REPLY Definition of root cause (SD6774)

[SDMAIL Jack Harich]

Posted by  Jack Harich <register at thwink.org>

Fabian Fabian wrote:
> Jack,
> If we refer to Donnella Meadows "Places to intervene in a System" , 
> one could assume that for designing powerful solutions/interventions 
> one should always dig down until reaching the paradigm level.
Thanks. I studied her paper awhile back, both the original and the later 
version. Based on my own definition of high leverage point (HLP) versus 
hers, I don't think she made a convincing case that "one should always 
dig down until reaching the paradigm level." Here's why, using a quote 
from elsewhere in the first draft of the paper I'm writing. This comes 
right after the previously quoted material:


Alex Leus pointed out the definition at 
http://en.wikipedia.org/wiki/Root_cause.

I've worked for so long with my definition and its role in a problem 
solving process, that to be honest, I didn't even lookup other 
definitions before making the post. So thanks Alex. This helps.

It's good to see "The basic concept is that solving a problem by 
addressing root causes is ultimately more effective than merely 
addressing symptoms or direct causes." - To me this is THE key point.

Also notice "... the preceding example highlights one difficulty with 
root cause analysis: knowing when to stop." and "There is little 
agreement as to the types of conditions that can reasonably be 
considered root causes." and "Practitioners of root cause analysis often 
define what the phrase "root cause" means for a particular setting and 
application." - That's why a system dynamics centric process needs a 
formal definition of root cause.

Getting more subtle, Wikipedia says "An issue closely related to solving 
an existing problem is to foster learning that will embed knowledge 
(within a person, group, or organization) that may help prevent similar 
problems from occurring in the future." - This brings up the concept 
that there are different types of root causes. This is an important 
point. There are two types in the Wikipedia entry: The normal root cause 
and the root cause of recurrence. The quote from the paper did not cover 
this, but the System Improvement Process handles this via three reusable 
subproblems, one of which is how to prevent problem recurrence. The 
three subproblems are described elsewhere in the paper, as well as at 
http://www.thwink.org/sustain/glossary/SystemImprovementProcess.htm.

Thus there are three distinctly different root causes to most difficult 
social problems. Unless you have found and resolved all three, the 
problem is unsolved. Worse yet, if you are looking for all three types 
of root causes at the same time without realizing it, which I suspect 
most are, then you are rarely going to find all three. Instead, you will 
satisfice and stop which you've found only one or two. And even then, 
due to the confusion induced by lack of awareness of the need to look 
for all three, whatever you have found will probably not even be one of 
the three true root causes.

The entry pointed to http://en.wikipedia.org/wiki/Root_cause_analysis. 
In here is, sure enough, a "General process for performing and 
documenting a root cause analysis based corrective action." The steps are:

  1. Define the problem.
  2. Gather data/evidence.
  3. Identify issues that contributed to the problem.
  4. Find root causes.
  5. Develop solution recommendations.
  6. Implement the recommendations.
  7. Observe the recommended solutions to ensure effectiveness.

Step 3 corresponds to step 2A of the System Improvement Process (SIP). 
Step 4 corresponds to step 2B. Step 5 corresponds to step 3 of SIP, 
which is solution convergence. Notice the big intuitive leap from step 4 
to 5 above. SIP says whoa, this is too big for most mortals to do 
reliably. They don't eat nails for breakfast. They eat Cheerios, or in 
my case, granola. :-) Plus there is no concept of leverage points. 
That's why SIP has steps 2C, 2D, and 2E.

So the seven step process above is for easy or medium problems. It's 
fine. I can see it's what I use for everyday problems, like fixing my 
car. But for difficult problems, especially  social problems like 
conflict, sustainability, obesity epidemics, etc, a much more powerful 
process is needed.

Thanks again Alex,

Jack
Posted by  Jack Harich <register at thwink.org>
posting date  Mon, 03 Mar 2008 21:04:20 -0500