공정개선을 위한 대용량 데이터 분석 및 데이터마이닝 적용사례 연구

Abstract

Due to the recent development of information technology, a large amount of process data may be collected directly from shop-floor in real-time. For the purpose of process improvement, it may be useful to utilize the information derived from those data since it may relieve the burden of additional data collection. However, it is quite difficult to extract meaningful information from those massive data mainly because of the presence of disturbing noise factors. Engineering knowledge and technical experiences may be useful to identify those noise factors and analyze the data effectively. In addition, a variety of data mining techniques, including clustering and decision tree analysis, may be helpful to better understand the nature and characteristics of processes. The main purpose of this thesis is to outline analysis procedures for process improvement utilizing engineering knowledge combined with data-mining techniques. The usefulness and applicability of proposed procedures are demonstrated through the case study from a semi-conductor manufacturing process. More specifically, clustering and decision tree analysis are employed to analyze the process data.