Please use this identifier to cite or link to this item:
Title: A knowledge-based performance measurement system for production planning and machine selection
Authors: Lam, Chau-yi Annie
Degree: M.Phil.
Issue Date: 2010
Abstract: In today's dynamic and competitive environment, there is intense pressure for manufacturing companies to face external and internal changes and uncertainties, including demand fluctuations, variations in customer requirements, decreases in product life cycles, and machine breakdowns. To react to these unpredictable changes, manufacturing companies must maintain a higher flexibility in their manufacturing system through the effective use of production assets. This, however, cannot be done without physical asset management (PAM), which plays an increasingly important role in achieving maximum lifetime effectiveness, utilization and return from physical assets, such as production machines, shop floors, and operating equipment. In the last decade, majority of researchers viewed PAM only from the maintenance perspective; they suggested various strategies and fault diagnosis tools to schedule maintenance activities and monitor machine health condition so as to reduce machine interruption. Although machine breakdowns are reduced after the provision of maintenance, it still does not guarantee that the machine can function effectively during its lifetime. To reflect the effective use of machine and increase its flexibility, it is necessary that the performance of physical asset operations be measured and evaluated. On the other hand, because unpredictable changes, such as customer requirements and machine breakdowns, often occur in the actual production shop floor, companies should promptly make complex decisions on manufacturing planning, including machine selection, process flow planning, production scheduling, and resource allocation. Often, such decisions rely on production supervisors' knowledge and experience. Thus, the design of a knowledge-based performance measurement approach is formulated to support performance evaluation of the utility of physical assets, as well as the process of decision making pertaining to manufacture planning. This thesis proposes a knowledge-based performance measurement system (KPMS) to support the decision-making processes of production planning and machine selection. The front-end module of KPMS uses data collection technologies such as radio frequency identification (RFID) and sensors to capture real-time manufacturing data from shop floors. The core module of KPMS involves a knowledge-based engine based on artificial intelligence (AI) techniques to derive useful knowledge from manufacturing operations data and performance indicators for decision support on production planning and resource allocation. In addition, a machine flexibility assessment function is included in the core module of KPMS to support the decision on machine selection. To validate the feasibility of using KPMS in providing reliable decision support for manufacturers in production planning and machine flexibility assessment, an industrial application case study was conducted in a helmet manufacturing company. The study found that with KPMS, the efficiency and reliability of manufacturing operations greatly improved and the firm’s flexibility in dealing with changes was enhanced significantly.
Subjects: Hong Kong Polytechnic University -- Dissertations
Production planning
Manufacturing processes -- Equipment and supplies
Pages: xii, 145 leaves : ill. (some col.) ; 30 cm.
Appears in Collections:Thesis

Show full item record

Page views

Last Week
Last month
Citations as of May 28, 2023

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.