Dear Chris,
The explanation is in the specification
text, which says something like “IDEF0 with the added capability to
represent multiple decompositions with each context”. That provides
the OR capability – choose a decomposition (i.e., the search program
chooses it using heuristic values) that maximizes the heuristic you want to use
in expanding the original (highest level) context.
It is true that the original formulation
of IDEF0 followed its predecessor in specifying only one decomposition per
context, but I added the capability to represent multiple decompositions
because no two projects are actually the same. Every project has some
twist.
Also, project planning requires the
planner to choose which mechanisms (e.g. which programmer, which DBA, which business
analyst, which server, …) are to be bound to the class symbols used as
mechanisms. That makes OR nodes in the graph of all reachable projects
given the best practices graph of past projects, metrics of each project’s
results, and other choices made in planning, and later further bound during
execution until each variable is ultimately bound to a single mechanism.
Likewise, the controls are variables in
the highest level context because different projects are subject to different
quality standards. Some are ISO standard, others are standards for
specific customers, and further narrowing of the variety of substitutions
occurs as the contexts are iteratively filled in from the planning stage to
delivery of the final result.
Inputs and outputs are variable (in best
practices IDEF0 databases) because they can be chosen from among available
signals. Radar inputs, text message formats, XML message types, and other
signals vary from project to project also. Ultimately, the best practices
IDEF0 database has to be bound to exactly one (perhaps even a new one to be the
product of some other IDEF0 activity).
It is the ability of an IDEF0 activity to
produce as output another IDEF0 activity which makes even wider variety
available to planners, managers and developers than can be totally represented
a priori in the database. So the activities to be specified during
development are very skeletal, and are then filled in during execution of the
project, after each TBD activity is finally completely specified and ready for
execution.
So IDEF0, in its original form of a
completely specified diagram set with no variables, is as you note, not an AND/OR
graph, but simply an AND graph. The substitution of variables with
partially constrained types, and the ultimately completed TBD activities and
objects, make it far more useful though.
The patent specification is couched in
terms of medical records databases which can be used to discover the variety of
treatments and medication practices stored therein, but my experience with it
was also in project representation, which is where I found a need to expand on
the original, more rigid formulation.
HTH,
-Rich
Sincerely,
Rich Cooper
EnglishLogicKernel.com
Rich AT EnglishLogicKernel DOT com
9 4 9 \ 5 2 5 - 5 7 1 2
From:
ontolog-forum-bounces@xxxxxxxxxxxxxxxx
[mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of Christopher Menzel
Sent: Sunday, August 07, 2011 2:03
PM
To: [ontolog-forum]
Subject: Re: [ontolog-forum]
intangibles (was RE: Why mostclassificationsare fuzzy)
On
Aug 7, 2011, at 1:58 PM, Rich Cooper wrote:
In
an And/Or graph (e.g., IDEF0: http://www.englishlogickernel.com/Patent-7-209-923-B1.pdf
figures 5 and 11A)
IDEF0 is a business process modeling
method that has no boolean logical constructs. I do see something
IDEF0-like inside the "index cards" labeled AND and OR in figure 11A,
but neither figure 5 nor figure 11A is an IDEF0 model.