weco-package: Western Electric Company Rules (WECO) for Shewhart Control...

In weco: Western Electric Company Rules (WECO) for Shewhart Control Chart

Description

Western Electric Company Rules (WECO) for Shewhart Control Chart

Introduction

WECO rules got the name from the quality control hand book published by the Western Electric Company in 1956. These rules detect the deviation from a stable process by continuously testing data against predefined abnormal patterns and have been accompanied with Shewhart control chart in statistical process control.

There are eight commonly adopted abnormal patterns, namely, eight WECO rules. It is of great interest to combine different WECO rules to increase the power to detect deviation from a stable process. This package implements the combination of multiple WECO rules.

Note that it is necessary to minimize the probability for making false alarms when the process is in control. Running length is the first time at which a process experiences out-of-control signals. The average running length (ARL) is frequently used as a measure in statistical process control for evaluating and comparing the performances of different methods.

Rules

This package considers the following eight WECO rules:

Rule 1

1 data point is greater than 3 standard deviation from the center line. (This rule is to identify single data point that is out of the acceptable range.)

Rule 2

9 data points in a row on the same side of the center line. (The ideal stable process is assumed to be up and down around the center line. A large block of data points on the same side of the center line indicates a process mean is shifted.)

Rule 3

6 data points in a row, all increasing or decreasing. (This rule is also an indicator of possible mean shift.)

Rule 4

16 data points in a row, alternating up and down. (When data points are routinely alternating up and down, it shows a high negative correlation between neighboring observations, which is abnormal for a stable process. For an in-control-process, it is not expected to observe correlation between neighboring data points.)

Rule 5

2 out of 3 data points on the same side are greater than 2 standard deviations from the center line. (For a normally distributed in-control-process, about 95 deviation. The chance to violate this rule is 0.00306. This rule is used to detect increase in process variation.)

Rule 6

4 out of 5 data points on the same side are greater than 1 standard deviation from the center line. (For a normally distributed in-control-process, about 62 deviation. This chance to violate this rule is 0.00553. This rule is also used to detect increase in process variation.)

Rule 7

15 data points in a row within 1 standard deviation of the center line. (Too many data points are within 1 standard deviation indicates the decrease in process variation.)

Rule 8

8 data points in a row are greater than 1 standard deviation of the center line. (This is another rule to detect increase in process variance.)

Graphical user interface (GUI)

This package provides a web-based Shiny GUI. See run.weco for details.

References

Introduction to Statistical Quality Control. Montgomery D.C. Third Edition. Wiley, New York. 1996

weco documentation built on May 2, 2019, 3:46 p.m.