RT Journal Article
SR Electronic
T1 Effect of Resolution of Measurements in the Behavior of Exponentially Weighted Moving Average Control Charts
JF PDA Journal of Pharmaceutical Science and Technology
JO PDA J Pharm Sci Technol
FD Parenteral Drug Association (PDA)
SP 288
OP 295
DO 10.5731/pdajpst.2013.00921
VO 67
IS 3
A1 Nora Meneces
A1 Silvino A. Olivera
A1 Carlos D. Saccone
A1 Julio Tessore
YR 2013
UL http://journal.pda.org/content/67/3/288.abstract
AB Process behavior is known through measurement. All measurements are discrete and transform a continuous process distribution into a discrete (multinomial) measurement distribution. In this article, properties of exponentially weighted moving average control charts (EWMA charts) of measurements are studied through Monte Carlo simulations. Detection capabilities (average run length curves) are presented as a function of measurement resolution and recommendations for proper design of a measuring system are given. EWMA charts measurement resolution requirements are compared to Shewhart charts, and two examples are presented to illustrate behavior differences. LAY ABSTRACT: Industry performs quality control activities in order to control raw materials, processes, and finished products with the purpose of delivering products to market complying with quality requirements. Different statistical methods are used in chemical, physical, and biological analysis and nearly all control methods will use measurement devices. Quality control is mandatory for many industries, for example, the pharmaceutical industry, and it is a very costly activity. The selection of suitable methods of control and appropriate measurement devices acquire special relevance for obtaining efficient results. Measurement devices require a certain measurement capability and must be fit for the purpose of controlling the quality of processes and products. The equipment capability includes, among other elements, the resolution, which means the “smallest change in a quantity being measured that causes a perceptible change in the corresponding indication”. Different control charts have been extensively used for statistical process control. Based on cases from the pharmaceutical industry and simulation tools, this paper studies the influence of resolution of measurement devices in exponentially weighted moving average control charts (EWMA charts). In addition, a comparison of EWMA charts and traditional Shewhart charts is presented in order to offer support to the practitioners.