As we all know there is variation everywhere and in everything. This includes people pencils, traffic at an intersection, customer needs, goods, services, and of course processes. If you can not predict this variation, I can bet you are compensating for it and that compensation is costing you money!!
In Six Sigma we use statistics to recognize and thus assess and understand that variation so we can predict it ahead of time. This reduces our costs and increases customer satisfaction. Statistics can help you “Picture” variation. Many times we look at data that we gather in various ways but some times we forget about the element of time. At any given point things may look great but over time many times things vary. This variation can not be seen in what we call a histogram of all the data or any other kind of plot that does not use time on one of the axis. For example below (Figure 1) is a plot of Systolic blood pressure of one individual over four months. This plots shows where the average Systolic Blood Pressure would be if it was Normal, Marginally High, or High. In this individuals case it does not look that great.
As you can see this individuals Systolic pressure is all over the place and does not look like a normal distribution. It may be Bi or Multi modal (two or more things effecting the results making several peaks in the Histogram). From this we can not draw to much of a conclusion but to tell him to see a doctor. Or should we??
If I take that same data and plot over time we get the following graphic.
Now clearly you can see a shift in the blood pressure over the 4 months. All for the better. We might not want the person to see the doctor but find out what might have cause this improvement. A different solution just because we looked at the data in a time sequence chart.
What these two graphs tell us is that if you have collected data over any given time period I’d suggest that you plot it in what is called an x or individuals chart (Figure 2 above) over time and see what you see in patterns there. In fact many times I plot the data in several different ways just to see what I can find. What kinds of patterns I see that can lead me to why the data varies so much.
If, you have questions or comments please feel free to contact me by leaving a comment below, emailing me, calling me, or leaving a comment on my website.
Bersbach Consulting
Peter Bersbach
Six Sigma Master Black Belt
http://sixsigmatrainingconsulting.com
peter@bersbach.com
1.520.829.0090
