**It’s Never About The Number, It’s What The Number Is Telling You**

**Successful Control Charts For Every Business Environment**

All successful leaders know how to see around corners. What is their secret? They use Key Performance Indicators (KPIs) complete with early warning systems. Both are components of any successful Lean Six Sigma program.

If it can be measured, it can be improved. But how do you know if the process is stable, and producing predictable results? The answer is simple, use control charts.

Many people often think the use of control charts as being limited to a manufacturing environment. This is an incorrect assumption. The application is much more widespread and successful, especially in today’s economic environment. These charts, commonly used in Lean Six Sigma, are appropriate for any process that can be measured.

A control chart is useful for…

- determining the capability of the process (expected range).
- determining if the process is under statistical control (stable).
- identifying special cause variation in the process.
- determining when and if corrective actions are required to the process.

By definition, a control chart is a graphical depiction of variance about a centerline over time. The mean (often referred to as an average) is commonly the centerline used in business applications. When using a mean as the centerline, the control chart is referred to as an `X (X bar) control chart. `X is nothing more than a fancy way of referring to the mean.

It is important to note that when a process is deemed to be out of statistical control, investigation is required. The cause of the variation must be identified and correction measures implemented when necessary.

The use of control charts can be extremely valuable when managing processes. However, to be of value these charts can’t sit in dusty binders in the manager’s office. The charts belong at the work location and should be reviewed during the leader’s Gemba Walk.

Furthermore, control charts should not be viewed as the only tool necessary for continuous improvement. Control Charts are only __one__ of many tools available in the lean enterprise’s arsenal in the war on waste.

While statistics are crucial in the measurement of business processes, it is never about the number. It is about what the number is telling the leaders about their business and what is being done differently because of the number.

**Constructing a Zone Control Chart**

To assemble an `X control chart, enough data must be collected to calculate the mean (average) and standard deviation (std dev). After calculating the standard deviation, you determine the mean (average) __+__ 1, 2, and 3 standard deviations (std dev). Finally, plot the Excel line chart including the data points, mean, 1 std dev, 2 std dev, and 3 std dev. Text labels are recommended for the right side of the graph to label the mean and __+__ 3 std dev (upper and lower control limits).

Process control charts can be broken into three zones on each side of the mean (average) for ease of analysis. Zone “C” is closest to the centerline and is bounded by the mean and __+__ 1^{ }times the standard deviation. Zone “B” is the area bounded by __+__ 1 and __+__ 2 times the standard deviation. Zone “A” is the area bounded by __+ __2^{ }and __+__ 3 times the standard deviation.

Having assembled the control chart, the next step involves understanding what the data is revealing about the business. More specifically, is the process data under statistical control? A process only has two states, in control and out of control. There is a series of rules that are used to detect abnormal conditions in which the process is said to be out of statistical control.

While there is some disagreement among experts over the rules for determining when a process is out of statistical control, these rules were defined by Dr. Douglas Montgomery in his 2005 book entitled Introduction to Statistical Quality Control.

1. One point or more points outside of the control limits (i.e. beyond Zone A)

2. Two out of three consecutive points three points > 2 std devs from the mean, on the same side of center, and within the control limits (i.e. in Zone A)

3. 4 out of 5 points > 1 std dev from the mean and on the same side of the mean (i.e. in Zone B or beyond)

4. Eight consecutive points on the same side of the center line (i.e. Zone C and beyond)

5. Six points in a row steadily increasing or decreasing

6. Fifteen points in a row on both sides of the centerline in Zone C.

7. Fourteen points in a row alternating up and down.

8. Eight points in a row on both sides of the centerline and beyond Zone C.

9. An unusual or nonrandom pattern in the data.

10. One or more points near a warning or control limit.

Learn more about the author by checking my LinkedIn profile at http://www.linkedin.com/in/roycewilliard

Reference

Montgomery, Douglas C. (2005). *Introduction to statistical quality control, Fifth edition*. USA: John Wiley & Sons, Inc..

Post Author: Royce Williard

Copyright 2009, The Williard Group LLC

## No comments:

## Post a Comment