Knowledge-based systems are computer-based programs that
incorporate human expertise and other documented knowledge with the facilities
for applying that knowledge to real-world circumstances. Knowledge-based systems
provide the benefit of and satisfy the requirement for documenting, developing,
and disseminating rules, processes, and/or guidance related to a specific domain
or problem area. Knowledge-based systems may be automated in embedded systems or
employed through a "user interface" where questions can be presented in a manner
similar to how they would be asked of a human consultant or expert.
Knowledge-based systems, sometimes called "expert
systems," are made up of two major components: the knowledge base and the
inference engine. The knowledge base is the repository of the knowledge, which,
in some systems, is expressed as a collection of facts together with related
"if, then" rules. The inference engine interprets and manipulates the combined
facts and rules in the knowledge base to arrive at the answer to a question.
Other components of a knowledge-based system include a knowledge acquisition
subsystem and a user interface. The knowledge acquisition subsystem facilitates
generation of the knowledge base. This process involves collecting information
from various sources including the human "experts" and translating this
information into facts and rules in the language of the knowledge base. This
process has been referred to as "knowledge engineering." The user interface
provides a mechanism for the effective exchange of problem-related information
between the end-user and the computer system.
To a large degree, knowledge-based systems are used to
extend and apply the expertise or documented knowledge of an acquired discipline
to areas where it would not be efficient, practical, or even possible using a
non-automated process. They are widely used by decision makers for strategic
planning and for identifying areas for improving productivity and process
quality. A knowledge-based system may also be used in applications associated
with automatic control or process monitoring. Systems used as expert assistants
are queried on an ad hoc basis whenever the knowledge of an expert is required
for satisfactory execution.
The decision to employ a knowledge-based system starts
with selection of the system itself. In some cases, it may be possible to
acquire a commercial-off-the-shelf system that is task-specific or
solution-specific to the target application. This approach might be common in
applications involving financial planning and medical diagnosis, where an
initial knowledge base exists. However, for applications related to leading edge
manufacturing, it is more likely that the system would need to be tailored to
meet the peculiarities and anomalies of the related processes. In this case, a
knowledge-based system shell would be used. This shell is a software development
environment containing generic system components for building the application
specific system. With the shell's knowledge acquisition subsystem, the knowledge
base and the inference engine are configured and instantiated using the
collection of knowledge and reasoning provided from representative experts and
other available facts pertinent to the processes.
A knowledge-based system can be used by an organization
to acquire, document, develop, and disseminate enterprise-wide design guidelines
and process capabilities. Using a knowledge acquisition system, expert knowledge
and opinions as well as lessons learned related to design practices and
processes can be captured to formulate the knowledge base. Engineers and
designers can then access the knowledge-based system when necessary to obtain
design guidelines and other information related to specific tasks at hand.
Managers may access the knowledge-based system to obtain information on current
process capabilities or to identify opportunities for process improvement. A key
benefit of employing knowledge-based systems is that consistency of the
information provided reduces error rates and improves the overall quality of an
organization's designs and processes.
Durkin, J., & Durkin, J. (1994). Expert Systems: Design and Development. MacMillan Collegiate Division.
Gonzalez, A. J., & Dankel, D. D. (1993). The Engineering of Knowledge-Based Systems: Theory and Practice. Englewood Cliffs, NJ: Prentice-Hall, Inc.
Jackson, P. (1999). Introduction to Expert Systems (International Computer Science Series). Addison-Wesley Publishing Company.
Turban, E., & Aronson, J. E. (1998). Decision Support Systems and Intelligent Systems. Prentice-Hall, Inc.