Pareto Analysis

last edited by: Frederic Casagrande on Dec 11, 2018 7:45 AM login/register to edit this page

Contents
1 Applications
2 Procedures
3 Instructions
4 References

A technique used to identify the few factors that have the greatest impact on an outcome, result, or on measures of quality, satisfaction or performance. Separating these "vital few" from the "trivial many" is generally known as the "80/20" rule. Applying this rule to sources of problems would mean that, on average, 80 percent of the problems can be traced back to 20 percent of all key causes. Identifying the high-volume impact factors in terms of frequency of occurrence helps to determine the focus of problem solving energy. In addition, the frequency distribution of these vital few can set the order of priorities of action. The Pareto diagram graphically depicts these distributions similar to a bar graph or histogram, showing the frequency of occurrence in descending order from left to right.

Applications

  • To provide a means for identifying the key factors that have the highest occurrences.
  • To enable prioritization and focus of problem solving efforts.
  • To aid in the analysis of the current situation with respect to activity performance, service, or quality.
  • To monitor the effectiveness of solutions once implemented.

Procedures

  1. Identify the problem or area of focus.
  2. Identify sources of data.
  3. Collect data.
  4. Tabulate data, using a table or matrix to display frequencies of occurrence.
  5. Transfer the tabulated data into a distribution chart.
  6. Plot the cumulative frequency.
  7. Analyze results and determine next steps.

Instructions

Identify the problem to be solved or select an area of focus to begin the analysis. In the following example, "delays in vendor payment" was identified as one of the key problem areas that needed to be addressed during a reengineering project. Problems can be identified by applying an appropriate technique (see Root Cause Analysis, Problem Analysis). If data is not available regarding the frequency of potential causes or other factors, design an appropriate experiment or conduct research to collect the data. Business cycles such as daily, weekly, monthly, quarterly, or annually are appropriate time frames to conduct the research. In the example that follows, 400 vendor payments were delayed over the course of one year. For each delay, the cause(s) of the delay were tabulated. The frequency of each factor was also identified.

Once the raw data has been collected, calculate the percentage occurrence for each factor or cause. Translate this data into a Pareto chart, plotting several of the most frequently occurring factors in descending order from left to right on the horizontal axis versus the respective cumulative frequency (or cumulative percentage) shown on the vertical axis. Label each axis. Make sure that the graduation on the vertical axis and on the horizontal axis is similar to give the diagram a proportionate look. One axis should not be excessively larger than the other. For example, the width of the "bars" can be adjusted or a space can be added between each bar. When there are many occurrences and many different causal factors, select an appropriate break-off point and show only the most frequently occurring factors. (Remember, the purpose is to show only the vital few.) Summarize the remaining factors using a category called "other." Plot the cumulative frequencies and connect the points to form a cumulative curve. Label each point. Place the total number of occurrences for each factor on each bar. Analyze the results. This analysis can focus the team on the areas that will realize the largest gains from the redesign effort (where to "get the most bang for the buck"). In the example, three factors - incorrect vendor name, incorrect vendor address, and missing invoice number - accounted for almost 80 percent of delays in processing a vendor payment. These "vital few" should be the focus of subsequent problem solving activities, such as determining additional causes for each of these factors (e.g., "Why is the vendor name so frequently wrong?" See Root Causes Analysis). In addition, these "vital few" should be used as a check for completion when reengineering is completed and/or solutions have been proposed (e.g., "Have the solutions proposed eliminated the causes of these three factors?").

The Pareto diagram assists in predicting the effectiveness of improvements because it shows the relative importance of the causes. It can also be used to determine the effectiveness of the solutions by comparing the original diagram to one constructed after the solutions have been implemented. In addition, once a baseline has been tabulated and graphed, data from other time periods can be overlaid to show progress and/or trends.

This technique can be applied in other ways. For example, it may be important to display which organization has the highest volume of mail to determine where a quality team might focus its efforts on cost reduction. The Pareto chart would plot the organizations on the horizontal axis and the cumulative percentage of the budget on the vertical axis.

References

  1. V. Daniel Hunt. Quality In America, How To Implement A Competitive Quality Program. Business One Irwin, 1992.
  2. Kazuo Ozeki, Tetsuichi Asaka. Handbook of Quality Tools, The Japanese Approach. Productivity Press, 1990.


last edited by: Frederic Casagrande on Dec 11, 2018 7:45 AM login/register to edit this page


Reviews (4)

Login/join to subscribe
ADVERTISEMENTS

"In the real world, the right thing never happens in the right place and the right time. It is the job of journalists and historians to make it appear that it has."

- Mark Twain

ADVERTISEMENT

Sponsors