Drawing Histograms

An important aspect of total quality is the identification and control of all sources of variation so that processes produce essentially the same result over and over again. A histogram is a tool that allows you to understand at a glance the variation that exists in a process. Although the histogram is essentially a bar chart, it creates a "lumpy distribution curve" that can be used to help identify and eliminate the causes of variations. Histograms are especially useful in the measure, analyze and control phases the Lean Six Sigma.

What can it do for you?
A histogram will show the central value of a feature produced by your process, and the shape and size of the dispersion on either side of this central value. The shape and size of the dispersion will help identify sources of variation, otherwise hidden. The data used to generate a histogram can ultimately be used to determine the ability of a process to produce an output that always falls within specification.

How can you do?
1. Decide the essential characteristic to the quality you want to examine. The CTQ must be measurable on a linear scale. In other words, the value added between the units of measurement must be the same. For example, a micrometer or a thermometer or a stopwatch that can produce linear data. Ask your customers to rate your performance as "poor" to "excellent" on a scale of five points will probably not.
2. Measuring the characteristics and record results. If the feature is continually being produced, such as the tension in a line or temperature in a furnace, or if too many items being produced to measure each, you will enjoy. Please be sure that your survey is random.
3. Count the number of individual data points. Add values for each data point and dividing by the number of points. This is the mean (or average) value.
4. Determine the highest value of data and the lowest value of data. Subtract the smaller number of higher. This is the range.
5. The next step is to determine how many "classes" or bars your histogram should have.

To make an initial decision, you can use this table:

Number of data points Number of classes
under 50 years 5-7
50-100 June to October
100-250 July-December
over 250 10-20

6. Divide the range by the number of classes of trial that you have selected. The resulting number will be your class interval of First Instance (the horizontal scale or width) for each bar on your card. You may round or simplify this number to make it easier to work, but the total number of classes must be within those listed above. To determine the number of classes and the class interval, consider how we measure the data. Increase or decrease the number of classes or modify the class interval until there is essentially the same number of measurement possibilities in each class.
7. Determine the class boundaries. You can do this by starting with the center of the range. If you have an odd number of classrooms, a center of the middle class at about the midpoint of the range, then alternately add or subtract the class interval to define the boundaries of other categories. If you have an even number of classes, begin the process of adding or subtracting the class interval at approximately the middle of the beach.
8. Count the number of data points that fall within each class. Add the frequency totals for each class. This number should equal the total number of data points. Divide the number of data points in each class by the total number of data points. This will give you the percentage of points for each class. Add the percentages of all classes. The result should be about 100.
9. Chart of results starting with Lowe.