By Issa Bass
 

Pareto analysis is simple; it is based on the principle that 80% of problems find their roots in 20% of causes. That principle was established by Vilfredo Pareto, a 19 th century Italian economist who discovered that 80% of the land in Italy was owned by only 20% of the population. Later empirical evidence showed that the 20/80 ratio was determined to have a universal application.

80% of customer dissatisfaction stems from 20% defects

80% of the wealth is in the hands of 20% of the people

20% of customers account for 80% of a business

When applied to management, the Pareto rule becomes an invaluable tool. In the case of a problem solving for instance, the objective should be to find and eliminate the circumstances that make the 20% "vital few" possible so that 80% of the problems are eliminated. It is worthy to note that Pareto Analysis is a better tool to detects and eliminate sources of problems when the those sources are independent variables.

The first step will be to clearly define the goals of the analysis. What is it that we are trying to achieve? What is the nature of the problem we are facing?

The next step in the Pareto Analysis is the data collection. All the data pertaining to factors that can potentially affect the problem being addressed need to be quantified and stratified. In most cases, a sophisticated statistical analysis is not necessary; a simple tally of the numbers suffices to prioritize the different factors. But in some cases the quantification might require statistical analysis to determine the level of correlation between the causes and the effect. A regression analysis can be used for that purpose, a coefficient of correlation or a coefficient of determination can be derived to estimate the level of association of the different factors to the problem being analyzed.

Then a categorization can be made, the factors are arranged according to how much they contribute to the problem. The data generated is used to build a cumulative frequency distribution.

The next step will be to create a Pareto Diagram or Pareto Chart in order to visualize the main factors that contribute to the problem, and therefore concentrate on the "vital few".

The Pareto Chart is a simple histogram; the horizontal axis shows the different factors while the vertical line represents the frequencies.

Since all the different causes will be listed on the same diagram, it is necessary to standardize the unit of measurement and set the time frame for the occurrences.

The building of the chart pre-requires a data organization. A four columns data summary must be created to organize the information collected. The first column will list the different factors that cause the problem, the second column will list the frequency of occurrence of the problem during a given time frame, the third column records the relative frequencies, in other words, the percentage of the total and the last column will record the cumulative frequencies, bearing in mind that the data are listed from the most important factor to the least.

The following data was gathered during a period of one month to analyze the reasons behind a high volume of customer return of cellular phones ordered on line.

The diagram itself will consist in three axes: The horizontal axis lists the factors, the left vertical axis lists frequency of occurrence, and it is graded from 0 to at least the highest frequency. The right vertical line is not always present on Pareto charts; it represents the percentage of occurrences, it is graded from 0 to 100%.

The breaking point (the point on the cumulative frequency line at which the curve is no longer steep) on this graph is at around "wrong product". Since the breaking point divides the "vital few" from "the trivial many", the two first factors "misinformed about the contract" and "Wrong products" are the factors that need more attention, by eliminating the circumstances that make them possible, we will eliminate about 71% of our problems.

About the author
Issa Bass is the managing editor of SixSigmaFirst. He can be reached at issa@sixsigmafirst.com

Tell us what you think about this article. Send a note to the Editor.

www.manorhouseassociates.com