The main point of continuous improvement program like TQM is acquire knowledge needed to improvement from the organizations and apply those to the production system. However, growing complexity and massive data make it more difficult to get those knowledge by use of existing process improvement technique. The advances in information technologies make it possible to extract useful knowledge from the enormous data through the use of Artificial Intelligence, especially Expert System.
In this study, the feasibility of applying Expert System to the one of fine ceramic manufacturing process was investigated experimentally. Inductive Learning tool(C4.5) was applied to finding rules for the optimal process condition. Artificial Neural Network(BPN5) was used to predict unknown results.
C4.5 showed the priority of the process variables and conditions of them through the decision tree to get specific result. BPN5 showed fairly good prediction rate of results under given conditions. The error rate and prediction rates were thought to be affected by the appropriate selection of the variables.
