Abstract: Quality of hot metal produced in a blast furnace is affected by multiple variables. Classical Statistical Process Control (SPC) methodologies are non-optimal to monitor and control these multiple variables as the effect of one variable can be confounded with effects of other correlated variables. Further, Univariate control charts are difficult to manage and analyze because of the large numbers of control charts of each process variable. An lternative approach is to construct a single multivariate T2 control chart that minimizes the occurrence of false process alarms. Multivariate control charts monitor the relationship between the variables and identifies real process changes which are not detectable with Univariate charts. This paper studies the application of Multivariate Statistical Process Control (MSPC) charts to monitor hot metal production process in a steel industry. T2 diagnosis with Principal component analysis (PCA) is applied to analyze the critical process variables.
Keywords: Control Chart, Regression Analysis, Statistical Process Control, Univariate, Multivariate, Principal Component Analysis, Correlation
Recieved: 23.12.2012 Accepted: 4.11.2013 UDC: 65.012.7