Flowcharts to Stock/Flow Diagrams

[John Gunkler jgunkler sprintmail]

Posted by ""John Gunkler"" <jgunkler@sprintmail.com>
I am doing much work in Lean Six Sigma consulting these days. One of the
fundamental things neophytes learn to do is to flowchart processes within
their organizations. And they ALL (almost without exception!) instinctively
do it backward from the way we create stock-and-flow diagrams! That is,
they all put flows in boxes and then connect the boxes with arrows.

I've discovered that, in their heads, all the arrow connectors usually mean
is ""this process happens before this other process.""

Here's how I take people the next step toward an understanding of dynamic
processes:

1. I teach them to redraw their flowcharts by taking the words out of their
boxes and putting them as labels on arrows (no boxes at all, at this stage.)
I help them learn to make sense of this ""backward"" way of thinking by saying
that a process is not a thing (or a ""noun"") but a verb, an action, a
transformation; and arrows look more like verbs, while boxes look more like
nouns, don't they?
2. Then, I remind them of the fundamental principal of processes: All
processes have inputs and outputs. In fact, the way you know that a process
exists is if there are ""things"" that change into other ""things."" Whenever
that happens, at least one process (""transformation"") must exist between the
two ""things."" They have already learned about the S-I-P-O-C diagram
(Supplier-Input-Process-Output-Customer.) So I'm merely reminding them of
the I-P-O in the middle.
3. So now I ask them to consider each arrow on their redrawn flowcharts and
ask themselves: What does the input to this process (this arrow) consist
of? Draw a box to the left of the arrow and put that description into it.
Then, what does the output consist of? Draw a box to right to contain this
description.
4. Then I teach them the ""boxes and arrows must alternate"" rule. They buy
it because they see that (a) you cannot have an arrow running into another
arrow -- because ""every process has an input and an output"" and (b) you
cannot have a box next to a box because, if the two boxes contain different
things the only way that could happen is if there is (at least one) process
in between to transform the contents of the first box into the contents of
the second box, and because if the contents of the two boxes does not
differ, then you really only have one box.

We never have to deal with CLD's at all coming at things this way. We start
with stock-and-flow diagrams (after a little shift in their mental models.)


----------------- Original message (ID=FB6A05C2) (65 lines) -------------------
Received: from smtp.albany.edu (mail1.its.albany.edu [169.226.1.105])
by listserv.albany.edu (8.13.0/8.13.0) with ESMTP id j479jMb3007926
for <SDMAIL@listserv.albany.edu>; Sat, 7 May 2005 05:45:22 -0400 (EDT)
Received: from hermes2.cf.ac.uk (hermes2.cf.ac.uk [131.251.0.113])
by smtp.albany.edu (8.13.0/8.13.0) with ESMTP id j479jHlv022632
for <SDMAIL@listserv.albany.edu>; Sat, 7 May 2005 05:45:18 -0400 (EDT)
Received: from maincf1p.cf.ac.uk ([131.251.4.20])
by hermes2.cf.ac.uk with esmtp (Exim 4.34)
id 1DULrn-0005tO-4O
for SDMAIL@listserv.albany.edu; Sat, 07 May 2005 10:44:59 +0100
Received: from Cardiff_University-MTA by MAINCF1P.cf.ac.uk
with Novell_GroupWise; Sat, 07 May 2005 10:44:59 +0100
Message-Id: <s27c9c2b.068@MAINCF1P.cf.ac.uk>
X-Mailer: Novell GroupWise Internet Agent 6.5.4
Date: Sat, 07 May 2005 10:44:22 +0100
From: ""Yi Wang"" <wangy15@Cardiff.ac.uk>
To: <SDMAIL@listserv.albany.edu>
Subject: REPLY Game theory and data mining (SD5253)
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
X-Virus-Scanned: Cardiff University Sophos AV
X-Scanned-By: MIMEDefang 2.49 on 169.226.1.44

Hi:
Thanks for the reply I am very interested in your research. Personlly I am
looking at blending system dynamics and game theory. Find their pros and cons
and try to find an better solution. Pls send anything you wish, I think I need
much help on thsi one.

best regards

yi wang
Posted by ""John Gunkler"" <jgunkler@sprintmail.com>
posting date Fri, 6 May 2005 23:12:31 -0500