Pareto Analysis or 80/20 Principle
Pareto Analysis or 80/20 Principle
Pareto Analysis or 80/20 Principle
Pareto Analysis is one of the most popular and highly impactful decision-making tools used in business
process improvement. Specially, when you are breaming about where to start for improvements in your
business, this is the tool to embrace as it will present you with both high and detailed level of data
analysis in a very lucid manner thereby making it easier to select your next improvement target. In fact,
you can not only identify the problems that have biggest impact on your savings target but also focus on
the solutions that will provide the most benefits to your business.
In this article, I will explain how to carry out a Pareto Analysis and how to use its findings to prioritize
matters that can potentially have most positive impact in terms of improvements and savings.
But first, where does it come from? Pareto Analysis, also called Pareto Principle (popularly known as
80/20 Rule) was first purposed by an Italian economist Vilfredo Pareto in 1896. This rule states that 80
percent of the problems are originated from 20% of the causes. In other words, if we solve 20% of the
problem contributors then we can save 80% of the loss (or eliminate 80% waste). Though it is mostly
used in industrial engineering environment but it can easily and equally be applied in other fields,
including in office and administration settings.
Here is a brief explanation of how to use this tool. I have presented an example of a boat maintenance
company, who has been hit with many service calls leading to loss of revenue of $80/Hour. A Microsoft
Excel table has been created to depict the data.
Wear &
Tear Dents Lube Cracks Total
Engine 20 2 4 0 26
Propeller 11 8 2 0 21
Door 7 6 2 4 19
Deck 5 5 0 1 11
Windscree
n 5 0 0 2 7
Dashboard 1 0 0 2 3
Total 49 21 8 9 87
The data has been highlighted in yellow to indicate that first analysis was done based on the part types
and here is snapshot of the corresponding Pareto Chart.
24/02/21 Author: Mahadev Pant, Consultant Engineer
15 60%
cumulative %
54%
10 40%
30%
5 20%
0 0%
Engine Propeller Door Deck Windscreen Dashboard
Parts
Now we know that only three parts (viz. engine, propeller, and door) are making up to 76% of defects
and the biggest contributor to the total defects is engine (at 30%). So, let us now go to next level of
analysis by focusing on defects in engine only.
77% 80%
15
No. of Defects
cumulative %
60%
10
40%
5
20%
0 0%
Wear & Tear Lube Dents Cracks
Defect type
From this graph, we can see that out of total number of defective engines, 92% are due to wear & tear
and lube. Now we got clear view on what story the parts and the defects are narrating here and already
know what to analyse in detail.
24/02/21 Author: Mahadev Pant, Consultant Engineer
Similarly, the same data can be looked from a perspective of defects type. The data for the defects type
has been highlighted in yellow, followed by the Pareto Chart.
Wear &
Tear Dents Lube Cracks Total
Engine 20 2 4 0 26
Propeller 11 8 2 0 21
Door 7 6 2 4 19
Deck 5 5 0 1 11
Windscree
n 5 0 0 2 7
Dashboard 1 0 0 2 3
Total 49 21 8 9 87
30 60%
56%
20 40%
10 20%
0 0%
Wear & Tear Dents Lube Cracks
Defect type
Once we know this high level chart, we can zoom into more details. In this scenario, wear & tear is the
biggest bar so it is worth to check that in detail.
24/02/21 Author: Mahadev Pant, Consultant Engineer
88%
78% 80%
15
63%
No. of Defects
cumulative %
60%
10
41% 40%
5
20%
0 0%
Engine Propeller Door Deck Windscreen Dashboard
Defect type
With all this analysis, we can now confidently tell that our first attention must be on solving the issues
related to engine and wear & tear.
This tool:
Improve profitability
Even though this tool is very handy and can be deployed from simple to relatively complex scenarios, it
is not a sine-qua-non and therefore it is a good idea to accompany this type of analysis by other
supporting analysis tools. We should not take figure 80 and 20 literally rather it should be taken as a
guide only.
So, in this article, we have seen that by slicing and dicing the Data horizontally and vertically, we were
able to pick 2 or 3 key problem areas to perform root cause analysis.
In the next article, we will learn about root cause analysis tools! Stay tuned!