Annop Lerttanaphiboon. A maintenance management system with rule-based expert system for machine fault diagnosis. Master's Degree(Technology of Information System Management ). Mahidol University. Mahidol University Library and Knowledge Center. : Mahidol University, 2009.
A maintenance management system with rule-based expert system for machine fault diagnosis
Abstract:
Nowadays, Thailand has become one of the largest electronics
manufacturing industries in the world. Electronic parts must be processed by a large
number of machines; therefore, machinery is important to the manufacturing industry.
Thus, machine maintenance is vital for the productivity and safety of operators, as
well as for the quality of products.
The objective of this thesis was to analyze, design, and develop a
maintenance management system which includes a prototype expert system for
machine fault diagnosis. The system consists of three modules: 1. A machine
downtime management module that has two functions; machine downtime monitoring
and analytical reports. 2. A PM (preventive maintenance) management module that
has three functions; equipment master list manipulation, PM plan and work order
management, and PM weekly recall generator and confirmation. 3. An expert system
module that has two functions; machine fault diagnosis and preventive maintenance
instruction. The knowledge for the expert system was acquired from domain experts
and machine manuals, analyzed into decision tree structure, and then represented as
rules. For the inference engine, forward chaining was used as a search strategy.
The system was tested and evaluated by engineers, technicians, and
machine operators and found to be satisfactory by all users. It can help technicians
reduce work in preventive maintenance processes by a total time of 1,180 minutes per
month to 128 minutes per month. Therefore, this research achieved its goal. Lastly,
this research is a prototype that can be applied to similar manufacturing industries.