Tool: Pareto Charts. The Pareto Principle This is also known as the "80/20 Rule". The rule states that about 80% of the problems are created by 20% of. - презентация

1 Tool: Pareto Charts

2 The Pareto Principle This is also known as the "80/20 Rule". The rule states that about 80% of the problems are created by 20% of the causes. The Pareto principle was developed by an Italian economist Vilfredo Pareto who noticed that 80% of the wealth was owned by 20% of the population.

3 The Pareto Principle The Pareto principle implies that we can frequently solve a problem by identifying and attacking the vital few sources. This principle can be applied to most systems and processes. The concept is used to dissect a large problem into smaller pieces and in order to identify the biggest contributors.

4 Pareto Principle Examples Most of the equipment breakdowns are due to a small percentage of the equipment. The majority of calls to a IT help desk are attributed to a small number of reasons. Most of the errors in any process occur in one or two process steps. Only a handful of students in the school district account for most of the tardy events.

5 What is a Pareto Chart? A graphical representation of the Pareto Principle. A series of bars whose heights reflect the frequency of the problem. A graph where data is categorized to expose patterns.

6 Bar height shows relative importance; in descending order Bars represent each stratified category Pareto Chart Features Vertical axis shows relative percentages Other category can be used. Its always last. Vertical axis shows count of data points Red line shows cumulative percentages

7 When use a Pareto Chart? In the measure phase to identify which problem should be studied. In the analyze phase to verify which potential causes of the problem are most significant. In the improve phase to determine if the problem has improved.

8 Analyzing the Pareto Chart The Pareto Principle applies if one or more categories account for a large percentage of the occurrences. Look for the bars that are much taller than the rest.

9 If Pareto Principle Applies Focus your improvement efforts on the largest category or categories of the Pareto Chart, in order to achieve the maximum gain. Only focus on a smaller bar if it has a larger impact or is easier to fix. Using only the data from the large categories of the Pareto Chart, determine if they can be further stratified into additional categories.

10 Lower Level Pareto Charts 2 nd level Pareto Chart If the data from the Pareto chart can be stratified further, create 2 nd or even 3 rd level charts. Analyze these charts to determine if the Pareto Principle applies. When youve narrowed down the problems on the deepest levels you will start finding root causes. 1 st level Pareto Chart 3rd level Pareto Chart

11 Analyzing the Pareto chart The Pareto Principle does not apply if all the categories account for an approximately equal percentage of the occurrences. All the bars are about the same height.

12 The Other Category Occurrences in the other category should be redistributed to existing categories or a new category should be created If you create an other category ensure that it is not one of the larger bars on the chart.

13 Collecting the Data Data has to be collected in categories. The categories should be determined before the data is collected. A sample size of 50 data points is recommended. If 50 data points are not available, obtain as many as possible. The data only represents a snapshot of the process at a period in time. Since processes change over time, the data needs to represent the process you are analyzing.

14 Keys to Success Apply the Pareto principle early in your process improvement efforts to identify the vital few causes. The Pareto principle applies if a small number of causes account for a large portion of the problem. The data needs to be representative of the period in time that you are analyzing. If the Pareto principle applies, see if you can further stratify the data into other categories. If the Pareto principle does not apply, look for other ways to stratify the data.

15

16 Thanks … LOGANATHAN K