PolyU IR
 

PolyU Institutional Repository >
Industrial and Systems Engineering >
ISE Theses >

Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/3542

Title: Generic methodology for the design and development of an intelligent optimization system
Authors: Leung, Wai-kei
Subjects: Hong Kong Polytechnic University -- Dissertations
Expert systems (Computer science) -- Industrial applications
Fuzzy logic -- Industrial applications
Mathematical optimization
System analysis
Issue Date: 2004
Publisher: The Hong Kong Polytechnic University
Abstract: In today's competitive business environment, optimization can be applied to various industrial processes so that, through their improvement, a company can increase its competitive advantage, gain a better profit margin and sustain the continual growth of company business. The determination of optimal process parameters, so the process time can be shortened and the product quality can be increased, is one of the critical issues in process improvement. The traditional method for finding process parameters, based upon experience and trial-and-error practice, has become inadequate, especially when a large number of process parameters are involved in the process. This research aims at investigating the most appropriate approaches to optimizing the industrial process, and illustrates the effectiveness of the proposed intelligent optimization system through a case study with an industrial process in a local SME. In this research, various approaches of the system modeling and optimization techniques are reviewed. The potentials and limitations of individual approaches are discussed. Based on the integration of Fuzzy Logic and Genetic Algorithm with the combination of On-Line Analytical Processing technology, an intelligent optimization system, named Fuzzy-GOLAP, was designed and developed. Three modules are entailed in Fuzzy-GOLAP: the Data Management Module (DMM), the System Modeling Module (SMM) and the System Optimization Module (SOM). With the aid of relevant data collected and managed in DMM, the industrial process model can be formed in SMM and the optimal process parameter set can be found in SOM. The process result is stored in DMM ready for the next optimization, thus closing the loop of continual improvement. Thus, a roadmap for implementing Fuzzy-GOLAP has been constructed for all local SMEs. In order to verify the suggested methodology and roadmap, a case study in a local SME was applied with the help of neural network theory to validate Fuzzy-GOLAP. Results of the validation indicate that the system can generate an optimal set of process parameters which can meet the company defect acceptance level and lead to the production of good quality products in a shorter time compared with when the traditional method is used. Implementation of Fuzzy-GOALP has demonstrated that not only can the production department benefit from this project, but the operational level and the management level can also reap advantages from it. The proposed methodology can be generalized for other local SMEs to adopt the system and roadmap for process optimization with considerable benefits.
Degree: M.Phil., Dept. of Industrial & Systems Engineering, The Hong Kong Polytechnic University, 2004.
Description: xviii, 157, [50] leaves : ill. ; 30 cm.
PolyU Library Call No.: [THS] LG51 .H577M ISE 2004 Leung
Rights: All rights reserved.
Type: Thesis
URI: http://hdl.handle.net/10397/3542
Appears in Collections:ISE Theses
PolyU Electronic Theses

Files in This Item:

File Description SizeFormat
b17726645_ir.pdfFor All Users (Non-printable) 9.3 MBAdobe PDFView/Open
b17726645_link.htmFor PolyU Users 162 BHTMLView/Open



Facebook Facebook del.icio.us del.icio.us LinkedIn LinkedIn


All items in the PolyU Institutional Repository are protected by copyright, with all rights reserved, unless otherwise indicated.
No item in the PolyU IR may be reproduced for commercial or resale purposes.

 

© Pao Yue-kong Library, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong
Powered by DSpace (Version 1.5.2)  © MIT and HP
Feedback | Privacy Policy Statement | Copyright & Restrictions - Feedback