Browse Prior Art Database

Method for Devising and Assessing Process Control Information-Stream Sampling Frequencies

IP.com Disclosure Number: IPCOM000132464D
Publication Date: 2005-Dec-17
Document File: 4 page(s) / 113K

Publishing Venue

The IP.com Prior Art Database

Related People

Kent Franklin: AUTHOR [+3]

Abstract

We propose an improved system that integrates a method to determine the most efficient sampling frequencies with a process control and monitoring system. This includes providing a method for assessing the effectiveness of the current sampling frequency to detect changes in the process. It also includes a method to recommend the most efficient sampling frequency to detect changes in the process. The methods are applicable to both stable and unstable processes. In addition, this method can help process engineers determine the correct sampling frequency to accurately assess the impact of process improvements. This concept is applicable to any manufacturing operation which utilizes a process control system. The information stream into the system may be collected manually or automatically. This system can be added to existing process control software or added as an independent program. The proposed tool can be used to determine the most effective sampling frequency to reliably detect changes in a data stream from a process. Since the system is integrated with the on-line process control system, this information is available in real time.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 46% of the total text.

Page 1 of 4

Method for Devising and Assessing Process Control Information-Stream Sampling Frequencies

Kent Franklin, Kevin McCormick, and Glenn Wire Kimberly-Clark Corporation, Neenah, Wisconsin

Introduction

In manufacturing operations, process control decisions are based in part on a stream of data coming from process and product inspections. Since these inspections come at a cost, there is pressure to minimize the sampling frequency to reduce the resources allocated to maintaining these information streams. On the other hand, the sampling frequency needs to be high enough so that accurate and timely information is provided to the operator to ensure good process control decisions. A method is needed to easily determine the most efficient sampling frequency and alert the operator if the sampling frequency is not adequate.

Existing statistical process analysis and control software (such as JMP®, Statistica®, and Minitab®) rely on the users to define the sampling frequency for control charts and capability studies. Statistical software generally provides tools to determine the appropriate sample size assuming a static process. However, these tools are not integrated with the process control software and data stream and also do not notify the operator or engineer that the sampling frequency is inadequate to meet process control or detection requirements. Also, tools are not provided to determine the appropriate control chart sampling frequency to detect changes in a dynamic process with a specified level of confidence.

We propose an improved system that integrates a method to determine the most efficient sampling frequencies with a process control and monitoring system. This includes providing a method for assessing the effectiveness of the current sampling frequency to detect changes in the process. It also includes a method to recommend the most efficient sampling frequency to detect changes in the process. The methods are applicable to both stable and unstable processes. In addition, this method can help process engineers determine the correct sampling frequency to accurately assess the impact of process improvements. This concept is applicable to any manufacturing operation which utilizes a process control system. The information stream into the system may be collected manually or automatically. This system can be added to existing process control software or added as an independent program.

The proposed tool can be used to determine the most effective sampling frequency to reliably detect changes in a data stream from a process. Since the system is integrated with the on-line process control system, this information is available in real time.

Details of the System

Appropriate sampling frequencies can vary over time according to operator needs. When the process is behaving well, the operator has a reduced need for information, so the rate of inspection can be relatively low. However, when the process is not behaving well, increased sampling...