REPLY Definition of root cause (SD6781)

[SDMAIL Kim Warren]

Posted by  "Kim Warren" <Kim at strategydynamics.com>

This may be a slightly different take on the question, and I don't know
about its theoretical validity, but some principles from an SD depiction
of how things work seems to produce the following line of reasoning - at
least in the business situations I work in. 

Working back from the performance outcome we are concerned with [growing
sales, improving profits, fixing service quality etc. ] a logical
disassembly of the arithmetical relationships involved produces a causal
chain structure that ends with one of three types of factor:
1 - one or more stocks [customers, capacity, staff] 
2 - one or more decision-items [price, marketing, salaries]
3 - one or more exogenous factors beyond the influence of either
management or the system itself [demographic changes, competitors'
prices ...]
At any instant in time, then, these three categories seem to cover all
'root causes' [not sure how this fits with the way the term is generally
defined]

If these three items remain unchanged, then performance remains exactly
the same. The only way for performance to change under constant
decisions and exogenous factors is for stock levels to change [win
customers, lose staff ...], and the only way for that to happen is for
flow-rates to be non-zero. So the causal investigation asks what drives
flow-rates. Adopting the same principles above, an arithmetic
investigation of what causes what behind each flow rate seems to arrive
at exactly the same three items: 
1 - one or more *existing* stock levels [current staff drive service
quality, which drives customer loss rate]
2 - one or more decision items [price drives customer loss rate] ..
though some flow-rates are themselves simply decisions, e.g. adding
capacity
3 - one or more exogenous factors [demographics affect changes in
customer numbers, competitors' price drives customer losses]

This logic also seems to answer the question 'where to stop' - we stop
when we get to a stock, a decision-item or a factor that is outside the
control of management or of the system itself. This does break the
'close-the-loop' philosophy - but we know how to depict decisions
emerging from policies influenced by system performance, and many
exogenous factors are genuinely 'out there' for all practical purposes -
nothing I can do about demographic change, and little I can do to alter
competitors' prices. 

Kim Warren
Posted by  "Kim Warren" <Kim at strategydynamics.com>
posting date  Wed, 5 Mar 2008 07:20:53 -0000