Proceedings of the 2016 International Conference on Industrial Engineering and Operations Management
Kuala Lumpur, Malaysia, March 8-10, 2016
Using several intelligences for complex industrial process
monitoring :detection and diagnosis
Nafissa Rezki, Leila Hayet Mouss, Djamil Rezki
LAP Laboratory Industrial Engineering Department
Batna University
Batna, Algeria
Nafissa_rezki@yahoo.fr , hayet_mouss@yahoo.fr , drezki@yahoo.fr
Okba Kazar , Laid Kahloul
LINFI Laboratory
Biskra University
Biskra, Algeria
kazarokba@yahoo.fr , kahloul2006@yahoo.fr
Abstract—The objective of the current paper is to present an intelligent system for complex process monitoring, based on
artificial intelligence technologies. This system aims to realise with success some complex process monitoring tasks that are:
detection, diagnosis. For this purpose, the development of a multi-agent system that combines multiple intelligences such as:
multivariate control charts, neural networks, has became a necessary. The proposed system is evaluated in the monitoring of the
complex process Tennessee Eastman Process.
Keywords—multivariate process; Hotelling T
2
control chart; multi-agent system; neural network
I. INTRODUCTION
Currently, the manufacturing processes become more and more complex and multivariate. In these systems, the operator
recuperates a vast data amount to be analysed. The high volume of data and the big number of process variables make the
operator task fastidious. To avoid such problems, the data based methods are more suitable for the process monitoring. The
multivariate control charts (Hotelling T
2
control chart, Multivariate CUSUM (MCUSUM), Multivariate EWMA (MEWMA))
have been used for the control of multivariate process and have proved their adequacy to reduce the complexity of such
process monitoring. Moreover, the monitoring of a multivariate process is a complex task, it can be devised into four subtasks
which are: the detection of abnormal situation, the diagnosis of the faults, the identification of variables that involved in the
faults, and finally the reconfiguration of the process [1].
Many researchers have used the control charts for process monitoring ([2],[3]). In this paper, we regroup the tasks of the
multivariate process monitoring in one approach. Our contribution is to determine the best combination of multivariate control
charts and neural networks. The result of this research is a multi agent system that applied to a multivariate process monitoring.
This multi-agent system use: multivariate control chart for abnormal detection, neural network for faults diagnosis.
The rest of this paper is organized as follows: section 2 explains the necessary backgrounds. The process-monitoring
approach is presented in Section 3 with the monitoring algorithm. In section 4, a case study of simulated Tennessee Eastman
Process is employed to illustrate the validity of the proposed approach, including the detection by MCCEA (Multivariate
Control Charts Executor Agent), diagnosis by DANNA (Diagnosis Artificial Neural Network Agent). Finally, conclusions
and future works are suggested.
II. BASIC CONCEPTS
A. Hotelling T
2
control chart
The Hotelling T
2
control chart is one of the most used charts for multivariate process monitoring. This chart gives good
results in the detection of deviation in the mean vector values of a studied process. In general, for a process with p variables,
the T
2
is given by “(1)”:
T
2
=n(x-μ)
T
Σ
-1
(x-μ) (1)
3351 © IEOM Society International