Class Twenty Five Notes

Class Twenty Five Notes - Page 1 of 22 Class Twenty Five...

This preview shows pages 1–3. Sign up to view the full content.

Page 1 of 22 1 Sheldon Kopp April 16, 2009 (3:51pm) E:\SWT\HA4315\Class notes\Spring 2009 version\C_25 spg09.wpd Reduce, reuse, restore, recycle HA 4315 “All the significant battles are waged within the self.” 1 I. Logistics and Objectives A. Memory Jogger II, pp. 36-51, introduction to control charts II. Run charts and control charts compared Last class I briefly took up run charts. Figure 1 shows a generic run chart in which the process exhibits only common causes. However, a run chart is not a complete tool. The main deficiency of a run chart per se is that variations can occur which are obviously special causes, but the run chart will not signal them as such. This is shown on Figure 2 . I put in an extreme point in the middle of the data. Obviously, the process that created that point is different from the process that created the points prior to it. Prior to the extreme high point, the graph would have looked like the one in Figure 1 . Indeed, it is exactly the same data with the exception of that one point being changed. Consider the rules for run charts. nine points in a row above or below the average seven consecutive points going up or down with no point falling below or above a previous point, respectively fourteen points simply repeating “up and down” with no two points in a row going in the same direction. These are indications of special causes. Let’s apply the run chart rules to Figure 2 . Generic run chart 0 5 10 15 20 25 30 35 40 45 9/1 1 /1 3/1 5 / 7 No special causes observed. Figure 1 Generic run chart 0 20 40 60 80 100 120 Still no special causes observed? Figure 2

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Page 2 of 22 April 16, 2009 (3:51pm) E:\SWT\HA4315\Class notes\Spring 2009 version\C_25 spg09.wpd Reduce, reuse, restore, recycle There no instances of nine points above or below the average line in a row. We do not see a seven point up or down trend or fourteen consecutive points up and down in a row. Thus, from the standpoint of run chart rules for when a special cause is in effect, we do not have one. Yet, anyone can clearly see that in Figure 2 that the process is changed. A control chart will show that event readily as a special cause. III. Control charts: an introduction Walter Shewhart first developed control charts back in the 1920s. What distinguishes a control chart from a run chart is the upper and lower control lines and zone lines which are between the upper and lower control lines. I put a sample one I developed below, Figure 3 . Figure 3 plots the number of cancer sites per patient for an oncologist’s practice over a number of weeks. What distinguishes it from the run chart is the addition of the horizontal lines. Each one represents a standard deviation. Interpretation of the control chart builds on the run chart ideas, but adds additional tests for statistically significant variation (special
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

Page1 / 22

Class Twenty Five Notes - Page 1 of 22 Class Twenty Five...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online