Got to catch them all!

A Pareto analysis is based on the principle suggested by Joseph Juran (and named after Vilfredo Pareto) that 80% of the problems/issues are produced by 20% of the issues. This is the famous 80/20 rule, and this principle is sometimes summarized as the vital few versus the trivial many. Process improvement professionals use the Pareto principle to focus limited resources (time and money) on a limited number of items that produce the biggest benefit.

Translate Pareto analysis into action through the use of a Pareto chart.

The Pareto chart is a combination of a bar and line chart. The vertical axis (left – y-axis) is the frequency of occurrence, cost, or the sum of an important unit of measure. The right vertical axis is the total cumulative percentage. Display categories as bars in descending order on the horizontal axis (x-axis) with the cumulative percentage displayed as a line.

If I wanted to catch 50% of the Pokemon I would need to catch all of the Drowze and Machop (first two categories).

Creating a Pareto chart (without using EXCEL or another tool):

1. Decide what categories you will use to group items.  This is where the magic happens. I often err on the side of greater granularity.  In our example I decided to count each type of Pokemon; I could have counted the Pokemon by type (that would have assumed I actually knew the type), but by taking a granular approach I had a broad range of options (using set-based design principals).
2. Decide what measurement is appropriate (cost, quantity, frequency, or others). In the example, counting quantity makes the most sense.  An analysis of utility costs would measure cost.
3. Decide what period of time to use for the measurement sample.  In the example, I count the number of Pokemon for two visits (yes I do know Pokemon are algorithmically generated and are not real). The time period should be long enough to provide significant results and short enough not to drive up the cost of measurement.
4. Collect the data, use a check sheet or collect preexisting data.
5. Subtotal the measurements for each category. In addition, calculate the percentage of the total for each category.
6. Determine the appropriate scale based on the data collected in step 4.  The maximum value will be the largest value calculated in step 5. Mark the scale on primary (left) vertical axis.
7. On the horizontal axis, plot the categories as bar beginning with the largest category and finishing with the smallest.
8. Draw a second vertical axis on the right-hand side of the sheet and creating a scale from 0 to 100%.  Both of the vertical axes should be the same size (scale) — this means you should be able to draw a straight line from 100% to the maximum value on the primary vertical axis.
9. Beginning with the largest category plot the cumulative percentage contributed by each category.  As you add the percentage of a column plot the result on the chart. The last column should equal 100%.  Connect each of the points you plotted to generate an upward sloping curve.

The Pareto chart is a great tool to visualize how data contribute to the whole and which categories contribute the most.