Back to results list
Please use this identifier to cite or link to this item:
|Title:||Development of a cooperative distributed process mining system for continual quality enhancement||Authors:||Ho, To-sum||Keywords:||Hong Kong Polytechnic University -- Dissertations
Total quality management
Management information systems
|Issue Date:||2007||Publisher:||The Hong Kong Polytechnic University||Abstract:||In today's competitive and unpredictable business environment, manufacturers face the challenge of demanding customers who strongly seek high-quality and low-cost products relevant to their specific needs. High-quality and high-reliability products play an important role in achieving customer satisfaction, and insisting on quality is always the only way for an enterprise to survive. In fact, to achieve high quality is not the responsibility of any one person or functional area; it is everyone's duty in the entire corporation. Poor process decisions from any individual may lead to poor customer satisfaction. The ultimate goal is to achieve better collaboration for making right decisions all the time in every process involved. Although numerous empirical and scientific approaches have been developed in the field of quality management, past research has not addressed this issue well enough, nor has actual practice managed to optimize the integrated workflow in order to make sure that all the participants have the possibility to act successfully in their processes. Traditionally, various functional disciplines have had their own information systems for quality control and monitoring in their own specific process. However, the fact that quality improvement is a distributed and cooperative problem-solving activity has been neglected. Therefore, attention should be paid to capturing the distributed process data to support knowledge discovery within the workflow of the enterprise. In order to facilitate continual quality enhancement, an intelligent system called the Cooperative Distributed Process Mining System (CDPMS) has been developed. It is equipped with the "distributed process mining" feature to discover the hidden relationships among all the process variables involved in a distributed and automatic manner. The proposed system consists of three modules: the Process Data Exchange Module (PDEM), the Backward Quality Tracking Module (BQTM) and the Forward Quality Optimization Module (FQOM). PDEM introduces a dynamic process data schema, thus enabling data interchange to take place among various departments of the enterprise. BQTM finds significant patterns in term of fuzzy association rules among centralized process data. The presence of process variables among these patterns in the workflow tends to imply the presence of quality problems. FQOM determines an optimal setup or configuration of process parameters within the integrated workflow. The new features of BQTM and FQOM are characterized with the incorporation of the newly designed iterative Bi-directional Process Mining (i-BPM) algorithm through the capabilities of the fuzzy-set concepts, association rule and genetic algorithm. These extract implicit generalized knowledge from distributed process data stored as quantitative values. i-BPM algorithm is also used to formulate nearly optimal sets of fuzzy rules to identify the possible solutions for defect reduction from the initial generalized fuzzy association rules. Process Quality Markup Language (PQML), which evolved from extensible Markup Language (XML) is also proposed to fill in the gap in the development of an expandable and flexible language for ensuring that relevant process information is quickly and accurately communicated throughout the enterprise.
To validate the feasibility of the proposed schema, two case studies have been conducted in two local companies using the approach suggested above. Furthermore, a generic methodology related to the design and implementation of the proposed system has been described. The proposed methodology is a step in the development of a generic model for effective communication with departments in order to achieve better collaboration and correct decisions. The major contribution of this research is to implement a system infrastructure that allows an analysis to be performed on widely distributed process data according to the corporate objectives and to facilitate correct decision making on diverse elements of business strategies. Besides, the deliverables of this research not only provide a means for developing CDPMS, but also open the door for incorporating process management in future quality information systems, enabling the operations staff to focus on distributed and cooperative process management.
|Description:||xix, 207,  leaves : ill. ; 30 cm.
PolyU Library Call No.: [THS] LG51 .H577P ISE 2007 Ho
|URI:||http://hdl.handle.net/10397/3133||Rights:||All rights reserved.|
|Appears in Collections:||Thesis|
Show full item record
Files in This Item:
|b21167436_link.htm||For PolyU Users||162 B||HTML||View/Open|
|b21167436_ir.pdf||For All Users (Non-printable)||5.37 MB||Adobe PDF||View/Open|
Citations as of Mar 18, 2018
Citations as of Mar 18, 2018
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.