|
QICID: 18214
Title: Using Factor Effects Analysis to Improve Statistical Process Control
Copyright: 2002, ASQ and Marcel Dekker, Inc.
Author: Flaig, John J.
Organization: Applied Technology, San Jose, CA (johnflaig@yahoo.com)
Subject: Cause and effect diagram,Design of experiments (DOE),Multivariate control charts,Statistical process control (SPC);
Series: Quality Engineering, Vol. 14, No. 4, June 2002, pp. 673-678
Abstract: [This abstract is based on the author's abstract.] The value of factor effects analysis on improving the diagnostic capability of the practitioner is significant, but it relies on the adequacy of the model. Quality practitioners know they are to take action when a control chart shows an out-of-control condition, but they may not know what action is appropriate. Several methods can be employed to generate the process model require in multivariate factor effects analysis. The best approach may be an amalgam of several techniques. A model based on design of experiments (DOE) helps develop a better understanding of the process and is valuable for both process control and improvement. A model involving the interpretation of physical laws governing the behavior of a system or deductive reasoning is frequently used because of its speed, simplicity, and cost. Yet another approach is to build a model based on analysis of observed process data. Despite the inherent risk of unknown and uncontrolled variables, this methodology has proved valuable in prescreening variables for more formal DOE. As Dr. Box noted, "All models are wrong, but some are useful." A useful model can add substantially to the practitioner's diagnostic tool kit.
Number of pages: 6
Price for ASQ Members: $5.00
Price for List/Forum/Division: $10.00
All electronic articles are sent as PDFs via e-mail. To view the documents, you will need Adobe
Reader (free download).
Orders placed during business hours are usually filled within one business day.
If you have questions please e-mail our Customer Care center at help@asq.org.
Browse QIC articles chronologically
previous next
|
