Statistical Process Control Monitoring Research Articles

Model Based Statistical Process Control for Continuous Systems

Statistical process control (SPC) is a set of methodologies for signaling the presence of undesired sources of variation in manufacturing processes. SPC methods for continuous processes may be developed by using stochastic models which do not assume that successive observations are independent. A method for applying SPC to continuous processes is presented. This method incorporates a computationally efficient procedure for the on-line identification and estimation of autoregressive with exogenous inputs (ARX) models. Two examples illustrating the method for SPC monitoring are presented.

JCPDS—International Centre for Diffraction Data statistical process control method development

A task group of the JCPDS—International Centre for Diffraction Data (ICDD) was established for the purpose of investigating a methodology which would be applicable for statistical process control monitoring of X-ray powder diffractometers. A procedure for collecting X-ray diffraction data for statistical process control purposes and the incorporation of these data into control charts are presented. The results of this task group show that, through the use of statistical process control methods, noncontrol situations for diffractometers were detected, the causes of these problems were identified, and these problems were corrected and noted on control charts.

Algorithmic Statistical Process Control: Concepts and an Application

The goal of algorithmic statistical process control is to reduce predictable quality variations using feedback and feedforward techniques and then monitor the complete system to detect and remove unexpected root causes of variation. This methodology seeks to exploit the strengths of both automatic control and statistical process control (SPC), two fields that have developed in relative isolation from one another. Recent experience with the control and monitoring of intrinsic viscosity from a particular General Electric polymerization process has led to a better understanding of how SPC and feedback control can be united into a single system. Building on past work by MacGregor, Box, Astrom, and others, the article covers the application from statistical identification and modeling to implementing feedback control and final SPC monitoring. Operational and technical issues that arose are examined, and a general approach is outlined.

SPC Analysis of Optimal Strategies for Restandardization of X-ray Fluorescence Analyses

AbstractQuantitative XRF analyses almost always depend on measurements of x-ray emission intensities of standards of known composition. The usual approach is a careful initial standardization with multiple standards lhat are checked for accuracy, stability, homogeneity, and internal consistency. This is followed by limited periodic restandardization or recalibration with one or two of the initial standards or a reference sample. The frequency and protocol of these periodic restandardizations tend to be rather arbitrary; for example adjustment on the basis of the average of a triplicate measurement of a standard at the beginning of the work day. There are certainly a large number of other possible restandardization strategies. They range all they way from no restandardization at all, through single or n replicate restandardization measurements before analysis of each group of M unknowns, to cumulative or moving averages, again with variable replicate size, frequency, and now averaging range. This paper examines the outcomes of some of these restandardization strategies applied to synthesized, realistic XRF analysis data which incorporate both abrupt and gradual changes in instrument response. SPC (Statistical Process Control) control charts are shown to be a very useful tool for determining optimum restandardization strategies. Optimal strategies (least scatter in calculated concentrations with fewest standards) which emerged from this study included not restandardizing at all except when indicated by an SPC chart of a reference standard intensity run concurrently with each unknown, and a moving range of 3, with standards run every 5 unknowns, coupled with SPC monitoring of the standards used for this restandardization.