Identification of Control Chart Deviations and Their Assignable Causes Using Artificial Neural Networks

In case of complex processes, the identification of out-of-control states, observed on control charts, and their specific assignable causes are very complicated tasks. To overcome these difficulties artificial intelligence techniques have been used. Among these methods, the artificial neural networks can develop intelligence by using process data without need to expert opinion. This paper proposes an original method for process monitoring based on control chart exploration using artificial neural networks applicable for high batch size production requiring high sampling frequency. The developed approach helps to identify the out-of-control states and the corresponding process defects that lead to their occurrences. Attention is given to three most frequently observed cases in industrial practices: shifts, ascending and descending trends, and cyclic phenomena. The developed neural networks use back propagation algorithm and one hidden layer. A real industrial case of study was used to evaluate the recognition and identification performances of the developed artificial neural networks. Results have shown excellent recognition rates that reached percentages of identification of both process deviations and assignable causes higher than 90%.

[1]  Lifeng Xi,et al.  A neural network ensemble-based model for on-line monitoring and diagnosis of out-of-control signals in multivariate manufacturing processes , 2009, Expert Syst. Appl..

[2]  Daniel W. Apley,et al.  Posterior Distribution Charts: A Bayesian Approach for Graphically Exploring a Process Mean , 2012, Technometrics.

[3]  Francisco Aparisi,et al.  Interpreting the Out-of-Control Signals of Multivariate Control Charts Employing Neural Networks , 2010 .

[4]  Joseph C. Chen,et al.  Development of fuzzy logic-based statistical process control chart pattern recognition system , 2016 .

[5]  Douglas C. Montgomery,et al.  Statistical quality control : a modern introduction , 2009 .

[6]  William H. Woodall,et al.  Bridging the gap between theory and practice in basic statistical process monitoring , 2016 .

[7]  James R. Evans,et al.  A framework for expert system development in statistical quality control , 1988 .

[8]  A. S. Anagun.  A neural network applied to pattern recognition in statistical process control , 1998 .

[9]  Yi-Chih Hsieh,et al.  A neural network based model for abnormal pattern recognition of control charts , 1999 .

[10]  Anders L. Madsen,et al.  Applications of object-oriented Bayesian networks for condition monitoring, root cause analysis and decision support on operation of complex continuous processes , 2005, Comput. Chem. Eng..

[11]  Adel Alaeddini,et al.  Using Bayesian networks for root cause analysis in statistical process control , 2011, Expert Syst. Appl..

[12]  Duc Truong Pham,et al.  Control Chart Pattern Recognition Using Combinations of Multi-Layer Perceptrons and Learning-Vector-Quantization Neural Networks , 1993 .

[13]  Melike Erdogan,et al.  Analysis and control of variability by using fuzzy individual control charts , 2017, Appl. Soft Comput..

[14]  Hui Yang,et al.  Interpreting out-of-control signals using instance-based bayesian classifier in multivariate statistical process control , 2017, Commun. Stat. Simul. Comput..

[15]  Nasreddin Dhafr,et al.  Improvement of quality performance in manufacturing organizations by minimization of production defects , 2006 .

[16]  Abdessamad Kobi,et al.  A Bayesian network dealing with measurements and residuals for system monitoring , 2016 .

[17]  Stelios Psarakis,et al.  The use of neural networks in statistical process control charts , 2011, Qual. Reliab. Eng. Int..

[18]  J. D. T. Tannock,et al.  A review of neural networks for statistical process control , 1998, J. Intell. Manuf..