| บทคัดย่อ(English) |
Downsizing the corporations computer systems is one interesting practice of organizations. The feasibility of downsizing and breakthroughs in computer technology reinforce this concept, and cause corporations to downsize for cost reduction and to increase the efficiency and effectiveness of business operations. However, this has both benefits and drawbacks. Many processes and related complex factors need to be considered. Furthermore, there is no tool to help make decisions on such issues. The product of this research, ~i"A Rule-Based System for Downsizing the Corporations Computer Systems (DOWNSIZINGX)"~i, is a tool that helps in the decision-making of downsizing computer systems. The architecture of DOWNSIZINGX is similar to a knowledge based system but uses a rule-based system. The information about downsizing is extracted and collected from many sources. A structured situation diagram and dependency diagram was designed that divided the downsizing into 14 major modules. A decision table analysis was used to analyze the results. About 2000 if-then rules and sets of questions, answers, and recommendations were developed and a knowledge base was constructed. Another component of DOWNSIZNGX is CLIPS, an expert system shell. It makes inferences about knowledge by using the forward chaining and depth first strategy as a conflict resolution technique. In addition, the user interacts with the system through Visual Basic Interface that is connected to CLIPS via CLIPS OCX. Furthermore, the system also has an explanation facility that allows the user to trace the recommendation results back to find what factors were involved in making them and to assist effective decision-making with reduced risks. The prototype was developed to work on Windows 98, and was implemented by Visual Basic programming using Microsoft Access as a database management. It was tested by simulating rules and comparing results with CLIPS Windows Interface. The simulated business decision-making situations were tested for all functions. The prototype was used to recommend some business decision scenarios and case studies. The results show that the system performs all functions correctly and also recommends situations effectively. Finally, it can give information that helps support the decision to downsize the corporations computer systems. The performance of the system is measured by the running time of the prototype by arbitrarily selecting the module and randomly answering the questions. We found that each modules performance depended upon the steps of inference and number of rules. It was satisfactory and made the overall performance of the system acceptable. |
