An approach of steel plates fault diagnosis in multiple classes decision making

Abstract

In the steel industry, specifically alloy steel, creating different defected product can impose a high cost for steel product manufacturer. This paper is focused on an intelligent multiple classes fault diagnosis in steel plates to help operational decision makers to organise an effective and efficient manufacturing production. Treebagger random forest, machine learning ensemble method, and support vector machine are proposed as multiple classifiers. The experimental results are further on compared with results in previous researches. Experimental results encourage further research in application intelligent fault diagnosis in steel plates decision support system. © 2014 Springer International Publishing.