Multivariate statistical process control of platinum: a case of mining company in Shabani, Zimbabwe

Mawonike, Romeo (2013) Multivariate statistical process control of platinum: a case of mining company in Shabani, Zimbabwe.

Full text not available from this repository.
Official URL: http://ir.nust.ac.zw/xmlui/handle/123456789/413

Abstract

Application of statistical methods in monitoring and control of industrial processes are generally known as statistical process control (SPC). Since most of the modern day industrial processes are multivariate in nature, multivariate statistical process control (MSPC), supplanted univariate SPC techniques. MSPC techniques are not only significant for scholastic pursuit; it has been addressing industrial problems in recent past. Monitoring and controlling a chemical process is a challenging task because of their multivariate, highly correlated and non-linear nature. In this paper, a series of techniques were applied. Time series plot was implemented to determine the stationarity of the data. The Box-Jenkins methodology of model identification, estimation and validation; was used to generate ARIMA models based on multiple non sequential data. As a result, the residuals from ARIMA models have shown four attributes: normally distributed, uncorrelated, independent and no autocorrelation between successive time points. Two MSPC techniques; Multivariate Cumulative Sum (MCUSUM) and Multivariate Exponentially Weighted Moving Average (MEWMA) were implemented as control charts for monitoring residuals. All the charts indicate the out of control signals in the process, which were believed to be from one or more variables combined together. The problem of which variable is causing the out-of-control process and when is that out-of-control happening was alleviated through the construction of individual Cumulative Sum (CUSUM) control charts. Elimination of out-of-control signals resulted in a successful in control process shown in both MCUSUM and MEWMA charts. Comparison between these two multivariate charts shows that MCUSUM is more powerful in detecting smaller shifts than the MEWMA chart. Therefore, monitoring of residuals provided a valuable proof-of-concept that validated the use of time series analysis in conjunction with MSPC tools in modeling and monitoring the behaviour of industrial processes.

Item Type: Article
Uncontrolled Keywords: Multivariate Statistical Process Control,Autoregressive Moving Average,Industrial processes,Mean shift
Divisions: Universities > State Universities > National University of Science and Technology
Depositing User: Mr. Edmore Sibanda
Date Deposited: 01 Dec 2015 02:32
Last Modified: 01 Dec 2015 02:32
URI: http://researchdatabase.ac.zw/id/eprint/1316

Actions (login required)

View Item View Item