Defect monitoring in iron casting using residues of autoregressive models


  • Vanusa Andrea Casarin Universidade Regional Integrada do Alto Uruguai e das Missões - URI Campus de Santo Ângelo
  • Adriano Mendonça Souza Department of Statistics, Department of Industrial Engineering (UFSM), Santa Maria, RS
  • Jaime Alvares Spim School of Engineering, Technology Center (UFRGS), Porto Alegre, RS



The purpose of this study is to monitor the index of general waste irons forecasting nodular and gray using the residues originated from the methodology Box & Jenkins by means of X-bar and R control charts. Search is to find a general class of model ARIMA (p, d, q) but as data have autocorrelation is found to the number of residues which allowed the application of charts. The found model was the model SARIMA (0,1,1)(0,1,1. In step of checking the stability of the model was found that some comments are out of control due to temperature and chemical composition.





Artigos (Ativos de 2011 até 2014)