Mark Fox position paper

Tom Gruber <>
Full-Name: Tom Gruber
Message-id: <2878499961-6068858@KSL-Mac-69>
Date: Wed, 20 Mar 91  15:19:21 PST
From: Tom Gruber <>
To: Shared KB working group <>
Subject: Mark Fox position paper
\title{The CARMEMCO Project}

\author{Mark S. Fox\\
Center for Integrated Manufacturing Decision Systems\\
Carnegie Mellon University}


\subsection*{Goal of the CARMEMCO Project}

CARMEMCO grew out of need to provide a single engineering and manufacturing
problem that can integrated our research efforts in engineering and
manufacturing decision systems.  Our short term goal for CARMEMCO is to
define a company, existing solely in the computer,  to support the
exploration of integration issues in design, planning, scheduling, etc.  Our
long term goal is to bootstrap the implementation of an on-campus factory
which will serve as a development- and test-bed for research projects in
automation of engineering and manufacturing.

\subsection*{What knowledge would be useful to share?}

In the industrial domain, a better question is what knowledge should NOT be
shared.  In our experience, defining partitions and limiting access to
knowledge is a naive view of the degree of overlap in the use of knowledge
that exists within an industrial firm.  While partitioning does occur in
reality, it does so to the detriment of the organization.

\subsection*{What form should it take?}

What has been called "upper level" onotologies would of course be useful but
the sharing of the lower level more domain specific terms and instances is
also necessary.  The problem here is that different parts of the
organization do not use the same terms even when referring to the same
concept.  Any attempt to standardize terminology fails.  Secondly, enforcing
the use of the same terminology can lead to inefficient problem solving for
a particular function. Thus arises the distinction between the language of
communication and the language used by a function to reason.

\subsection*{What are the barriers to sharing?}

Within the industrial domain, significant efforts are underway to
standardize organization, product and process descriptions.  ICAM, CAMI,
CIM-OSA, PDES Inc. etc are representative of efforts to provide "upper
level" terms for manufacturing organizations.  Barriers arise when these
"reference" models are instantiated within a particular organization,
as mentioned above.  The focus on a language of communication between
organizational entities, like the PDES effort in design is not as well accepted
for production level functions.

\subsection*{How should the community cooperate?}

See the paper by Fox and Tenenbaum from the Santa Barbara meeting.
[Editor's note: we plan to bring copies of this paper to the Dunes