VARIABLES, MEASUREMENTS & DATA COLLECTION

A.

CHAPTER OBJECTIVES

B.

NATURE OF VARIABLES & CRITICAL TO QUALITY (CTQ) CONCEPTS

C.

MEASUREMENT AS A SOURCE OF INFORMATION

D.

DATA COLLECTION (DISCRETE Vs. CONTINUOUS)

E.

THE ADVANTAGE OF CONTINUOUS DATA

F.

CATEGORIES OF SCALE

G.

PLANNING AND DEVELOPMENT

H.

SAMPLING

I.

RANDOM SAMPLING

J.

SEQUENTIAL SAMPLING

K.

STRATIFIED SAMPLING

L.

DATA ORGANIZATION

M. TRANSACTIONAL TIME STUDY

9-1

CHAPTER OBJECTIVES
The objectives of this chapter are as follows:
Develop an understanding of dependent, independent and
leverage variables.
Become familiar with the Critical to Quality (CTQ) concept.
Understand how measurement is used as a source of
information.
Learn the difference between discrete and continuous
data the advantage of continuous data.
Become familiar with the categories of scale.

9-2

CHAPTER OBJECTIVES
The objectives of this chapter are as follows:
Establish uniformity in data collection.
Apply techniques in presenting statistical data for easy
comprehension.
Understand the different types of sampling; Random,
Sequential & Stratified.
Ensure proper data organization.

9-3

NATURE OF VARIABLES
Variables are elements subject to variation and describe a
cause and effect system.
Value taken by dependent variable (Y-output of process)
depends on state or condition of independent variables
(Xs-input of process).

9-4

LEVERAGE VARIABLES
Think Tracy,
think!

Vital
Few

Trivial
Many

Leverage variables (Xs) exert large influence on Y.

20% of X variables contribute 80% to the variation (output)
in Y (Pareto principle).
Objective is to identify/contain the vital few variables.
Powerful tool - DOE.

9-5

CRITICAL TO QUALITY (CTQ)

CONCEPT
We must understand our customer needs and translate
them into what we must do internally.
Customer needs can be defined in terms of CTQs such
as quality, delivery, price, etc.
Most CTQs are customer (internal & external) driven.
Other CTQs can be the result of:
High Risk priority factor (FMEA).
Economic benefits from defect reduction.
Regulatory on safety.

9-6

CRITICAL TO QUALITY (CTQ)

CONCEPT
Anything that impacts customer satisfaction is critical to quality.
On-time delivery.
Correct and complete orders.
Product functions properly.
Accuracy and readability of billing statements.
Timeliness of invoices.
Helpfulness in responding to and solving problems.
Courteousness.

9-7

CRITICAL TO QUALITY (CTQ) CONCEPT - DEFECTS

When determining a CTQ in a process, ensure that the defect is

truly defined.
A defect is any instance or event in which the product or process
fails to meet a customer requirement.
Speak to customers (internal or external) to establish the use of the
product or process.
For example:
an independent focus group is utilized to provide information for
an acceptable range of color on an ivory wallplate.
Anything outside that range is considered a defect.
A distributor is questioned on an acceptable time for receiving a
rebate on an invoice.

9-8

MEASUREMENT AS A SOURCE
OF INFORMATION
Another concept related to critical-to-quality (CTQ)
characteristics is measurement.
Without measurements, we do not have facts, only opinions.
To understand and improve our processes, we must measure
appropriate critical features.
There are features that cannot be measured with the precision
of a micrometer, but it is important we measure them.
Scales of measure help us to understand how to assess these
features and yield useful information.
Data must be transformed into relevant information.

9-9

MEASUREMENT AS A SOURCE
OF INFORMATION

Data does not provide information - you

have to torture data for it to give information.

And the instrument of torture is statistics.

Dr. Mikel Harry

9-10

MEASUREMENT AS A SOURCE OF INFORMATION

1. Information Sources:

2. What type of data?

Types
of problems

Words (A to Z)
Pictorial
Verbal

Types
of data

Numeric (0-9)

Types of
statistical tools

Region of interest
Grouping

Measurement

Discrete

Continuous

Nominal

Scale

Standard

Opportunity

Ordinal
Interval

3.

Ratio

Characteristic

Density

There are many sources of information, but numeric is most used.

We must:
Determine what we want to know.
Select a scale of measurement.
Collect and analyze the data.

9-11

DATA COLLECTION
(DISCRETE vs CONTINUOUS)
Business decisions entails the handling of large quantities
of data.
Unless collected properly, organized and simplified, data
cannot be used intelligently.
Proper data collection is very important to determine
where we are and where we want to go.
The precision, consistency and strategies during the
collection of data has a large bearing on analyzing,
improving and controlling a product or process.
Garbage in = Garbage out

9-12

DATA COLLECTION
Gather and select objective data in a form
appropriate for the tool used in the analysis of the
data.

9-13

TWO TYPES OF DATA

Qualitative data, subjective evaluation of a visual
characteristic which is not directly measurable. Such as
color, Go-NoGo or Pass-Fail (also termed Attribute Data).
Quantitative data consists of measures that take
numerical values for which descriptions such as means and
standard deviations are meaningful. They can be put into
an order and further divided into two categories: discrete
and continuous.

Discrete data are countable data, for example, the

number of defective items produced during a days

production.
Continuous data, when the parameter (variables) are
measurable, are expressed on a continuous scale. For
example measuring the length of a part.
9-14

DISCRETE VS CONTINUOUS DATA

DISCRETE

CONTINUOUS

NO-GO

GO

Caliper

SHIPPING ORDER
QTY

UNIT

DESCRIPTION

TOTAL

$10.00

$1.50

$4.50

10

$10.00

$10.00

$5.00

$10.00

FAIL

$10.00

ERROR

Time

PASS

TEMPERATURE
Thermometer
Electrical Circuit

Two kinds of data can be used for measuring purposes:

Discrete (Attribute) = Qualitative data
Continuous (Variable) = Quantitative data

9-15

DISCRETE DATA (ATTRIBUTE)

Count the frequency of occurrence (i.e.; number of
times something happens or fails to happen).
This data cannot be meaningfully subdivided into
more precise increments.
Includes Go/NoGo and Pass/Fail data.

9-16

CONTINUOUS DATA (VARIABLE)

Data that characterizes a product/process in terms of
weight, volts, size, etc.
Data can be meaningfully divided into finer increments of
precision.
Continuous data provides much more information than
discrete data.
Considered to cost less to collect since we require less of
it to make a decision.

9-17

THE ADVANTAGE OF CONTINUOUS DATA

To Obtain the Same Level of Understanding

e
cret
Dis

Sparse
Information

Continuous

Rich With
Information

More advantageous to seek continuous data even if several forms

of discrete data is available.
Validity of inferences using discrete data is highly dependent on
number of observations.
Sample size required to characterize a discrete product/process
feature is much larger than required sample size when continuous
data is used.
9-18

THE ADVANTAGE OF CONTINOUS DATA

Discrete data tends to be:
Defect related; available only after the fact that a defect has occurred.
Reflective of long term trends with no guidance as to what went
wrong.
Expensive to collect because so much of it is needed.
Deficient in the amount of information obtained.
Continuous data is often:
Available earlier during the process (you can design measurement
strategies to collect the data following the process flow).
Reflective of short term trends, allowing immediate corrective action.
Less expensive than discrete data, because less is required for a
given level of understanding.
Rich with useful information.

9-19

CATEGORIES OF SCALE
Categories of scale

Description

Examples

Nominal: unrelated categories which

represent membership or
non-membership.

Discrete data (usually)

Grouping/sorting
Yes/no, pass/fail
Arithmetic not possible

Categories
Labels

Ordinal: ordered categories with no

information about distance
between categories.

Discrete data
Ranking
Seldom used
Very little arithmetic possible

Interval: ordered categories with

equal distance between
categories, but no
absolute zero point.

Continuous data
Most common scale
Use arithmetic with caution

Temperature scales
Dial indicator
5-1=4-1

Ratio:

Continuous data
Proportional relationship
Most forms of arithmetic apply

Velocity = distance/time
Ruler
6=32

ordered categories with

equal distance between
categories with an
absolute zero point.

1st, 2nd, 3rd

Relative height
Alphabetic order
1<2<3<4

During a project, different measurement scales may be used

as the level of knowledge increases from the early stages
through the complete understanding of all critical variables.

9-20

NOMINAL SCALE
Manpower

Equipment

Application
Poor paint
adhesion

Environment
Application form
Select one from each group
Nationality

Marital status

Occupation

Materials
List of Field Reps
Access to
a digital
camera

Without
access to
a digital
camera

Bob T.

Fred W.

Jim C.

Bill S.

Joe W.

John D.

Diane A.

Sam C.

Use nominal scale to establish different types of defects prior to

computing any statistics.
Use of nominal scale is very common in transactional areas of
business.
There is no relationship between the various categories and their size,
importance, etc. In effect there is no scale of measure.
9-21

ORDINAL SCALE
Relative Size

Example 1: Pareto Chart - Paint Adhesion Test

Example 2: Customer Survey

Question:How would you
rate our service?

r
ato
Ope
r

Pai
Typ nt
e
App
lica
tion
Hum
idity

Pre
p

ara
ti

on
Prim
Typ er
e

Order of
importance

Excellent
Very Good
Good

Ordinal Scale

Fair
Poor

Useful to rank data into ordered lists or sequences (Pareto

chart).
Used in customer surveys to measure satisfaction ratings.
Puts nominal data into ordered categories.
Involves only discrete data.

9-22

INTERVAL AND RATIO SCALE

Examples of interval scale:
(no absolute zero)
1. Displaced Scale
50
40
30
20
10
0

Examples of Ratio Scale

(absolute zero)
2. Dial Gage
0.10
0
0.20
Gage
block

3.Relative Velocity

1. Ruler
100
90
80
70
60
50
40
30
20
10
0

2.PositionvsTime
at Constant Speed

3.W eight As a Function

of Number of Bricks

Interval scale does not consider the absolute zero; ratio scale does.
Use interval scale if we want to accurately measure small variations in a
large object.
Use the ratio scale when we use a ruler or tape measure to measure
the length of an object.
Most data falls into these categories and is rich with information. 9-23

LOGARITHMIC SCALE

Linear Vertical Scale

Logarithmic Scale

Numbers are masked by lack of sensitivity on the linear vertical

scale.
Present results on a logarithmic scale to provide a better view of
9-24
progress; it reveals small variations.

LIKERT SCALE
Example Rating Scale Ranges
Five-point school grading system (A B C D E)
Seven-point numerical rating (1 2 3 4 5 6 7)
Verbal scale (excellent, good, average, fair, poor)
Types of Questions for Questionnaires
Strongly agree
Somewhat agree
Neither agree or disagree
Somewhat disagree
Strongly disagree
Much better than expected
Somewhat better than expected
As expected
Somewhat less than expected
Much less than expected

One of the best of all

Better than most
About the same as most
Worse than many
One of the worst of all

Common method of rating various topics through use of surveys.

Quantitative values can be attached to survey questions to derive a
numerical score.
9-25
Technique is effective for transactional processes.

PLANNING & DEVELOPMENT

Data must be collected according to a well-developed plan
if valid information on a study is to be obtained.
The plan must identify important variables related to the
study and specify how they are to be measured.
From the data collection plan, a statistical model can be
formulated from which inferences can be drawn.

9-26

PLANNING
1. What you need to know.
2. The format that you would like to see or need the information.
3. Identify where to get information.
4. Identify who can get this information.
5. Identify how data will be organized.
6. Design the data collection sheets.
7. Collect Data (keeping traceability).
8. Record and retain the data.

9-27

DEVELOPMENT
I LOVE
SIX SIGMA!

A good data collection system should

have the following properties:
Planned and coordinated.
Uses standard data collection sheets.
Clear work instructions for collecting
the data.
Well-organized and easily
comprehended.
Consistent data interpretation.
9-28

TYPES OF DATA
COLLECTION SHEETS
Inspection sheets are forms where CTQ characteristic to be
measured or monitored are identified and recorded.
Data sheets are simply tabular or columnar format for
recording data.
Check sheets are forms which provide process data and
trends.
Checklists refer to forms which simply list steps to be
taken.
Although different in appearance and use, they ensure data
is collected, recorded and interpreted the same way.

9-29

DATA COLLECTION SHEET EXAMPLE

(Inspection Sheet)
Procedure: Measure CTQs BY1, BY2 and BY3 using a dial indicator. Record measurements on the form.

Project ID
Component Number
SPANWISE BEAM # 3 ASSY 007U5003
CTQ

CTQ

Ser No.

BY1

BY2

BY3

Ser No.

BY1

BY2

BY3

CTQ

BY2

BY1

Measurements

BY3

Lower Chord

-A-

(W/L)
Upper Chord
Lower
OUTBD

KC 101U9123.BY
CTQ W/L
CTQ
BY

-A-

BY1
BY2
BY3

B/L

114.87 LBL 110

STA
1043.54

114.87 LBL 6.00 1043.54

114.87 RBL 108 1043.54

Critical characteristics (CTQ) to be inspected or tested.

Method of inspection and testing. (i.e. Visual, electrical, dimensional etc) and instruments to be used.
The location of the measurement in a graphical format (Blueprint, diagrams, instructions or procedures).
The method of recording and retaining the data collected.

9-30

DEVELOPMENT OF
DATA COLLECTION SHEETS
Data Collection Sheets are best developed using a crossfunctional team approach.
Best way is to hold a brainstorming session on the
different aspects needed to measure the CTQs. After that,
a form is developed..
Data entry should be formatted in the same fashion as it
will be analyzed. Therefore reformatting is unnecessary.
To ensure consistency a form should be developed which
is unique to a specific component, product type or
process. Therefore, each form is individually tailored for a
specific application.

9-31

SAMPLING
Sampling plays a most important role in statistical
analysis. Often it is simply not possible, or not desirable to
examine the entire population of interest.

Sampling saves money, time and effort.

Careful collection of data from a sample will often provide

better information than a less careful study that tries to
look at everything.

9-32

THREE BASIC TYPES OF SAMPLING

Random sampling is used to ensure that the measurements or

units selected taken are the result of pure chance.

Sequential sampling is the selection of sample units in the order

of production or receipt.

Stratified sampling is used when a lot is the result of several

combinations of inputs such as: multiple employees, machines,
shifts, etc. In other words, the product is the combination of
several smaller lots.

9-33

RANDOM SAMPLING
A random sampling is one where the conditions are such
that each unit of the population has an equal chance of
being chosen.

9-34

Example - Random Sampling

Check Ball Process
Check Balls

Process
Output

3
22
62

16
37
71

Random Number List

The assembly department has been having problems with the assembly of an
engine check valve due to variation in the diameter of the check balls.
At the manufacturing process, data was collected on the diameter of the ball. The
data indicated that the process was capable of producing good parts.
The sampling plan was reviewed and it was noted that the operator measures the
first six parts produced for each run of check balls which apparently did not
represent the quality of the whole run.
Samples need to be selected at random to increase the likelihood that the
measurements of the samples are representative of the process population.
It improves the validity of population estimates and statistical tests such as
capability analysis or the calculation of the probability of defect.
Sampling only the first few parts does not show how the process performs over
time.
Changes throughout the day such as heat and wear have a big impact to this
process capability.

9-35

SEQUENTIAL SAMPLING

In sequential sampling, samples are selected in the

order in which they are produced.

9-36

EXAMPLE - SEQUENTIAL SAMPLING

Bolt Process

0.250 +- 0.001

5th hour
4th hour

Process
Output

3rd hour
2nd hour
1st hour

x5

x4 x3 x2

x1

A team is studying the process of manufacturing bolts. The bolt

diameter is identified as the CTQ characteristic, and to determine
the process baseline of the CTQ, the project team set-up a
sampling plan.
The process runs throughout the day so they wanted to make sure
that the samples would collectively represent the true output of the
process.
Data were collected from samples done in sequence. By comparing
the variance between samples, they gained insight into the shortterm process capability.
9-37

EXAMPLE - SEQUENTIAL SAMPLING

Cont.
Sequential Sampling simply involves collecting the
samples in sequence to ensure that we can see the
influences of time on the output.
A sequential sample of five bolts was pulled each hour.
The team used a Data Collection Sheet to document its
measurements. On the back of the document, detailed
instructions were included. Space was provided on the
Data Collection Form to record traceability information on
the samples selected. It was very important in the analysis
of the data to be able to determine when the sample was
selected.

9-38

STRATIFIED SAMPLING
Stratified sampling is used when the lots are known to be the
result of multiple sub-lots. In other words, the lots are the
combination of differences such as shift, employees, machines,
etc.
In taking many stratified samples from a given lot, the average
number of defectives will be the same as if the samples were
drawn at a random.

9-39

EXAMPLE - STRATIFIED SAMPLING

I

Random
Numbers

II

37 16
22 71
62 3

III

8
2
24

66
55
70

32 56
23 10
33 74

A manufacturer of ball bearings has determined that one of the

CTQs is the spherical diameter of the balls. The manufacturer has
three processes, which produce the balls simultaneously, and
therefore is unsure as to the proper sampling technique. A
sampling plan has to be developed.

9-40

EXAMPLE - STRATIFIED SAMPLING

Cont.

Since the lot is the result of multiple operators and machines, the
product is actually the combination of a several smaller lots. Therefore,
a Stratified Sampling plan was chosen.
Stratified Sampling will increase the likelihood that the measurements
of the samples are representative of the combined lots' process
population. It improves the validity of population estimates and
statistical tests.
In Stratified Sampling, the samples are selected proportionately from
each process. Within each lot we are pulling a random selection of
balls. Our Stratified Sample is made up of the combination of these
random selections. This specific type of Stratified Sampling is better
known as Stratified Random Sampling. Stratified Samples can also be
selected sequentially. This is known as Stratified Sequential Sampling.
The team used a Data Collection Sheet to document their findings and
included detailed measurement instructions. Spaces were provided on
the Data Collection Sheet to record traceability information on the
samples selected.

9-41

DATA ORGANIZATION
Expandable

Data
Collection

Forms

Sheet

.158
.886
.479
.478
.490
.159
.645
.712
.159
.331
.068
.684
.919
.030
.757
.227
.985
.632
.900
.062
.946
.960

.077
.753
.555
.031
.936
.185
.085
.657
.550
.456
.648
.674
.001
.340
.049
.984
.562
.419
.788
.858
.465
.518

.926
.073
.003
.540
.209
.224
.499
.823
.800
.772
.537
.951
.654
.081
.037
.464
.716
.322
.754
.483
.396
.076

.407
.406
.193
.493
.909
.373
.838
.061
.185
.099
.181
.481
.162
.429
.092
.420
.537
.994
.660
.408
.948
.430

.693
.579
.781
.490
.576
.332
.560
.544
.618
.859
.619
.543
.723
.823
.363
.727
.358
.383
.397
.409
.271
.062

.951
.220
.091
.016
.340
.883
.266
.806
.491
.405
.132
.984
.373
.633
.200
.282
.000
.221
.043
.721
.313
.835

.015
.677
.623
.729
.207
.558
.209
.707
.393
.564
.132
.421
.495
.863
.795
.431
.606
.593
.408
.918
.795
.181

.522
.194
.590
.883
.888
.674
.346
.504
.228
.697
.362
.077
.701
.966
.760
.369
.992
.928
.763
.697
.568
.913

.565
.902
.808
.182
.812
.773
.112
.490
.443
.941
.583
.990
.610
.591
.058
.654
.254
.892
.192
.701
.613
.114

.591
.630
.465
.894
.082
.362
.155
.774
.822
.562
.735
.373
.891
.492
.581
.832
.078
.607
.202
.349
.077
.332

.345
.663
.837
.050
.939
.269
.332
.249
.535
.298
.467
.706
.384
.341
.711
.405
.335
.344
.617
.280
.693
.853

.358
.933
.161
.028
.667
.668
.836
.874
.772
.638
.451
.986
.607
.163
.666
.044
.772
.339
.416
.404
.194
.626

.222
.018
.891
.379
.583
.899
.017
.171
.430
.244
.539
.124
.893
.741
.575
.071
.081
.908
.371
.994
.335
.036

.566
.684
.302
.762
.187
.298
.866
.433
.602
.256
.442
.389
.412
.053
.495
.713
.290
.311
.633
.161
.046
.581

Logical

Need

Flexible

Traceable

Meaningful
Information

9-42

WHY ORGANIZE DATA?

It prevents numerous pages of data which tend
not to be consistent and not always recorded in
the same format.
Data can be compared and contrasted clearly
and accurately to determine process variations.
It provides a logical flow for easy access and
manipulation.

9-43

ORGANIZE DATA
Logical - format is meaningful to the user.
Expandable - enlarged without changing the anatomy.
Flexible - any portion of the anatomy can be manipulated for a
specific task.
Traceable to its source.
Transformable and built in way that data can easily migrate to
other applications.

9-44

TRANSACTIONAL TIME STUDIES

Conducting transactional time studies are an effective way
of gauging the robustness of a process.
Time traps become during the time study.
Performing transactional time studies either in a service
area (ie. Customer Service, Marketing, Engineering, etc.)
or manufacturing, success is based on the initial planning
and implementation of the study.

9-45

TRANSACTIONAL TIME STUDIES

PLANNING
Several steps along with supplies must be taken into
consideration when planning a time study.
A detail level and/or top level process map of the process.
This should initially be the as-is map. In most
instances, a detail level process map is required.
Personnel involved in the process.
If possible, evaluate at least two people performing the
same process to reduce bias in the study.
Stop watch(es) for timing the process steps.
Data collection sheets structured for easy recording.
Department personnel might be relied upon to complete
the data sheets in an easy to use/understand format.

9-46

TRANSACTIONAL TIME STUDIES

PLANNING
Once planned, review all process maps with
department personnel and management.
Provide and explanation of how the study will be
conducted.
This will prevent nervousness, confusion and errors
during the study.
Keep department work-in-progress disruptions to a
minimum.
If available, have independent observer(s) attend the
audit. They can also act as assistants/time keepers.

9-47

TRANSACTIONAL TIME STUDIES

EXAMPLE
The following example demonstrates the most commonly
performed transactional time study involving manual paper
transactions.
Conducted in the Corporate Claims Department on the
Home Depot freight allowance process.
Team consensus determined the process could be
improved by automating the manual system, thus a dollar
figure had to be established for the cost of the manual
system.

9-48

HOME DEPOT FREIGHT ALLOWANCE

EXAMPLE
The first step in planning the time study involved
establishing each process step based on a process map,
as depicted on the next slide.
A brainstorming session with key department team
members provided details for each step, where company
policies were noted along with time queues.
Process steps that require timing personnel were noted on
the process map.
Brian, the modern
caveman, thinks
Transactional project
good, ugh.

9-49

SIX SIGMA DEBIT MEMO PROJECT

HOME DEPOT FREIGHT ALLOWANCES
DETAIL PROCESS FLOW MAP

TIMED PROCESS
$.0062 per
transaction or
$6,200 per million

CLAIMS DEPT
Claims Associate sorts
through charge-back
notifications to isolate
freight debits (transit
facilities) and discards
debits under $50.*
(POLICY)

CLAIMS DEPT
Claims Associate keys
invoice numbers from
305 report to verify that
charge is from a transit
facility.

Home Depot
Transit Facility?

TIMED PROCESS
$.0428 per
transaction or
$42,776 per million

HOME DEPT CORP

Charge-back notification
sent via mail.

NO

CLAIMS DEPT
Based on investigation
debit code changed
(corrected).

YES

CORP CREDIT
Claims Clerk sorts credit
memos and returns to
Claims Associate
(Beth).

TIMED PROCESS
$.0248 per transaction
or $24,800 per million

CORP CREDIT
Claims Associate
reconciles (keys) credit
memos with Home
Depot freight chargebacks in CARMS.
END OF PROCESS

TIMED PROCESS
#.0347 per transaction
or $34,700 per million

CLAIMS DEPT
Claims Associate sums
Home Depot dollar
amounts and enters one
or more credits (PLION)
to Home Depot account.

TIMED PROCESS
$.302 per
transaction

CLAIMS DEPT
Claims Associate waits
for EDI check
remittance from Home
Depot and is notified via
email internally. Occurs
approx 1 week later.

TIMED PROCESS
$.302 per
transaction

WAIT = 0-1 DAY

CLAIMS DEPT
Claims Associate
requests 305 report
which includes freight
debits for all customers
(email to Corp Credit)
Run time = Overnight

CORP CREDIT
Vic approves all credits
in PLION less than
$10,000.

WAIT = 1 DAY
CLAIMS DEPT
Claims Associate sorts
Home Depot debits on
305 report and verifies
credit sum is less than
$10,000

CLAIMS DEPT
Plion generates credit
memos. Overnight batch
run.

WAIT = 1 DAY FOR CREDIT MEMO PRINTOUT

WAIT = 2 DAYS TO ENTER CARMS

Total =
$.7125 per transaction or
$712,500 per million
*Of 1,607 transactions
processed, 1,303 were
under $50 debits

9-50

HOME DEPOT FREIGHT ALLOWANCE

EXAMPLE
The next step involves replicating the information from the
process map to a data collection sheet for each step.
Process steps that contributed an insignificant amount of
time are excluded from timing.
Some examples of exclusions are approving an on-line
pending request or requesting a report.
When performing cost calculations, time queues that have
an associated cost should be included in the study.

9-51

DATA
COLLECTION
SHEET

9-52

HOME DEPOT FREIGHT ALLOWANCE

EXAMPLE
Once the planning stage is complete, implementing the study is
the next step.
Establish a time study date and block of time with department
management and personnel will limit interruptions during the
study.
Ensure that all employees understand their role and explain that
the process is under evaluation, not their performance.
This will prevent unnecessary additional stress that can skew
the time study results.
Using a stopwatch, time each process step along with counting
the number of transactions and record all data on the data
collection sheet.

9-53

HOME DEPOT FREIGHT ALLOWANCE

COST CALCUATIONS
Step 2 was calculated based on the number of transactions, the
time to complete those transactions & the average salary for a
Claims Associate, broken down to the amount of salary per
minute.
$20.00 hourly wage for Claims Associate (includes benefits)

1.

1,607 trans
= 48.7 trans/min
33 min

2.

$20.00 hour
= $0.333/min
60 min

3.

$.333
1 min
X
= $0.0068/trans
1 min
48.7 trans

4.

$0.0068
X 1 min = $6,837.70/million trans
1 trans

9-54