EXPERT SYSTEMS AND THE USE OF INFORMATION IN BUILDING DESIGN AND CONSTRUCTION
| Published date | 01 February 1988 |
| Pages | 91-118 |
| Date | 01 February 1988 |
| DOI | https://doi.org/10.1108/eb026820 |
| Author | COLIN H. DAVIDSON,PHILIPPE L. DAVIDSON,KALEV RUBERG |
| Subject Matter | Information & knowledge management,Library & information science |
THE
Journal of Documentation
VOLUME 44 NUMBER 2 JUNE 1988
EXPERT SYSTEMS AND THE USE OF INFORMATION
IN BUILDING DESIGN AND CONSTRUCTION
COLIN
H.
DAVIDSON
School
of
Architecture,
University
of
Montréal,
Montréal,
Québec,
Canada,
H3C
3J7
(to
whom all correspondence should be addressed)
PHILIPPE
L.
DAVIDSON
Information Technology
Section,
Laboratory for
Intelligent
Systems,
Division
of
Electrical
Engineering,
National Research Council
of Canada
KALEV RUBERG
Institute for
Research
in
Construction,
National Research Council
of Canada
The building industry, through its structure and its mandate, faces endemic
information
problems;
expert
systems
are expected to impact
positively.
Expert
systems are suited to situations of uncertainty; knowledge and reasoning are
separated, allowing easier updating. Knowledge acquisition from human
experts
is
difficult and problems of information reliability arise, suggesting the
scope for cooperation between knowledge engineers and documentalists
familiar with the domain. In building, prevailing conditions seem to indicate
the appropriateness of expert systems, particularly during the design phase;
however, written documentation and general research results are rarely
consulted. This highlights the need for an information 'refining' stage between
production and use. It is easier to set up expert systems for specialised sub-
domains;
however,
on-going research
is
attempting
to develop
a comprehensive
approach to project-specific information that would
be
operational from initial
design through
to
completed construction. Criteria for a comprehensive design
information system can be listed.
Journal
of
Documentation,
Vol. 44, No.
2,
June
1988,
pp. 91-118.
91
JOURNAL OF DOCUMENTATION Vol. 44, no. 2
INTRODUCTION
FROM AN INFORMATION SCIENCE POINT OF VIEW, the building
industry is a particularly interesting applications
field
for two reasons: (i) the
nature of
its
organisational structure, and (ii) the nature of
its
mission.
Organisation.
The building industry, in management jargon, is a 'multi-
industry', and each building project is undertaken by a 'temporary multi-
organization' [1]. The industry as a whole consists of a large number of
enterprises, both professional and consultant practices, and manufacturing
and construction companies. Each of them exists over a long period of time,
but must form a team with others for short periods to participate in particular
building
projects.
Long-term survival depends on a proper sequence of short-
term activities. As a result, each firm has its own long-term
modus operandi
and its own
ways
of ensuring its presence on the market
place;
it develops and
maintains some form of in-house information system, if only to record its
acquired experience. Also, each short-term project team must develop
effective coordination,
by
contract and
by
inducement, so that the
firms
called
upon to work together (and who may never have worked together before),
produce
the
required building within
the
imposed
constraints;
within
the
short
time span, a project-specific information system must
be
designed and utilised
by all participants.
Mission. The tasks of designing and constructing a building involve a
complex sequence of operations as the project moves from an expression of
intent to the reality of 'bricks and mortar'. Objectives and constraints are first
determined explicitly by the intending building owner, and are translated into
a functional and technical programme of requirements by specialist con-
sultants; these requirements are transformed by other professionals into
designs and details, accompanied by contract documents which, in turn,
become instructions for the manufacturers and contractors who actually
organise and execute the on- and off-site building operations. Finally, the
building is handed over to the owner, who must make it work effectively to
serve his
purpose.
All along
this
process,
interlinking decisions
are made
under
conditions of
interdependence
and
uncertainty [2]
- where the common thread
is the project information system, built up each time from scratch. The initial
inputs for this information system come from two unrelated sourses
(unrelated
to
each other and -
because each
building project
is
different- often
unrelated to the particular task at hand), namely: (a) the accumulated
knowledge and know-how of each participant, and (b) the sum of general
knowledge ('non-project-specific' information) vested in the various scientific
and technical libraries and reference systems. As we will see, both these
sources have some inherent limitations.
In this context of organisational and operational complexity, it is hardly
surprising that the control of information flow is seen to hold the key to
92
To continue reading
Request your trial