Basic Process

Improvement Training
Version 1.5
3rd March 2011

1
 Objectives

• The objective of Process Improvement is

to produce effective and efficient
processes

– Effectiveness : Does the process produce

the desired result and meet the customer’s
needs?
– Efficiency : Does the process use the
absolute minimum machine, manpower and
time to produce the desired results for the
customer?
2
Process Improvement Roadmap (using
DMAIC)

Any Project where you are trying

to improve the end to end process
time or cycle time, use Step 8,
otherwise use Step 9

3
Step 1 – Identify Voice of the Customer

4
 What is VOC?

Voice of the customer (VOC) is a term to

describe the in-depth process of capturing a
customer's expectations, needs, wants,
preferences and aversions

The output of a VOC exercise are these

“raw” customer sound bites that captures
their expectations, needs, wants,
preferences and aversions.

5
 1.1 Identify Customers (Groups)

• Who are the customers?

• Does it make sense to divide them into customer groups or
segments? Typical groups:
– Customer type
– Type of usage (direct customer / end user, consumer)
– Product/service type they buy
– Importance to your business (revenue, profit, volume
purchased from your company)
– Region
– Demographics (age, education, gender…)
– other
 1.2 Gather Customer Data
Reactive data Proactive data
• The data comes to you • You’ve got to go to get the
data

For the customers/customer groups it’s important to find out what’s important to
them regarding your products or services:

What do they consider as good performance vs. bad performance

What do they need your product/service for?
How would a different performance of your product/ service make their life
easier?
 Reactive Data
Examples Customers
• Revenue analysis • Current and former
• Complaints customers
• Hotline reports
• Service call reports Data on
• Warranty claims • Issues and difficulties
• Web site hit-rate • Unmet needs
• Specific interests
 Proactive Data
Examples Customers
• Comment/registration cards • Future, current and former
• Interviews, focus groups customers
• Gemba visits
• Market research Data on
• Benchmarking • Met and unmet needs
• Product/service features’
relative importance to the
customer
• General interests
 1.3 Organize the VOC

• Organize the information as statements in

customer language (cards/post-it notes)
• Group the customer needs into groups
(affinity diagram)
 Statements in Customer Language

• Write the customer statements on cards

or post-it notes (one statement per card).
Fast Pick up
Fast Pick up
immediately
immed
Tasty pizza

On time Friendly

Example: VOC for a Pizza Delivery Service

 Grouping of Needs
Timeliness Behavior
Fast delivery
Delivery as Courier
On-time
announced smiles
delivery

Not stressed GM
deliverer shows up
Miscellaneous

No damage Box looks

of pizza box good Appearance
Right
Location Shape Color

Round
Salami cut Right color
evenly
Example: VOC for a Pizza Thin pizza
shell
delivery service
 Exercise 1 – Structured Interview

1. Work in Pairs
2. Work out a list of questions you want to
answer the customers of the process you
want to improve
3. One of the Pair will play the role of the
customer and the other will play the role of
the interviewer
4. Collect the VOC
5. Organize the VOC

13
Step 2 – Identify Process

14
 2. Identify the Process (that causes the
Problem)
A simple process map helps in defining the project:
• It brings the process view into the project
(What is the project’s related process?).
• It clarifies the project scope
(What are the process boundaries under consideration?).
• It lists the customers and their requirements to the process (outputs).
• It gives a first idea of possibly important inputs (Xs) of the process and who
supplies them.

Suppliers Inputs Process Outputs Customers

Start

Stop
 Creating a High-Level Process Map
1. Start by clarifying the process boundaries (start and stop)
2. List the process customers (internal and external, direct and
indirect)
3. List the process outputs (physical outputs as well as features
and measurables)
4. List the main process steps (not more than 5-8), use verb / noun
structure to describe the steps (e.g. “initiate payment” and not
“operations”)
5. List the inputs necessary to perform the process steps
6. List the suppliers that provide the inputs
Suppliers Inputs Process Outputs Customers
Start

Stop
 SIPOC Example:
Medical Service Process
Oncological Customer Service Process:
Servicing hospitals with problems related to oncological product

Supplier Input Process Output Customer

Hospital Trouble report from customer Receive call Machine running Hospital

Engineering Problem solving tools for call Define problem (including Problem solved
center severity defined)

Internal Service Group Training for call center reps Try to solve the problem over the Satisfaction
phone

Training Department Training for FSE (Field Service Contact FSE

Engineers)

Spare Parts Work Team Tools for FSE Organize visit (including
suspected spare parts)

Spare Parts Onsite trouble shoot

Replace spare part

Check of system
 Exercise 2 – Draw the SIPOC of the Process
you are trying to improve
Supplier Input

start
Process

end
Output Customer
s

18
Step 3 – Produce Initial Project Charter

19
 Project Charter
Name of Project
Name of Process
Problem Statement
What is the definition What % of the Output
of defective? is presently
defective?
Goal Statement
Process Owner Project Sponsor
Process’ Customer Key Stakeholders
Project Leader Start Date
Team Members Planned End Date
Project Coach Actual End Date

20
 Example of initial Project Charter
Name of Project Improve the Timeliness of Home Deliveries for Mac Donalds
Singapore
Name of Process Home Delivery Process
Problem Statement From Jan to March 2011, we received 300 complaints from
customers that their orders failed to arrive on time (30 mins from
Order to Delivery).
What is the definition More than 30 mins What % of the Output To be measured
of defective ? from Order to Delivery is presently
defective?
Goal Statement Pending baseline measurement
Process Owner Mr Donalds Project Sponsor Mrs Mac
Process’ Customer People who order Key Stakeholders Mr ABC, Mrs CDS. Mr
meals over the phone XYZ
Project Leader Mr White Belt Start Date 12 Apr 2011
Team Members Mr Team Member 1 Planned End Date 15 June 2011
and Ms Team Member
2
Project Coach Mr Black Belt Actual End Date

21
 Initial Project Charter
Name of Project Fill this
Name of Process Fill this
Problem Statement Fill this
What is the definition Fill this What % of the Output To be determined in
of defective? is presently Measure Phase
defective?
Goal Statement To be determined in Measure Phase
Process Owner Fill this Project Sponsor Fill this
Process’ Customer Fill this Key Stakeholders Fill this
Project Leader Fill this Start Date Fill this
Team Members Fill this Planned End Date Fill this
Project Coach Fill this Actual End Date Fill this

22
 Exercise 3 – Do your Initial Project Charter
Name of Project
Name of Process
Problem Statement
What is the definition What % of the Output
of defective ? is presently
defective?
Goal Statement
Process Owner Project Sponsor
Process’ Customer Key Stakeholders
Project Leader Start Date
Team Members Planned End Date
Project Coach Actual End Date

23
Step 4 – Translate VOC to CTQ

24
 What is VOC and What is CTQ?
Voice of Customer Critical to Quality
Voice of the customer (VOC) is a A metric that reflects what the
term to describe the in-depth process customer wants and that
process of capturing a customer's can be measured internally
expectations, needs, wants,
preferences and aversions 1.A key concept of Six Sigma is to
identify what is “Critical to Quality”
to the customer as a metric (CTQ)
The output of a VOC exercise are 2.The CTQ is always a Y (output)
these “raw” customer sound bites metric
that captures their expectations, 3.The CTQ must be measurable
needs, wants, preferences and 4.Quality is defined by the
customer and not by the process
aversions. expert

25
Two Kinds of Data that you must be
able to Distinguish
A type of data is discrete if there are only a
finite number of values possible or if there is
a space on the number line between each 2
possible values.

a.Space on the Number Line between 2

values
• Count of Anything e.g., 1 Apple, 2
Apples, 3 Apples. There is no such
thing as 1.5 Apples
a.Proportions & Percentages – derived from
Count Data
• Proportion - Number of Fails / Total
• % - Number of Fails / Total x 100
a.Finite Number of Possible Values
• Pass or Fail
• Strongly Disagree, Disagree, Agree,
Strongly Agree
26
 Two Kinds of Data that you must be
able to Distinguish
A type of data is continuous if it can take
any value along a number line.

In order to capture continuous data, a

measurement system has to be used.

The usefulness of the data depends on the

quality of the measurement system

Some common examples:

Use the half test. Half of
any continuous value is a.Elapsed Time – 30 minutes, 30.1 minutes,
always meaningful (Half 30.11 minutes, 30.111 minutes. This can go
of $10 = $5, Half of $5 = on forever.
$2.50 and so on. Half of b.Height – 1.8 meters, 1.85 meters, 1.857
2 persons = 1 person, meters.
Half of 1 person is ½ c.Temperature – 30 degrees C, 30.6
person (no meaningful, degrees C, 30.68 degrees C
hence count is not d.Etc 27
continuous. Count is
 Examples of Data Types
Type How Obtained Examples
Continuous Measuring instrument or a Elapse cycle time, weight,
(or “variables”) calculation length, speed
Discrete: Count occurrences and Proportion of : defective
Percentage or proportion non-occurrences items, late applications
Discrete: Count occurrences in an Number of : errors,
Count area of opportunity complaints, accidents
Discrete: Observation Type of : application,
Attribute product, customer
Discrete: Observation or ranking Customer rating
Ordinal

28
 From VOC to CTQ
VOC Needs CTQ Type of Specificati Operational
& Want CTQ on Definition

I get really Accurat % Accuracy Discrete* Not Data is accurate if

annoyed when I e Data of XYZ data Applicable it matches the
come to 3000 in the source’s value on
XTRA and I find product the same day of
the value of this population
XYZ data is
wrong
It took IT Fast Cycle Time Continuous 4 hours or Time start = when
Helpdesk almost Service from less IT Helpdesk
2 weeks to get Service receives the email
back to me on Request to (not read, but
my email Resolution received by the
request for of Problem email system)
assistance and is less than Time end = when
resolve my or equals to the customer
problem 4 hours sends an email to
Only Continuous CTQs need specification say the issue has
29
been resolved
 What are specifications?

• For Continuous CTQs, you have to set specifications to determine what is good and what is bad.
• Example train service:
• CTQ: Difference between planned arrival time and actual arrival time in minutes
• Specification: Every actual vs. planned arrival difference of more than 10 minutes is a defect
(“I’m not on time for the meeting”).
• Example bank service:
• CTQ: Time waiting in line before being served
• Specification: Waiting time of more than 7 minutes is a defect (“Must be able to do a bank
transaction plus having something to eat in my lunch break”).
Step 5. Measure Baseline

31
 Why measure the Baseline?

• The baseline measurement tells us how is the

process is performing (before the improvement
action) vis-à-vis the customer expectations
• It gives us a reference line from which we can later
claim whether any improvement have taken place
• This is obvious but worth stating – if you do not
have a baseline, it is impossible to claim that a
process has improved. Improved always has this
connotation of improved from what?

32
 Measure the Baseline Performance

1. Define what is a defective

2. Determine the time period to obtain one
data point of measurement (1 shift, 1 day,
week, 1 month? 1 day is recommended)
3. Determine how many time periods you can
reasonably collect
4. Obtain the # of output per Time Period
5. Separate the output into Defective and Non
Defective
6. Compute the % Defective
7. Plot the % Defective in a Time Series Plot
33
 Example 1 – Discrete Y example
1. Define what is a defective - Incomplete Data Entries. Data Entries are complete only
if all five fields are populated.
2. Determine the time period to obtain one data point of measurement (1 shift, 1 day,
week, 1 month? 1 day is recommended) – Every Working Day
3. Determine how many time periods you can reasonably collect – 10 working days
4. Collect Data at Job Level – See next slide
5. Obtain the # of output per Time Period - see table below
6. Separate the output into Defective and Non Defective – see table below
7. Compute the % Defective – see table below

Day 1 2 3 4 5 6 7 8 9 10Total

# of Output 10 15 12 20 5 10 11 15 12 10 120
7. Plot the % Defective in a Time Series Plot – See slide after the next slide
# of Defective 2 3 1 0 1 2 1 0 1 4 15

% Defective 20% 20% 8% 0% 20% 20% 9% 0% 8% 40% 13%

34
 Instructions on Data Collection (Discrete)
Collect data at Job Level like so

Date Job Number Field 1 Value Field 2 Value Field 4 Value Field 5 Value Completed Entry?
1-Jan-11 2011185ABC Company $1,235,000.00Latin America Not Listed Yes
1-Jan-11 2011186CDF Company USA Listed No
1-Jan-11 2011187ZZZ Company $230,000,012.00Asia Listed Yes
1-Jan-11 2011188JJ Corp $7,845,341.00EMEA Listed Yes

Use Pivot Table Function in EXCEL to obtain this kind of consolidated data

Count of Completed Entry? Completed Entry?

Date No Yes Grand Total % Defective
1-Jan-11 1 5 6 0.1666667
2-Jan-11 1 6 7 0.1428571
3-Jan-11 1 2 3 0.3333333
Grand Total 3 13 16

Use the % Defective column and the Date column to plot the Time Series

* Note – this is just a small example using very few data points. In your project,
you would be collecting much more data over many days
35
 Baseline Performance Time Series Plot (so
the baseline is 13% defective or 87% Pass)

On average, 13% (the P-Bar) of the data is incomplete on any

given day. You can expect between 0% and 40% of the data to
be incomplete each day. 36
 Example 2 – Continuous Y example
1. Define what is a defective – More than 30 mins elapsed from the Order to
Delivery
2. Determine the time period to obtain one data point of measurement (1 shift, 1
day, week, 1 month? 1 day is recommended) – Every Working Day
3. Determine how many time periods you can reasonably collect – 10 working
days
4. Collect Data at Job Level – See next slide
5. Obtain the # of output per Time Period - see table below
6. Separate the output into Defective and Non Defective – see table below
7. Compute the % Defective – see table below

# of Output 30 20 22 32 44 30 23 22 15 30 268
# of Defective 4 1 2 5 3 4 2 5 2 5 33
% Defective 13% 5% 9% 16% 7% 13% 9% 23% 13% 17% 12%
P-Bar 12% 12% 12% 12% 12% 12% 12% 12% 12% 12%
7. Plot the % Defective in a Time Series Plot – See slide following next

37
 Instruction on Data Collection (Continuous)
Collect data at Job Level like so
Date Date Time Start Date Time End Cycle Time (Mins) Pass or Fail?

1-Jan-11 1/1/2011 13:45 1/1/2011 15:10 85Fail

1-Jan-11 1/1/2011 13:50 1/1/2011 14:15 25Pass

2-Jan-11 1/2/2011 9:45 1/2/2011 11:11 86Fail

Use Pivot Table Function in EXCEL to obtain this kind of consolidated data
Count of Pass or Fail? Pass or Fail?
Date Fail Pass Grand Total % Fail
1/1/2011 1 1 2 50%
1/2/2011 2 2 100%
Grand Total 3 1 4

Use the % Defective column and the Date column to plot the Time Series

* Note – this is just a small example using very few data points. In your project,
you would be collecting much more data over many days
38
 Baseline Performance Time Series Plot (so
the baseline is 12% defective or 88% Pass)

On average, 12% (the P-Bar) of the deliveries are not timely on

any given day. You can expect between 5% and 23% of the
deliveries to be not on time each day.
39
 What is the technical name of this time
series plot of % defective?
• It is called a P Chart.
• It plots the proportion of nonconforming units.
• For example, use to examine whether wait times are
less than 5 minutes, flights depart on time, bicycle
tires are flat, printed logos are smudged, and so on.
• Please note (many people get this wrong) – the
P-bar or average proportion is not the sum of the
individual proportions / sample size. It is the sum of
all defectives / total output.

40
 How to set the Goal?

1. The how many X improvement method

• Suppose your % defective 10%
• Therefore % Non Defective is 90%
• If you want to improve this by 2X, then you take 10% /
2 = 5%
• Hence, the goal will be to improve from 10% defective
to 5% defective or from 90% Non Defective to 95%
Non Defective
1. Ask your sponsor what the goal should be, given
the baseline
2. Ask the customer

41
 6. Set the Goal
CTQ Baseline Goal Method used to
determine goal
% Accuracy of XYZ 90% 95% 2X improvement
data in the product

Cycle Time from 88% 99% Internal Target from the

Service Request to Process Owner
Resolution of
Problem is less
than or equals to 4
hours

42
 7. Update the Project Charter
Name of Project Improve the Timeliness of Home Deliveries for Mac Donalds
Singapore
Name of Process Home Delivery Process
Problem Statement From Jan to March 2011, we received 300 complaints from
customers that their orders failed to arrive on time (30 mins from
Order to Delivery).
What is the definition More than 30 mins What % of the Output 25% of Orders
of defective ? from Order to Delivery is presently failed to be
defective? delivered within 30
mins
Goal Statement Improve the % of Orders to be Delivered by 30 mins from
75% to 99% by June 2011
Process Owner Mr Donalds Project Sponsor Mrs Mac
Process’ Customer People who order Key Stakeholders Mr ABC, Mrs CDS. Mr
meals over the phone XYZ
Project Leader Mr White Belt Start Date 12 Apr 2011
Team Members Mr Team Member 1 Planned End Date 15 June 2011
and Ms Team Member
2
Project Coach Mr Black Belt Actual End Date 43
Step 8. Process Analysis (CTQ is Cycle Time)

44
 What is Process Analysis?

• Process Analysis involves the use of Process Maps to

understand what are the root causes of failure and
use this understanding to improve the process
• Process Maps are used to create a common
understanding of any of the following:
– HOW WE THINK IT IS Map: What you or the team
or management think the process is
– AS IS Map: What the process really is
– TO BE Map: What the process could be
– SHOULD BE Map: What the process should be

45
 Two Schools of Thought about Process
Analysis

Find out what needs to be

Map the As-Is Process
changed / Improved on Map the To-Be Process Implement the Changes

Analyze the As-Is Approach

Gather the Map the As-Is Process Create a Transition Plan

Map the Should Be
Requirements from As-Is to Should Be
Process
and implement
(Ignore the As-Is)

Clean Slate Approach 46

 Which approach is better?
Clean Slate Approach Analyze the As-Is
Approach
Innovation Better for Innovation Suffers from in-the-box
Better for Out of the Box thinking
Thinking Less likely to result in real
innovation
Ease of Change Harder Easier

Speed of Harder Easier

Change
Use when The As-Is process is The As-Is process more or
completely useless and less correct. Just needs step
needs to be completely wise improvement.
overhauled OR
transformational change
is needed
47
 Process Mapping

• Every Process Map should contain the following

information:

– The triggering event that starts the process (Oval Symbol)

– Steps which transform the job from one state to another
state or Process Steps (Rectangle Symbol)
– Decision points where the job is channeled to different
routes within the process (Rectangle Symbol)
– The end state where the job exits the process (Oval
Symbol)

48
Decision symbol can be
Yes / No type OR it can
be IF type
No Process Mapping (Basic)
Write the Triggering Event here –
Yes Start e.g., Customer applies for Credit
Card

If …..
Process Steps – use a Verb Noun
format e.g., Check Creditworthiness
Process Step
If …..

Decision – State the

Decisio
Question e.g., Credit Process Step
Worthy? n

Process Steps – DO
Process Steps – use a NOT write an essay in
Verb Noun format e.g., Process Step the box ! Keep it Simple.
Issue Card Noun Verb. That’s it.

Write the End Event

here – e.g., Customer
receives Credit Card End End
49
 Thinking Process for Process Analysis differs
if you are trying to improve Cycle Time vs.
Not Cycle Time
Cycle Time Not Cycle Time
1. Value Stream Map the 1. Map the Process
Process AND identify the 2. Do an FMEA on the Process
Process Wastes 3. Collect a sample of the
2. Look for the bottleneck and defectives and for every defect,
remove the bottleneck identify which process step
3. Look for wastes within the caused it’s failure AND which
process and remove them Failure Mode
4. Look for manual steps and 4. Use Pareto Charts to visualize
try to automate them the data
5. Brainstorm preventive and
detective controls
6. Implement controls

50
What to do if you want to reduce cycle time?

1. Value Stream Map the Process

AND identify the Process Wastes
2. Look for the bottleneck and
remove the bottleneck
3. Look for wastes within the process
and remove them
4. Look for manual steps and try to
automate them
51
 Value Stream Map* the Process
Indicate the Operating Hours of this process e.g.,
Start 8 am to 10 pm (Available Time)

Indicate the Arrival Volumes e.g., Vol per day = 30 (Demand)

Document also the End to End
Cycle Time. Note there is a
difference between Cycle Time Process Step You can add more information about the
& Processing Time.
Processing Time = process in the Process Step. Typical Info
FTE =
Processing Time is the time Automated? =
are Processing Time and Number of Full
taken to complete processing WIP(in tray) = Time Equivalent (Staffing)
one job without wait time,
interruptions etc.
Process Step
End to End Cycle Time or Processing Time =
Cycle Time is the time it start Decisio FTE =
Automated? =
for one job to move from start n Indicate the flow WIP(in tray) =
to end of a process. time (if possible)

Total Cycle Time – Total Indicate if the

Processing Time = Total Process Step flow is manual or
Wasted Time automated
Processing Time =
FTE =
Automated? =
WIP(in tray) =
*this is just a
process map with
more process End End
related info
52
 Example of a Simple Value Stream Map
Step 1 – Draw the Process Flow Card Application
Step 2 – Key in the Value Stream
Metrics.
Step 3 – Compute the Takt Time
Credit Check & Evaluate
300 Credit Card Applications
Processing Time = 2
Operating Hours = 8 hours
mins
Takt Time = (8 x 60) min / 300 =
FTE = 1
1.6 mins
Automated? = Semi
WIP (in-tray) = 50 Send Rejection Letter
Step 4 – Find out which step is Processing Time = 1
the Bottleneck No
min
Approved?
FTE = 1
Find the step which Automated? = Semi
Processing Time / # of FTE is
Yes WIP (in-tray) = 0
higher than Takt Time
Service Activation
That step will be the bottleneck
and it will likely have Work-in- Processing Time = 1
Progress stuck at the start of the min
Process FTE = 1
Automated? = Semi
Why is that? See the next few WIP (in-tray) = 0
slides.
End End

53
From a Cycle Time Perspective, the ideal is
Continuous Flow
Application
Form
Every Ten Every Ten
Minutes Minutes
10 mins 10 mins
One Job One Job
Arrives leaves
Application Into the the
Form
Value Stream Value Stream

E.g., 10 mins
10 mins

Application
Inter Arrival Time = 10 mins
Form

10 mins
10 mins

Evaluate & Recommend Approval Service Activation Print Credit Card

PT = 10 minutes PT = 10 minutes PT = 10 minutes PT = 10 minutes

From Start to End, this Value Stream will always take no more than 40 minutes
54
 How to achieve Continuous Flow in
Transactional Processes?
• This is an idealized scenario. No real process actually works like
this. However, in order to understand continuous flow, try to
understand this.
• Imagine if every 10 mins one job arrives.
• Imagine that the 1st step takes 10 mins to process the incoming
job.
• That means, as the 1st job is exiting the 1st step, the 2nd job is
arriving into the 1st step at exactly the same time. Hence, there is
no wait time for the 2nd job.
• Now the 1st job would have entered into the 2nd Step and that
would also take 10 mins to complete the processing AND SO ON.
• Hence, if you have such an idealized process, there will be no
bottleneck and consequently, you can get continuous flow.

55
 Why thrive towards Continuous Flow?

1. More Speed – because in a Continuous Flow, no Work-In-

Progress is ever kept Waiting.
2. Consistency – Cycle Time from Order to Delivery is kept
consistent and predictable because there is no excess
WIP in the Value Stream to cause Lead Time to fluctuate
or grow excessively
3. Less Cost – There is no waste of Transportation and
Excess Inventory hence the organization does not incur
cost associated with the management of them

56
 But your process is not idealized. Not every step
has the same processing time. How can we
achieve continuous flow?
Every 6 mins,
one job arrived Processing Time = 16
mins
FTE =2

Question – will this process step become a bottleneck?

Answer – So long as Processing Time / # of FTE is less than the inter arrival
time of jobs, it will not become a bottleneck

16 / 2 = 8 mins > 6 Mins

Hence, it will become a bottleneck

Question – How can we improve this process so that it ceases to be a

bottleneck?
a.Reduce the Processing Time to 12 mins or Less
b.Increase the # of FTE by 1 (16/3 = 5.3 mins)
c.Increase the interarrival time of jobs (generally out of your control) 57
 Note on FTE

• FTE stands for Full Time Equivalent

• It represents how many staff are manning the
process on a full time basis
• It is not the same as number of employee
• Say you employ 10 staff in your team. The team
handles three processes. Your staff divide their
time equally between these three processes
during the available time.
• Hence, the FTE per process is NOT 10 but 10/3 =
3.333.

58
 How do I know what is the interarrival time
of jobs for my process?
• In reality, no transactional process has a constant inter arrival time
of jobs (i.e., every X mins, one job arrives)
• However, let’s say for one day, your process has to complete
processing X jobs (let’s call this it’s Demand)
• And let’s say, your process operates for Y mins a day (let’s call this
available time)
• Then, you can say that practically, every Y / X mins, one job
arrives.
• This Y / X or Avail Time / Demand is called the Takt Time.
• In order words, if you know the Takt Time of your process, you can
identify the bottleneck of your process.

59
 Note on Available Time

• Available Time has nothing to do with the

staffing level of the process
• It refers to the time the process is
available for processing of jobs
• Hence, let’s say we have two Mac Donalds.
They are both opened 24 hours. The 1st
Mac Donalds has 10 staff, and the 2nd has
20. Their available time is the same – 24
hours. Staffing has nothing to do with it.

60
 Let’s use an example that looks complicated like
your own production process to convince you that
this works
P/T = 25 P/T = 5
mins mins
FTE = 2 FTE = 0.5

50% P/T = 2
P/T = 15
mins
mins
FTE = 0.5
FTE = 0.5
50%

P/T = 5 P/T = 2 P/T = 45

mins mins mins
FTE = 0.25 FTE = 0.5 FTE = 3

Start End End

61
So which step will be the bottleneck?
 Step 1 – Compute the Takt Time

Daily Demand = 60
P/T = 25 mins P/T = 6 mins
FTE = 2 FTE = 0.5 Avail Time = 10 hrs
TT = 10 TT = 10 Takt Time = 10 x 60 / 60 =
PT/FTE = PT/FTE = 10 mins

50% P/T = 2 mins

P/T = 15 mins FTE = 0.5
FTE = 3 TT = ?
TT = 10 PT/FTE =
PT/FTE =
50%

P/T = 5 mins P/T = 2 mins P/T = 45 mins

FTE = 1 FTE = 0.5 FTE = 3
TT = 10 TT = ? TT = ?
PT/FTE = PT/FTE = PT/FTE =

Start End End

62
 Step 2 – Look at the points where the flow is
separated and compute the Takt Time for each new
stream
Daily Demand = 60 / 2 = 30
P/T = 25 mins P/T = 6 mins
FTE = 2 FTE = 0.5 Avail Time = 10 hrs
TT = 10 TT = 10 Takt Time = 10 x 60 /30 =
PT/FTE = PT/FTE = 600 / 30 = 20 mins

50% P/T = 2 mins

P/T = 15 mins FTE = 0.5
FTE = 3 TT = 20
TT = 10 PT/FTE =
PT/FTE =
50%

P/T = 5 mins P/T = 2 mins P/T = 45 mins

FTE = 1 FTE = 0.5 FTE = 3
TT = 10 TT = 20 TT = 20
PT/FTE = PT/FTE = PT/FTE =

Start End End

63
Step 3 – Compute the PT/FTE for every step

P/T = 25 mins P/T = 6 mins

FTE = 2 FTE = 0.5
TT = 10 TT = 10
PT/FTE = 12.5 PT/FTE =12

50% P/T = 2 mins

P/T = 15 mins FTE = 0.5
FTE = 3 TT = 20
TT = 10 PT/FTE =4
PT/FTE = 5
50%

P/T = 5 mins P/T = 2 mins P/T = 45 mins

FTE = 1 FTE = 0.5 FTE = 3
TT = 10 TT = 20 TT = 20
PT/FTE = 5 PT/FTE = 4 PT/FTE = 15

Start End End

64
The 1st process step where the PT/FTE is greater than the Takt Time will be the
bottleneck. All subsequently steps where PT/FTE is greater than the Takt Time
will in turn become bottlenecks when the bottleneck higher up in the flow
becomes unclogged.

P/T = 25 mins P/T = 6 mins

FTE = 2 FTE = 0.5
TT = 10 TT = 12
PT/FTE = 12.5 PT/FTE =10

50% P/T = 2 mins

P/T = 15 mins FTE = 0.5
FTE = 3 TT = 20
TT = 10 PT/FTE =4
PT/FTE = 5
50%

P/T = 5 mins P/T = 2 mins P/T = 45 mins

FTE = 1 FTE = 0.5 FTE = 3
TT = 10 TT = 20 TT = 20
PT/FTE = 5 PT/FTE = 4 PT/FTE = 15

Start End End

65
A simple and quick way to improve this process is to take staff from those
steps that are have too much FTE (those in Green) and assign them to the
bottlenecks

P/T = 25 mins P/T = 6 mins

FTE = 2 FTE = 0.5
TT = 10 TT = 12
PT/FTE = 12.5 PT/FTE =10

50% P/T = 2 mins

P/T = 15 mins FTE = 0.5
FTE = 3 TT = 20
TT = 10 PT/FTE =4
PT/FTE = 5
50%

P/T = 5 mins P/T = 2 mins P/T = 45 mins

FTE = 1 FTE = 0.5 FTE = 3
TT = 10 TT = 20 TT = 20
PT/FTE = 5 PT/FTE = 4 PT/FTE = 15

Start End End

66
 Generic Ways to remove a bottleneck?

• Reducing Processing Time to synchronize with

Takt Time
– Simplification
– Automation
– Line Balancing
• Increase FTE
– Optimize the Staff allocation by identifying
process steps that are over staffed and
assigning those excess staff to bottlenecks
• Reduce the Demand (normally not possible)
67
 What is Line Balancing?
1. Plot the 2. Draw the line that
processing time for represents the Takt
every process step Time
Each Process Step 3. Identify the
is separated into its Bottleneck Step
Work Elements.
4. If possible shift Work
E.g., Update Elements from one
Records = Boot Up process step to another
System (WE1) + so that the entire line is
Key in the Record “balanced”
(WE2) + Save the
Record (WE3)

68
 Thinking about Waste

• A simple way of thinking about improving cycle time –

finding and eliminating waste. So the task is two fold:
– We must know how to find waste
– We must know how to reduce or eliminate waste
• What is Waste?
– Anything other than the absolute minimum of
material, machines, and manpower required
to add value to the product.
– Everything that is not value-added is waste

69
 Value and Non Value Added Activities
Whether an activity is value adding depends on how the customer sees it
Typically, over 80%

Necessary but Non

Value Added Activities Non Value Added
Value Added

Activities that do not add Activities that do not add

Activities core to the
value from customer’s value from customer’s
What is it? product or service that
perspective but is perspective and is not
the customer wants
required by the business willing to pay for

Does this activity:

Is the Product or Service 1. Does this activity create a
1. Reduce Financial Risk?
Transformed by this activity? known type of waste?
2. Comply with some
How to see regulation?
2. Is the customer willing to pay
it? Does it add a feature, function or for it?
3. Comply with some internal
form to the product or service that 3. Is it rework or corrective
control?
the customer is willing to pay for? action?
4. Does anything bad happen if
we remove this step?

Try to separate this activity from the

Action Mode Increase the Quality of this Step main flow so that the customer is Reduce and ultimately eliminate it
not impact by the time it consumes

70
 How much Waste in Process?
Application Application
Application Application
Form
Application Form
Application
Form
Application Form
Application
Form Form
Form Form
Out Box In Box
Application for
Credit Card = Wait for Collection = Wait in In Box =
Transport to Mail Room = Sorting and Wait in Transport =
10 mins 60 mins 60 mins
20 mins Mail Room = 180 mins 20 mins

Application
Application
Form
Application
Form
Application Application
Form Application
Form Form
Application
Form
Application
Form
In Box Form
Out Box
Wait in In Box = Transport = Sorting and Wait in
60 mins Transport = Credit Check
20 mins Mail Room = 180 mins Wait in Out Box =
20 mins 10 mins
60 mins Application
Application Application
Application Form
Application
Form
Application Form
Application
Form
Application Form
Form Form
Form
In Box
Out Box
Wait in In Box =
Approval = Wait in Out Box = Transport = Sorting and Wait in 60 mins
2 mins Mail Room = 180 mins Transport =
60 mins 20 mins
20 mins

Application
Application
Form
Application
Form
Application
Form
Form
Out Box
Batch Mailing Card Activation
To Customer = 30 mins Sorting and Wait in Transport = Wait in Out Box = Print Card =
Mail Room = 180 mins = 3 mins
20 mins 60 mins 10 mins

Two Days Later Customer Gets Card

Spoilage =
5%
71
 Which Activity is Value added?
Application Application
Application Application
Form
Application Form
Application
Form
Application Form
Application
Form Form
Form Form
Out Box In Box
Application for
Credit Card = Wait for Collection = Wait in In Box =
Transport to Mail Room = Sorting and Wait in Transport =
10 mins 60 mins 60 mins
20 mins Mail Room = 180 mins 20 mins

Application
Application
Form
Application
Form
Application Application
Form Application
Form Form
Application
Form
Application
Form
In Box Form
Out Box
Wait in In Box = Transport = Sorting and Wait in
60 mins Transport = Credit Check
20 mins Mail Room = 180 mins Wait in Out Box =
20 mins 10 mins
60 mins Application
Application Application
Application Form
Application
Form
Application Form
Application
Form
Application Form
Form Form
Form
In Box
Out Box
Wait in In Box =
Approval = Wait in Out Box = Transport = Sorting and Wait in 60 mins
2 mins Mail Room = 180 mins Transport =
60 mins 20 mins
20 mins

Application
Application
Form
Application
Form
Application
Form
Form
Out Box
Batch Mailing Card Activation
To Customer = 30 mins Sorting and Wait in Transport = Wait in Out Box = Print Card =
Mail Room = 180 mins = 3 mins
20 mins 60 mins 10 mins

Two Days Later Customer Gets Card

Spoilage =
5%
72
 % Value Added Time

• Total Time = 4225 Minutes*

• Value Adding Time = 23 Minutes
• % Value Adding = 0.5%
One Business’ Reaction
• Waste = 99.5%
“When we stepped back and evaluated our processes to
determine what steps created customer value, we were
shocked to discover that almost 99% of our work adds no
value for customers. In fact we were so demoralized by this
finding that we consider adjusting this criteria”

VP of a Bank

*Actually it is more because we have not even counted the time lost after the work day has ended

73
 Production Wastes
7 Types of Waste + 2 Service Examples Manufacturing Examples
Errors and Rework
Defects Work not meeting Requirements Scrap, Defects
Missing Info
Job waiting in In-Box, Out-Box
Equipment Downtime
Waiting Equipment Downtime
Waiting due to Batch Processing
Waiting for Info, Approval etc

Unnecessary Steps Processing Step that adds any feature that the
Over Processing Multiple Handoffs customer does not value

Walking
Unnecessary Motion Searching for Files
Walking to get parts

Excessive Raw Material, WIP and Finished

Inventory Excessive Backlog / WIP
Goods that has not been sold yet

Unnecessary Transportation Routing around the office Routing around the Factory

This does not really happen in Service Producing more than what is require to meet
Over Production Environments demand

Abilities Not using a staff to his or her full potential Not using a staff to his or her full potential

Customer’s Time Queues, Asked to redo a task due to rework Delays due to Stock-out

74
 Value Added Analysis
Steps that are essential
even when everything
• Value-added steps: works correctly move Steps that would not be needed
– Customers are down the left side. if everything worked right the
first time move horizontally
willing to pay for it across the right side.
– It physically
changes the Value-Added Non-Value-Added
Yes
product
– It’s done right the Take Original
Copier Yes
in Use?
Wait?
No
Leave
first time
No

Yes
• Non-value-added Place No Glass Clean
Original Dirty?
steps:
– Not essential to Select
produce output Size Think of how to remove all these
– Does not add Select
non value adding steps
value to the Orientation
output
No No No
Select Find Box Find Open
Number Paper? Knife?
Paper Open? Knife Box

Yes Yes Yes

Yes Paper No Find

Loaded? Help 75
Step 9. Process Analysis (CTQ is Quality)

76
 What to do if the CTQ is Quality

1.Map the Process

2.Do an FMEA on the Process
3.Collect a sample of the defectives and for
every defect, identify which process step
caused it’s failure AND which Failure Mode
4.Use Pareto Charts to visualize the data
5.Brainstorm preventive and detective
controls
6.Implement controls

77
Decision symbol can be
Yes / No type OR it can
be IF type
No Step 1 – Map the Process
Write the Triggering Event here –
Yes Start e.g., Customer applies for Credit
Card

If …..
Process Steps – use a Verb Noun
format e.g., Check Creditworthiness
Process Step
If …..

Decision – State the

Decisio
Question e.g., Credit Process Step
Worthy? n

Process Steps – DO
Process Steps – use a NOT write an essay in
Verb Noun format e.g., Process Step the box ! Keep it Simple.
Issue Card Noun Verb. That’s it.

Write the End Event

here – e.g., Customer
receives Credit Card End End
78
 Step 2 – Conduct an FMEA on the Process?

• A failure modes and effects analysis (FMEA), is

a procedure in operations management for analysis of
potential failure modes within a process for classification by
the severity and likelihood of the failures.
• A successful FMEA activity helps a team to identify potential
failure modes based on past experience, enabling the team
to design those failures out of the process.
• Failure modes are any errors or defects in a process
especially those that affect the customer, and can be
potential or actual. 
• Effects analysis refers to studying the consequences of
those failures.

79
 Template of an FMEA
S O D R Res-
Process Potential Failure Potential Failure Actions
E Potential Causes C Current Controls E P ponsi- Actions Take
Step Mode Effects Recommended
V C T N bility
 FMEA – How to Conduct

S O D R Res-
Process Potential Failure Potential Failure Actions
E Potential Causes C Current Controls E P ponsi- Actions Tak
Step Mode Effects Recommended
V C T N bility

 List each process step

 Identify potential failure modes for each process step

 Identify potential effects of each failure and rate its
severity
 Identify potential causes of the effects
and rate their likelihood of occurrence
Rate - in consideration of given design or process 
controls - the likelihood of detection of each failure
mode
Multiply the three numbers to determine the risk of 
each failure mode (RPN = Risk Priority Number)
Identify ways to reduce or eliminate risk associated with high RPNs 
 Ratings
Severity Rating Meaning
1 No effect
2 Very minor (only noticed by discriminating customers)
3 Minor (affects very little of the system, noticed by average customer)
4/5/6 Moderate (most customers are annoyed)
7/8 High (causes a loss of primary function; customers are dissatisfied)
9/10 Very high and hazardous (product becomes inoperative; customers angered; the failure may result
unsafe operation and possible injury)
Occurrence Rating Meaning

1 No effect
2/3 Low (relatively few failures)
4/5/6 Moderate (occasional failures)
7/8 High (repeated failures)
9/10 Very high (failure is almost inevitable
Detection Rating Meaning
1 Almost certain
2 High
3 Moderate
4/5/6 Moderate - most customers are annoyed
7/8 Low
9/10 Very remote to absolute uncertainty
 Example of an FMEA
Process Step Potential Potential SEV Potential OCC Current DET RPN Actions Recommend
Failure Mode Failure Causes Controls
Effects
Application for Applicant did Unable to 2 Customer 2 None 1 4 Bring the application process online. Create an
Credit Card not supply perform Oversight automated check to prevent application
identification Credit submissions that do not contain a valid national
number Check identification number

Assignment of Assigned a Risk that 9 Operator 3 None 5 135 Bring the application process online. Create an
Credit Limit higher Credit customer Manual automated credit limit assignment algorithm
Limit than is may not be Error based on the verified income info of the client
allowed able pay
based on the up on loan
customer's amount
income group
Send Credit Credit Card Someone 9 Wrong 1 None 9 81 Install a process such that the card cannot be
Card via Mail posted to the else may Address used unless activated by the client with an
to Client Wrong try to use was keyed identity check process
Address the Card into our
system

Credit Card Someone 9 Delivery 1 None 9 81 Install a process such that the card cannot be
lost in the else may agent's used unless activated by the client with an
Mail try to use process identity check process
the Card failure
Step 3 - Collect a sample of the defectives and for
every defect, identify which process step caused
it’s failure AND which Failure Mode

Job Why is it Which Failure Mode Which Process

Identifier defective? Step?
1232 Inaccurate Data Supplier supplied the 1
Value wrong data
3798 Missing Data Value Analyst failed to key in the 4
data
4589 Missing Data Value Analyst does not have all 5
info to populate the value
9876 Inaccurate Data Data accurate in the 3
Value database but failed to flow
to product

This type of exercise makes the assumption that the team is able to
assign the correct failure mode for every defective. This is not always
the case and in the event the team is not sure, it is much better not
to use guesswork.

84
 Step 4 – Use Pareto Charts to visualize the
data
• A Pareto chart is a type of chart that
contains both bars and a line graph,
• The left vertical axis is
the frequency of occurrence.
• The right vertical axis is the
cumulative percentage of the total
number of occurrences.
• In this example, in order to lower
the amt of late arriving by 80%, it is
sufficient to solve just the first 3
issues.
• In our case, we will use this to
highlight, the most common failure
modes in the process

85
 Step 4.1 – How to do Pareto Analysis in
EXCEL
Defective Job Failure Mode Number Count of Failure Mode Number
Number Failure Mode Number Total
374 12 1 6
444 10 2 11
197 12 3 10
454 12 4 60
206 12 5 32
171 12 6 30
138 5 7 9
326 5 8 10
303 12 9 12
172 5 10 11
364 5 11 8
366 6 12 37
217 12 13 9
421 12 14 7
225 12 15 6
452 14 16 11
262 16 17 7
363 5 18 9
406 5 19 10
20 6
Grand Total 301
1. For every defective job,
assign the Failure Mode
that caused that defective 2. Use Pivot Table function in EXCEL to
to occur create this summary table

86
 Step 4.1 – How to draw a Pareto Chart in
EXCEL
Count of Failure Mode Number Failure Mode Number Total % of Total Cummulate %
Failure Mode Number Total 4 60 19.93% 19.93%
4 60 12 37 12.29% 32.23%
12 37 5 32 10.63% 42.86%
5 32 6 30 9.97% 52.82%
6 30 9 12 3.99% 56.81%
9 12 2 11 3.65% 60.47%
2 11 16 11 3.65% 64.12%
16 11 10 11 3.65% 67.77%
10 11 3 10 3.32% 71.10%
3 10 8 10 3.32% 74.42%
8 10 19 10 3.32% 77.74%
19 10 18 9 2.99% 80.73%
18 9 7 9 2.99% 83.72%
7 9 13 9 2.99% 86.71%
13 9 11 8 2.66% 89.37%
11 8 17 7 2.33% 91.69%
17 7 14 7 2.33% 94.02%
14 7 1 6 1.99% 96.01%
1 6 20 6 1.99% 98.01%
20 6 15 6 1.99% 100.00%
15 6 Grand Total 301 100.00%
Grand Total 301

3. Sort the Table into 4. Compute the % Total and

Descending order of Cumulative %
Freq Count 87
 Pareto Analysis in EXCEL
Failure Mode Number Total % of Total Cummulate %
4 60 19.93% 19.93% If we can If we can
prevent these
12 37 12.29% 32.23% prevent these
4 failure 12 failure
5 32 10.63% 42.86% modes, 53% modes, 80%
6 30 9.97% 52.82% of the of the
9 12 3.99% 56.81% defectives will defectives will
go away
2 11 3.65% 60.47% go away
16 11 3.65% 64.12%
10 11 3.65% 67.77%
3 10 3.32% 71.10%
8 10 3.32% 74.42%
19 10 3.32% 77.74%
18 9 2.99% 80.73%
7 9 2.99% 83.72%
13 9 2.99% 86.71%
11 8 2.66% 89.37% The main point of Pareto Analysis is
17 7 2.33% 91.69% understand how to prioritize our
14 7 2.33% 94.02% energy. It is obvious here that we
1 6 1.99% 96.01% should start with Failure Modes 4, 12,
20 6 1.99% 98.01% 5 and 6 and improve those first as they
offer the biggest “bang for the buck”
15 6 1.99% 100.00%
88
 5. Brainstorm preventive and detective
controls
• The best strategy when dealing with
Failure Modes is to prevent them from
happening all together. That is called
Preventive Control.
• If Preventive Control is not possible, the
second best strategy is to detect them
when they occur so that the defectives can
be removed from the production line so
that they do not go to the customer. That
is called Detective Control.
89
 Examples of Preventive Controls from
Everyday life
Examples from everyday life:

Failure Mode – Gas is left Failure Mode – Client forgets

Failure Mode – Person
running without a flame to take the dispensed cash
forgets to turn off the tap
from the ATM
Preventive Control - if the Preventive Control – the
flame does not burn, the Preventive Control – you
Sink will not overrun due to
gas is switched of cannot take your card back
the presence of the hole
automatically unless you take the cash first
near the top of the Sink
OPTIONAL TOPIC –
ORGANIZE YOUR ROOT
CAUSES
91
 Cause-and-Effect Diagram
Narrowly defined
High # of High # of problem forms
grammar errors misspelled words head of the fish;
causes listed on
Reporters & editors do
not use spellchecker
the diagram
Reporters & Copy editors busy
editors do not tracking down facts
R
kn epo
T
jo hin
k
S
us pel should potentially
ow rt b l
review grammar e t o ’si t e r-f che contribute to this
ab rs ca c ri e c k
T ou & e tc op nd er
jo hin
b k
to i t’ s
Reporters unaware of ts d
pe itor
h y
er d
ro es
ly no
t problem.
correct grammar llc s
Arrows indicate ca c
tc op Reporters have he do
c k ’tn
rs k ’s
h y
the direction of er d
ro es
poor typing skills er

potential cause-
rs k’
s High Number of
and-effect. “Major Errors”
Results of accuracy checks unclear
Hard to detect errors

rd
da
ce e

n tan
cs
ur us

io
f o de
s

ifi
at

ti o s
so s

in vi

ec
rm

ta of
y er

ct pro

in t
sp
o
an rt

no ack
en n
m epo

rre s

g
s
co ce

L One “spine” leads

r
cu rte
R

t
in ur

do epo
So

Accuracy checks not done

into head
R

Causes grouped by
ot

(problem statement).
ks n

fa ot
e c rs

ng n

relationship to
s
ct
o

ki ers
c h it

Contributing causes
g ed

ec rt
ch epo

each other.
in e
do -lin

are arranged on
R
In

Potential causes at smaller and smaller

High # of wrong names, lower level would “bones.”
numbers and facts contribute to cause
92
at next level up.
 The “Might Cause” Check

– Helps to confirm the items listed are potential

causes
– Helps to check relationships between items

<smallest bone>

might cause
<next largest bone>

which might cause

<next largest bone>

which might cause

<main bone>

which might cause

<head (problem statement)>
93
 Might Cause Example

Thinking it’s copy editor’s job to catch errors

<smallest bone>

Reporters and editors to not use spellchecker

might cause
<next largest bone>

which might cause

<next largest bone>

High number of misspelled words

which might cause
<main bone>

High number of Major Errors

which might cause
<head (problem statement)>
94
Step 10 & 11 – Improvement Ideation &
Implementation

95
 Creativity Techniques
Thinking out of the box
Creativity techniques force us out of our
usual ways of looking at things.

´Normal´ Thinking Thinking out of the box

Starting with the problem, Starting with something that has

what ideas do we have to nothing to do with the problem what
solve it? ideas do we have to solve it?
 Brainstorming
Sequence:
1. Title the brainstorming
2. Set time (20´)
3. Conduct brainstorming involving
all team members
4. Clarify questions and summarize
ideas
Brainstorming rules:
– Visualize the ideas (use of flip
chart, post-it notes, etc.)
– Don’t critique others’ points
– It’s allowed to build on the ideas
of others
– Facilitate
 Brainwriting 6-3-5
The name Brainwriting 6-3-5 comes from the process of having 6 people write 3
ideas in 5 minutes. Each person has a blank 6-3-5 worksheet.

Sequence
1. Write the problem statement at the top of the worksheet.
2. Write 3 ideas on the top row of the worksheet in 5 minutes.
3. After 5 minutes (or when everyone has finished writing) pass the
worksheet to the person on your right.
4. Add three more ideas.
5. Continue until the worksheet is completed. There will now be a total of
108 ideas on the 6 worksheets.

Problem Statement: How to...

  Idea 1 Idea 2 Idea 3
1      
2      
3      
4      
5      
6      
 Pictures as Idea Triggers
Start with this creativity technique after the most obvious ideas
were explored, e.g. by brainstorming or brainwriting 6-3-5.

Sequence:
1. Put the problem statement on view.
2. Show a picture unrelated to the problem
• e.g. a painting, a poster, an advertisement
1. Write down participants´ associations to this picture
• elements, attributes, emotions,
1. Use those associations as triggers to find potential
problem solutions.

Variation:
Use a lexicon instead of a picture: Open the lexicon on any page, choose
spontaneously a word and brainstrom associations to this word. Use the
associations as triggers to find ideas related to your problem.
 Analogies
Sequence:
1. Find an agreed problem definition in terms of
• 'How to make X'. 'How to prevent Y', 'How to speed up Z‘
1. Generate a list of items (people, situations, objects, processes,
actions, places, etc.) that is 'like' it in some way
• e.g. analogies to 'making X‘: having a baby, making a pudding, a
robot car factory, ...etc.
1. Pick one of these analogies that seems interesting (preferably
where the problem and analogy are from different domains)
• e.g. a biological analogy for a mechanical problem.
1. Describe the analogue
• how it works, what it does, what effects it has, how it is used
• size, position, etc.
1. Use this description to suggest ideas relevant to your problem
• Does the analogue have features you can use directly? Do the
differences suggest other ways of looking at your problem?
 Morphological Chart
The morphological chart is based on an analysis approach: the problem
is divided into smaller units. For each subproblem potential subsolutions
are developed, which are combined to create an entirely new solution.
Sequence:
1. List the attributes of the problem.
2. Below each attribute, brainstorm as many alternates as you
can think of
3. When completed, make many random runs through the
alternates, picking up a different one from each column and
assembling the combinations into entirely new forms of your
original subject.

Example: Developing a new pen

Shape / Cylinder  Material Cap Ink Source
Faceted Metal Attached Cap No Cartridge
Square Glass No Cap Permanent
Beaded Wood Retracts Paper Cartridge
Sculptured Paper Cleaning Cap Cartridge made of ink
 Grouping the Ideas – Affinity Diagram
After collecting lots of ideas with the help of creativity techniques, you
should group them for a better overview.

How to develop an Affinity Diagram Frequency Media

(compare VOC) Once a
every month
1. Transfer data onto cards or post-it 14 days
monthly
Intranet
notes news-
E-mail
paper
2. Group the cards to find the “affinity”
people involved
3. Label the groups of cards
4. Optional: group the clusters employees Management
field
5. Draw the diagram service shop exe-
adminis- floor cutives
6. Combine ideas into solutions tration

Example: Improve communication process

To avoid long discussions, try the following:
Grouping the ideas the participants are not
allowed to talk to each other.
 A Solution Selection Matrix
t
en
s em
es pl e
Specific en Im c or
Problem Root Solutions it v to S n
Causes (What) Tasks fe
c se ost t al t io
(How)
f
E X a
E X C = To Ac

SCALE: 1-None 2-Somewhat 3-Moderate 4-Very 5-Extreme

– Shows relationship of problem statement, root causes and proposed solutions

– Evaluates which solution(s) to implement by rating the effectiveness of the
solution and the ease of implementing (feasibility) the selected tasks

103
Step 12 – Before & After Comparison

104
 How to do it?

• Go to the lesson in 5. Measure the

Baseline
• The steps are exactly the same
• Continue to collect data after the
improvements have been implemented
• Monitor to observe if there is any change
in the % Defective

105
 Before & After Comparison

Before

After

Important thing to understand – the average has to shift. If the average

does not shift in a convincing way, it the process has not improved 106
Step 13 – Plan to Control the Gains

107
 Standardization Core Principle
– Nothing happens on a reliable, sustained basis unless we build a
system to cause it to happen on a reliable, sustained basis.

• Standardization is what allows high quality to happen on a reliable, sustained

basis.

108
 Standardization

– Making sure that important elements of

a process are performed consistently in
the best possible way
– Changes are made only when data
shows that a new alternative is better
– Documentation is up to date and used
– What images come to mind when you
think of process standardization?

109
 Benefits of Standardization

• Standardization helps us compete more

successfully in the marketplace by providing:
– Increased reliability
– Reduced costs
– Improved employee performance
– Increased safety
– Processes that remain in control
– Continuous improvement
– Flexible practices that allow for quick
response to customer needs
110
 Standardization = “Standard Practices”

• A standard practice is…

– a definition of a work method wherein all variables of
the method have been specified in detail
– a written agreement between the worker and the
company regarding how the job will be done.
– a controlled document that can have different names
depending on your business
– Procedure, Work Instruction, Standard Operating
Practice, etc.
– In TR, we call them Method of Work or MOW

111
 Uses for Standard Practices

1. To reduce variation among individuals or groups (and so

make process output more predictable)
2. Provide “know-why” for operators and managers now on
the job
3. Provide a basis for training new people
4. Provide a trail for tracing problems
5. Provide a means to capture and retain knowledge
6. Give direction in the case of unusual conditions

112
 Do this - Documenting the Standard
Practice / MOW
– Update detailed process maps with
improvement actions
– Determine which process step(s), if any,
require an additional written work instruction
to ensure standardization
– Define the written work at the level that makes
sense for your situation
– Use your business’s approved format to
document the standard practice. Typical
formats include:
– unique document name/number
– process name, scope and purpose
– responsibility and description of process steps
– revision number/date
– document approver
113
Step 14 – Celebrate

114
 How to Celebrate

1. Get the Team together

2. Buy a lot of Beer
3. Buy some pizza etc
4. If possible, get drunk

Note – Responsible adults do not drink &

drive

115
END OF TRAINING

116