Histogram in 7 QC Tools

 

 - Histogram is a visual interpretation of numerical data showing the number of data points falling within a specified range of values.

 - It is a frequency distribution tool for process improvement.

 - We can used only one parameter in variable data to construct this chart.

 - Histogram is similar to bar graph but histogram shows no gaps between the bars.

 - Histogram is represented by a curve known as Frequency Distribution.

Parts of Histogram

1. Title

 - Title describes the information included in histogram.

2. X-axis

 - This axis shows the scale of values which the measurements fall under.

3. Y-axis

 - This axis shows the number of times the values occurred within the set of interval of X-axis.

4. The Bars

 - The height of bar shows the number o times that the values occurred within the interval and the width of bar shows the interval that is covered.

5. Legends

 - Legend provides additional information like where the data came from and how the measurements were gathered.

What is a use of histogram?

 - Summarize the data in graphical format.

 - Analyze the variation in process.

 - Assist in decision making.

How to Make Histogram?

1. Collection of Data and Compute the Range

 - Here we take example of 50 records. Make a table of data and calculate the range by subtracting Largest and Smallest value.

 

2. Find the Number of intervals and interval width

 - Here we have to find the number of intervals required and interval width.

3. Summary of data as per interval size and construct the graph

 - Here we summarized the data as per interval width.

 - Then we construct the graph. Y-axis shows frequency of data points and X-axis shows Interval width.

Types of Histogram Patterns

1. Normal Distribution Pattern

 - It's a bell shaped curve where peak point is at center of the curve.

2. Skewed Distribution Pattern

 - In this curve peak point is either at right side or left side of the curve and the tail stretches away from it. Right side peak called Right skewed distribution and left side peak called Left skewed distribution.

3. Double Peak Pattern

 - In this curve two peak point observed like two humped camel.

 - This curve shows combination of two process or records.

4. Multi Peak / Plateau pattern

 - In this curve several process are combined.

 - Range selection is not proper. Top distribution looks like plateau.

5. Edge Peak Pattern

 - This curve is same as normal distribution except that it has a large peak at one tail.

 - This cause by faulty construction of histogram or instrument error.

6. Truncated Pattern

 - The curve looks like normal distribution with the tails cut off.

 - The supplier might be supplying the parts which are within specified limit only by producing normal distribution at their end.

7. Dog Food Pattern

 - The dog food pattern is like missing the results near the average values. 

 - This curve shows remaining / missing part of truncated pattern which was sorted by supplier to avoid rejection at customer end.

Benefits of Histogram

 - Histogram allow to compare data easily.

 - Equal interval allows easy data transfer from frequency to histogram.

 - Helps focus on the general shape of the data.

Disadvantage of Histogram

 - Cannot read exact values from histogram because data is grouped into categories

 - Distribution shape affected by change in interval width