Data Mining has been successfully applied to many practical marketing problems. Some examples are credit risk analysis for banks, attrition scoring for financial institutions, fraud detection, target marketing, customer profiling, warranty analysis, and so on. However, this thesis attempts to apply Data Mining into problems related to manufacturing factories, and solve them. Since most of the companies deal with the data as confidential, the data that is analyzed using Data Mining is generated from a simulation model which is a hypothetical job-shop manufacturing company. Even though there are many problems that occur in manufacturing factories, the follow three problems related to the hypothetical job-shop company are chosen: (1) What is the most important machine in terms of increasing profit? (2) What is the best batch size in terms of increasing profit? (3) What is the best product mix in terms of increasing profit? The other purpose of this thesis is to prove that the results from the data collected by a more advanced management philosophy is better than the data collected by a traditional management philosophy. (Abstract shortened by UMI.)
展开▼