Friday, September 12, 2008

An Introduction to Pareto Diagrams

Do you need a simple chart to convey the idea that quality corrective action is required? Pareto diagrams offer an effective, illustrative, and analytical tool for identifying quality problems.

A Pareto diagram is a modified histogram performance report. It receives its information from work results such as data figures, repair data, maintenance figures, or scrap rates.

Instead of grouping results by intervals as in histograms, Pareto diagrams lump similar quality defects together in order to identify the most common errors for corrective action. This is based upon the 80/20 rule where 80 percent of the defects are caused by 20 percent of the problems. Pareto diagrams are useful for measuring machine output and time reliant processes. There are three uses and types of Pareto diagrams.
  • Basic Pareto analysis - A basic Pareto analysis identifies the key contributors to the quality problem as a single diagram, with a percentage line showing each category as a portion of the cumulative total. The basic Pareto diagram identifies the most common problems.
  • Comparative Pareto analysis - A comparative Pareto analysis looks at a problem as a "before and after" comparison, using two Pareto diagrams. These diagrams show the change in the number of problems identified for each problem category.
  • Weighted Pareto analysis - Weighted Pareto analysis gives significance to unapparent factors such as cost, time, or criticality.
Once a Pareto diagram is constructed, the key defect areas become obvious, so you can reduce these defects to a more acceptable level. Then, after implementing corrective action, a new Pareto diagram may be constructed as a comparison to show that the key defects were greatly reduced.

Pareto diagrams are an extremely useful tool in quality decision making. The diagrams make it clear what quality changes need to be made and whether the remedy was effective. Keep in mind that Pareto diagrams do not identify causes, only problems, so other analytical methods may be required to gain further insight into the problem.

1 comment:

Anonymous said...

Your blog keeps getting better and better! Your older articles are not as good as newer ones you have a lot more creativity and originality now keep it up!