REPLY Getting a Good Problem Statement (SD6602)

[SDMAIL George Richardson]

Posted by  George Richardson <gpr at albany.edu>

On Sep 8, 2007, at 7:10 AM, SDMAIL Jim Hines wrote:
> You're right that defining a problem in terms of reference modes
> pre-supposes that people have already decided to use SD.


I don't think that's quite right.  It presupposes that you want to  
address a dynamic problem, and to define to the problem dynamically,  
that is, in terms of graphs over time, but that does not mean that  
you are constrained from that point on to use the rest of the system  
dynamics approach.  I would think there could be any one of several  
follow-ons to starting by defining the problem dynamically, including  
micro-simulation (agent-based modeling).

We have seen several settings in which just asking a client group to  
draw graphs over time of the issues that concern them helps the group  
with their problem(s), usually beacuse they had not previous taken a  
look at the problem specifically over time.

I'd definitely agree with Jim, if I understand him rightly, that  
asking folks to draw graphs over time of what they're interested in  
is absolutely essential for getting a good problem statement to  
approach the problems using system dynamics tools.  But I also tend  
to think its a good idea not matter what approach the group ends up  
using.  Thinking dynamically draws us naturally into putting events  
in a context that moves over time, and that's smart thinking whether  
one tries to see those dynamics in stock-and-flow / feedback terms or  
not.

..George 
Posted by  George Richardson <gpr at albany.edu>
posting date  Sun, 9 Sep 2007 10:54:16 -0400