By evaluating a histogram, you can see the spread of a data distribution, as well as how many and which measurements are hitting upper specification limits. To create a histogram, you’ll need to take regular measurements of a specific process. The value of each measurement becomes a bar on the chart. For example, this histogram charts measurements of a location C on a specific part. By looking at the measurements on a histogram, we can see that:
- There is a wide spread across the distribution.
- Numerous measurements are hitting the upper spec limit.
- An expected loss of about 5% will occur if the process continues in the current manner.