Basic Tools for Process Improvement

B asic Tools for Process Im provem ent
What is a Pareto Chart?
A Pareto Chart is “a series of bars whose heights reflect the frequency or impact of
problems. The bars are arranged in descending order of height from left to right.
This means the categories represented by the tall bars on the left are relatively more
significant than those on the right” [Ref. 5]. The chart gets its name from the Pareto
Principle, which postulates that 80 percent of the trouble comes from 20 percent of
the problems. Viewgraph 1 highlights the elements of this definition.
Why should teams use Pareto Charts?
You can think of the benefits of using Pareto Charts in economic terms (Viewgraph
2). “A Pareto Chart breaks a big problem into smaller pieces (and) identifies the
biggest contributors. . . (It can) help us get the most improvement with the resources
available by showing where to focus efforts in order to maximize achievements. The
Pareto Principle states that a small number of causes accounts for most of the
problems. Focusing efforts on the ‘vital few’ causes is usually a better use of
valuable resources” [Ref. 1a].
When should we use a Pareto Chart?
A Pareto Chart is a good tool to use when the process you are investigating
produces data that are broken down into categories and you can count the number of
times each category occurs.
No matter where you are in your process improvement efforts, Pareto Charts can be
helpful, “. . . early on to identify which problem should be studied, later to narrow
down which causes of the problem to address first. Since they draw everyone's
attention to the ‘vital few’ important factors where the payback is likely to be greatest,
(they) can be used to build consensus. In general, teams should focus their attention
first on the biggest problems—those with the highest bars” [Ref. 5].
Making problem-solving decisions isn’t the only use of the Pareto Principle. Since
Pareto Charts convey information in a way that enables you to see clearly the
choices that should be made, they can be used to set priorities for many practical
applications in your command. Some examples are:
! Process improvement efforts for increased unit readiness
! Skills you want your division to have
! Customer needs
! Suppliers
! Investment opportunities
2
PA RETO CHA RT

---

B asic Tools for Process Im provem ent
What Is a Pareto Chart?
• Bar chart arranged in descending
order of height from left to right
• Bars on left relatively more important
than those on right
• Separates the "vital few" from the
"trivial many" (Pareto Principle)
PARETO CHART
VIEWGRAPH 1
Why Use a Pareto Chart?
• Breaks big problem into smaller pieces
• Identifies most significant factors
• Shows where to focus efforts
• Allows better use of limited resources
PARETO CHART
VIEWGRAPH 2
PA RETO CHA RT
3

---

B asic Tools for Process Im provem ent
How is a Pareto Chart constructed?
To construct a Pareto Chart, you need to start with meaningful data which you have
collected and categorized. You may want to turn to the Data Collection module at
this point to review the process of collecting and categorizing data that you can chart.
Now you’re ready to follow the steps for constructing a Pareto Chart (Viewgraphs 3
and 4). The steps below have been adapted from Joiner [Ref. 1b].
Step 1 - Record the raw data. List each category and its associated data count.
Step 2 - Order the data. Prepare an analysis sheet, putting the categories in order
and placing the one with the largest count first.
Step 3 - Label the left-hand vertical axis. Make sure the labels are spaced in
equal intervals from 0 to a round number equal to or just larger than the total of all
counts. Provide a caption to describe the unit of measurement being used.
Step 4 - Label the horizontal axis. Make the widths of all of the bars the same
and label the categories from largest to smallest. An "other" category can be
used last to capture several smaller sets of data. Provide a caption to describe
them. If the contributor names are long, label the axis A, B, C, etc. and provide a
key.
Step 5 - Plot a bar for each category. The height of each bar should equal the
count for that category. The widths of the bars should be identical.
Step 6 - Find the cum ulative counts. Each category's cumulative count is the
count for that category added to the counts for all larger categories.
Step 7 - A dd a cum ulative line. This is optional. Label the right axis from 0 to
100%, and line up the 100% with the grand total on the left axis. For each
category, put a dot as high as the cumulative total and in line with the right
edge of that category's bar. Connect all the dots with straight lines.
4
PA RETO CHA RT

---

B asic Tools for Process Im provem ent
Constructing a Pareto Chart
Step 1 - Record the data
Step 2 - Order the data
Step 3 - Label the vertical axis
Step 4 - Label the horizontal axis
Step 5 - Plot the bars
PARETO CHART
VIEWGRAPH 3
Constructing a Pareto Chart
Step 6 - Add up the counts
Step 7 - Add a cumulative line
Step 8 - Add title, legend, and date
Step 9 - Analyze the diagram
PARETO CHART
VIEWGRAPH 4
PA RETO CHA RT
5

---

B asic Tools for Process Im provem ent
Step 8 - A dd title, legend, and date.
Step 9 - A nalyze the diagram . Look for the break point on the cumulative
percent graph. It can be identified by a marked change in the slope of the graph
(see Viewgraph 11). This separates the significant few from the trivial many.
NOTE: The significant few-trivial many principle does not always hold. No matter
how many data are categorized, they can be ranked and made into a Pareto
diagram. But sometimes no single bar is dramatically different from the others, and
the Pareto Chart looks flat or gently sloping. To attack the tall bar in that situation is
no help. You need to look for another way to categorize the data.
Now let's look at an exam ple to illustrate the Pareto Chart construction process:
You recently inherited $10,000 and would like to apply it to some of your
outstanding bills. Here is what you owe:
Home improvement loan balance
$1,956
Visa
$2,007
Mastercard
$1,983
Church building fund pledge (monthly
installments of $83.33 for two years)
$2,000
Balance of car loan
$1,971
School tuition (monthly installments of
$169.17 for one year)
$2,030
Viewgraphs 5 and 6 show how this would look when recorded on an analysis sheet
and plotted on a Pareto Chart.
6
PA RETO CHA RT

---

B asic Tools for Process Im provem ent
Analysis Sheet Example
Outstanding Debts
Category
Amount ($)
School tuition (monthly installments)
2,030
Visa
2,007
Church pledge (monthly installments)
2,000
Mastercard
1,983
Balance of car loan
1,971
Home improvement loan balance
1,956
Total
11,947
PARETO CHART
VIEWGRAPH 5
Pareto Chart Example
Outstanding Debts
$ 12,000
10,800
d
9,600
e
w
8,400
O
7,200
6,000
ount
4,800
m
3,600
A
2,400
1,200
0
Tuition
Visa Church
MC
Car
Home
Type of Debt
LEGEND: AMOUNT OWED ON OUTSTANDING DEBTS AS OF 12 FEB 95.
PARETO CHART
VIEWGRAPH 6
PA RETO CHA RT
7

---

B asic Tools for Process Im provem ent
You probably noticed that no single bar is dramatically different from the others.
Looking at your outstanding debts in this way isn't much help. Is there a different way
the data could be categorized to make it more meaningful? What if you were to
consider the interest rates on your outstanding debts?
Viewgraphs 7 and 8 show what that would look like. A much clearer picture of your
outstanding debts now emerges, and you are able to make a better decision on how to
manage your money.
NOTE: In this example, we opted not to show a cum ulative line (Step 7) on the right
side of the Pareto Chart because it might be confusing to create a percentage of
percentages.
How do we interpret a Pareto Chart?
When you look at a Pareto Chart, you can see break points in the heights of the bars
which indicate the most important categories. This information is useful when you are
establishing priorities.
As you can see in the example we've just looked at, you can detect two big breaks in
the heights of the bars when you categorize the data in a different way:
!
The first break point is between the second and third bars. The difference
between these two bars is much more noticeable than the other differences.
This shows the relative importance of the first two bars in relation to the
others.
!
The other break point occurs after the fourth bar. Addressing the third and
fourth bars will give a higher payoff than addressing the last two bars.
You will have an opportunity to develop your interpretation skills when you do the
practice exercises that follow.
8
PA RETO CHA RT

---

B asic Tools for Process Im provem ent
Analysis Sheet Example
Interest Rates on Outstanding Debts
Category
Int. Rate (%)
Visa
21
Mastercard
18
Balance of car loan
9
Home improvement loan balance
6
School tuition (monthly installments)
2
Church pledge (monthly installments)
1
PARETO CHART
VIEWGRAPH 7
Pareto Chart Example
Interest Rates on Outstanding Debts
57.0
51.3
45.6
s
39.9
te
a
34.2
t
R
28.5
s
22.8
re
te
17.1
I
n
11.4
5.7
0.0
Visa
MC
Car
Home Tuition Church
Type of Debt
LEGEND: INTEREST RATES CHARGED ON OUTSTANDING DEBTS AS OF 12 FEB 95.
PARETO CHART
VIEWGRAPH 8
PA RETO CHA RT
9

---

B asic Tools for Process Im provem ent
How can we practice what we’ve learned?
Now let's practice constructing and interpreting Pareto Charts using the data provided
for the scenario in each exercise. Analysis sheets and Pareto Charts are provided as
answer keys.
EXERCISE 1: You are part of a team responsible for the quality of the BEQ/BOQ.
You want to improve the service you provide but are not sure where to begin or where
to concentrate your efforts. You decide to keep track of the complaints received over
a three-month period. Viewgraph 9 records this data.
Rank order your data in an analysis sheet and draw a Pareto Chart.
Now answ er the follow ing questions:
1. Does the Pareto Principle apply?
2. Where should you concentrate your efforts?
3. Do your proposed efforts pass a sanity check? In other w ords, do the
top problem s really stand out as m ost im portant to you and your
custom ers?
10
PA RETO CHA RT

---

B asic Tools for Process Im provem ent
Data on BEQ / BOQ Complaints
Complaint
Feb
Mar
Apr
Total
Loud stereo noise after 2300
30
50
17
97
Insufficient hot water
23
20
11
54
Towels too small and/or thin
12
8
12
32
Inadequate lighting
175
100
75
350
Poor quality TV reception
10
13
60
83
Worn-out furniture
1
4
10
15
Insufficient storage space
25
52
50
127
Cockroaches
324
265
373
962
Rooms too warm or too cold
300
110
95
505
PARETO CHART
VIEWGRAPH 9
PA RETO CHA RT
11

---