

2. As we all here in the SD mailinglist are either very experienced
systemdynamicists (as we all have noticed also the ""big shots"" are
taking place in the discussion - thanks a lot for all your
committment:-)) or practioners coming from company background there
tends to be a focus on ""our"" problem which we perceive as that.

Shouldn't we rather challenge the assumption and ask what the connection
to the experience of non-systemdynamicists is? For them our discussion
often sounds superficial and so they draw themselves back into what they
are familiar with (even though this won't lead them to any sustainable
change in their behavior, processes or policies:-().

So what could be done?

My proposal would be to spread the ""word of mouth"" into other virtual
""societies"" in order to grasp real life experience beside the academic
and consultant point of view (which of course is necessary and helpful
to bring the field forward).

One way of initiating discussion in such a way is implying topics from a
system dynamics /systems thinking point of view in networks (such as the
mentioned above) to attract people from diverse academic and working
background and I am more than sure there will be ""new"" questions and
approaches on how to connect system dynamics to daily work around the
globe -not necessarily simulation and perfect modelling but to get a
sound understanding what the system dynamics ""stuff"" is really about.

That is just my five bucks (two cents wouldn't do it this time;-)) on
the topic.

Cheers

Ralf

PS.: Still on the learning journey -always will be:-))

PPS.: As John Morecroft has pointed out in his comment dated (At 6:41
Uhr -0400 13.05.2008): creativity and time are essential to good
modelling -it depends additionally on the background:-). - Thanks a lot
John for your excellent post and the story around Jay's lecture which
definitely makes a great point!

Best regards

Ralf
Posted by Ralf Lippold <ralf_lippold@web.de>
posting date Wed, 14 May 2008 18:44:39 +0200
_______________________________________________

[Jack Harich <register@thwink.]

Posted by Jack Harich <register@thwink.org>

> Posted by ""Jack Homer""
> People should be encouraged to ask any question they want...

Perhaps it is time for a Frequently Asked Questions page?

Jack (Harich)
Posted by Jack Harich <register@thwink.org>
posting date Wed, 14 May 2008 08:36:42 -0400
_______________________________________________

[""Jack Homer"" <jhomer@comcas]

Posted by ""Jack Homer"" <jhomer@comcast.net>


Christina Spencer says:
> I agree that it is perfectly reasonable that people with questions that
> may have come up before should investigate before posting old questions

Let me state again that I haven't asked newbies to investigate before
posting questions, but rather that newbies show discretion when
RESPONDING to questions, recognizing that most questions, especially
questions about best practices in SD, have already been well answered
over the past 50 years.

I agree that a wiki would be the right place for an archive of compiled
wisdom regarding best practices. If somebody (or, let's hope, multiple
somebodies) had the time and were so inclined, they could work their way
through the exemplary published works of SD literature (both books and
articles) to find the nuggets (short excerpts) of wisdom, quote them
verbatim, and give them topical headings to make them searchable. I
would suggest this ""backward"" approach rather than a ""forward"" approach
of attempting to develop a list of frequently asked questions and then
going on a search for the answers; or, even more problematic, trying to
develop answers from scratch and reinvent the wheel. A ""compiled
wisdom"" approach just seems much more feasible to me than a ""FAQ""
approach. In a wiki, it could start small and grow as more people see
the value and become involved in the project. I don't know whether
there are copyright issues to worry about, but I would think most people
in our field would be only to happy to have brief excerpts of their
works quoted as part of Compiled Wisdom (!), as long as proper
attribution is given.

One resource that could perhaps help steer the effort, balancing the
backward with a bit of forward, is the study of best practices by
Ignacio Martinez presented at the SD conference in Atlanta in 2001 along
with a large panel discussion. This study involved eliciting best
practice ideas from SD experts, and ended up identifying 41 best
practices in the areas of problem definition, conceptualization, and
model formulation. This could at the least help identify some good
topical headings for the compiled wisdom.

Jack Homer
Posted by ""Jack Homer"" <jhomer@comcast.net>
posting date Thu, 15 May 2008 09:07:24 -0400
_______________________________________________

[""Bobby Powers"" <bobbypowers]

Posted by ""Bobby Powers"" <bobbypowers@gmail.com>



On Fri, May 16, 2008 at 1:11 PM, SDMAIL Jack Homer <jhomer@comcast.net> wrote:
> ... A ""compiled
> wisdom"" approach just seems much more feasible to me than a ""FAQ""
> approach. In a wiki, it could start small and grow as more people see
> the value and become involved in the project.

Lucky for us, Bob Eberlein set up a System Dynamics wiki:
http://www.systemdynamics.org/wiki/index.php/Main_Page

All it needs is some people to start filling it with their wisdom

-Bobby Powers

Posted by ""Bobby Powers"" <bobbypowers@gmail.com>
posting date Fri, 16 May 2008 13:59:31 +0200
_______________________________________________

[Tom Fiddaman <tom@ventanasyst]

Posted by Tom Fiddaman <tom@ventanasystems.com>

At 06:18 AM 5/15/2008, SDMAIL Christina Spencer wrote:
> I agree that it is perfectly reasonable that people with questions that
> may have come up before should investigate before posting old questions
> - but how are they supposed to do that?

One source of questions & answers is the archive of this list, at
http://www.ventanasystems.co.uk/forum/
(Specifically,
http://www.ventanasystems.co.uk/forum/f ... php?fid=34 )

It's easy to search, but not at all organized, so it could be hard to
find answers to general questions with nonspecific terminology. It might
be a good place to mine for material for the wiki.


One further thought: forum search tools typically aren't as good as
google. For example, if you search the SD archive for ""limits to growth""
there are many relevant results. However, if you seek ""discrete
continuous"", you get nothing, presumably because the two words aren't
normally adjacent. You can search the SD archive direct from google with
the following search string:

site:www.ventanasystems.co.uk/forum discrete continuous

This approach yields many useful results.



Tom
Posted by Tom Fiddaman <tom@ventanasystems.com>
posting date Mon, 19 May 2008 10:05:42 -0600
_______________________________________________