Browse Prior Art Database

Method to Automatically Calculate and Set Control Limits

IP.com Disclosure Number: IPCOM000124185D
Publication Date: 2005-Apr-11
Document File: 3 page(s) / 156K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a method to automate the statistical process control limit calculation and setup. Benefits include integration of data stability check into the control limit calculation.

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 54% of the total text.

Method to Automatically Calculate and Set Control Limits

Disclosed is a method to automate the statistical process control limit calculation and setup. Benefits include integration of data stability check into the control limit calculation.

Background

Currently, calculating and implementing control limits is a manual process, fully dependent on an engineer’s experience and expertise. There is no clear procedure defined which leads to the implementation of many different approaches for control limit calculation. This can cause inefficient limits to be implemented that can jeopardize the effectiveness of the statistical
process control.

General Description

The disclosed method automatically calculates the control limits, while ensuring that the limits calculated have a statistically valid sample size, are calculated from a  stable dataset, will drive for process continuous improvement and will not jeopardize production due to a high out of control rate. The following are the steps for the disclosed method:

§         Sample Size verification. A minimum sample size is required to have valid control limits.  If this sample size cannot be achieved, a warning message is given and limits are not calculated.

§         Stability Check. The dataset used to calculate the control limits is stable. If data is found to be unstable, a warning message is given and limits are not calculated.

§         Comparison of Control Limits with Spec Limits. The control limits are less than the spec limits. This step can be modified so that it ensures that a minimum Cpk is needed in order to implement the control limits.

§         Drift Detection in Control Limits. No slow process degradation is taking place. Slow process degradation can be seen as a slow increase of a UCL or decrease of a LCL. 

§         Trigger Rate calculation. New control limits are validated to ensure the trigger rate (quantity of points out of contro...