Files in this item

FilesDescriptionFormat

application/pdf

application/pdf9236571.pdf (4MB)Restricted to U of Illinois
(no description provided)PDF

Description

Title:Knowledge-based decision support system for scheduling in a flexible flow system
Author(s):Piramuthu, Selwyn
Doctoral Committee Chair(s):Shaw, Michael J.
Department / Program:Business Administration
Discipline:Business Administration
Degree Granting Institution:University of Illinois at Urbana-Champaign
Degree:Ph.D.
Genre:Dissertation
Subject(s):Business Administration, General
Artificial Intelligence
Computer Science
Abstract:Decision Support Systems (DSSs) are necessary resources for most complex decision making situations. The environment under which the DSSs function are, for the most part, evolving and, hence the DSSs should have the capability to adapt themselves as per the changes in the characteristics of the environment. An adaptive DSS that can function in a dynamic environment is proposed in this thesis, incorporating simulation modeling and inductive learning, to improve the overall performance of the system.
We develop an adaptive DSS incorporating learning in this thesis. We also attempt to refine the learned knowledge as is stored in the knowledge-base, thus reducing the effects of noise in the training examples on learning. The proposed framework is illustrated by scheduling a flexible flow system (FFS). Example of an application environment representing an FFS is the surface mount technology (SMT) facility used in printed circuit board (PCB) manufacturing. Scheduling in a PCB assembly facility involves decisions to be taken both at part-release and dispatching at machines stages. We develop a bi-level DSS to accomodate these interactions. The performance of the resulting system is shown to improve over systems using just one best heuristic for scheduling.
Issue Date:1992
Type:Text
Language:English
URI:http://hdl.handle.net/2142/20587
Rights Information:Copyright 1992 Piramuthu, Selwyn
Date Available in IDEALS:2011-05-07
Identifier in Online Catalog:AAI9236571
OCLC Identifier:(UMI)AAI9236571


This item appears in the following Collection(s)

Item Statistics