Page 296 - Sensors and Control Systems in Manufacturing

254
Cha p te r
F i v e

that a language can be used at all levels of the DSS. In the
design and use of an LS for the user interface, one can con-
sider objectives such as ease of communication, the level of
complexity that can be presented by the LS, and the time
needed for the user to learn it.
• Knowledge systems. The basic function of a knowledge system
(KS) is the representation and organization of the “knowl-
edge” in the system. Two possible approaches are storing in
the information in database form or representing the data as
a base for artificial intelligence using methods from, for
example, predicate calculus. The objective of KS is to ease
accessibility of data for the DSS. The KS should be able to
organize and classify databases and problem domains accord-
ing to objectives that are sensible and convenient for the user.
Some of the objectives in the design of the KS are to reduce
the amount of computer memory required, increase the speed
with which the data can be retrieved or stored, and increase
the number of classifications of data and problem domains
possible.
• Problem processing systems. The problem processing system of
a DSS provides an interface between the LS and the KS. The
primary function is to receive the problem from the user via
the LS and use the knowledge and data from the KS to
determine a solution. Once a solution is found, the PPS sends
it through the KS to be translated into a form the user can
recognize. More importantly, in the model formulation,
analysis, and solution procedure of PPS, the conflicting
objectives of stated problems must be considered. The PPS
should provide methodology that can optimize all conflicting
objectives and generate a compromise solution acceptable to
the user. Some of the objectives in the development of such
multiobjective approaches are to reduce the amount of time
that the user must spend to solve the problem, increase the
degree of interactiveness (e.g., how many questions the user
should answer), reduce the difficulty of questions posed to
the user, and increase the robustness of the underlying
assumptions and procedures.

5.6 Analysis and Design of CIM with Sensors and
Control Systems
Many manufacturing systems are complex, and finding a place to
begin a system description is often difficult. Breaking down each
function of the system into its lowest possible level and specifying
objectives for each level and their interactions will be an effective