Day 3 - Quality Management System

Tools and Implementation

 Quality Management Tools

1. Process Flow Chart

2. Run Chart / Control Chart
3. Cause-and-Effect Diagram (Fishbone diagram)
4. Cause-and-Effect Martix
5. Check Sheet and Histograme
6. Pareto Analysis
7. Scatter Diagram
8. Failure Modes & Effects Analysis (FMEA)

Copyright 2009 John Wiley & Sons, Inc. 2-2

 1. Process Flow chart

Copyright 2009 John Wiley & Sons, Inc. 2-3

 2. Run Chart / Control Chart

Performance

Time
 3. Cause-and-Effect Diagram

 Cause-and-effect diagram (“fishbone” diagram)

◼ chart showing different categories of problem causes

Copyright 2009 John Wiley & Sons, Inc. 2-5

 4. Cause-and-Effect Matrix

 Cause-and-effect matrix
◼ grid used to prioritize causes of quality problems

Copyright 2009 John Wiley & Sons, Inc. 2-6

 5. Check Sheets and Histograms

Copyright 2009 John Wiley & Sons, Inc. 2-7

 6. Pareto Analysis

 Pareto analysis
◼ most quality problems result from a few causes

Copyright 2009 John Wiley & Sons, Inc. 2-8

 7. Pareto Chart

Copyright 2009 John Wiley & Sons, Inc. 2-9

 8. Scatter Diagram

Copyright 2009 John Wiley & Sons, Inc. 2-10

Case: shortening telephone waiting time…

• A bank is employing a call answering service

• The main goal in terms of quality is “zero waiting time”

- customers get a bad impression
- company vision to be friendly and easy access

• The question is how to analyze the situation and improve quality

 The current process

Customer Receiving
Operator
A Party

Customer
B How can we reduce waiting
time?
 Fishbone diagram analysis

Absent receiving Working system of

party operators

Absent Too many phone

calls
Out of office Lunchtime

Not at desk Absent

Makes
custom
Not giving er wait
receiving party’s Does not
coordinates Lengthy talk understand
Does not know customer
Complaining organization
well
Leaving a Takes too much
message time to explain

Customer Operator
Reasons why customers have to wait
(12-day analysis with check sheet)

Daily Total
average number
A One operator (partner out of office) 14.3 172
B Receiving party not present 6.1 73
C No one present in the section receiving 5.1 61
call
D Section and name of the party not given 1.6 19
E Inquiry about branch office locations 1.3 16
F Other reasons 0.8 10
29.2 351
 Pareto Analysis: reasons why customers have
to wait

Frequency Percentage

300 87.1%

250 71.2%
200
49%
150
100

0%
A B C D E F
 Taking lunches on three
different shifts

Ideas for Ask all employees to

leave messages when
improvement leaving desks

Compiling a directory
where next to
personnel’s name
appears her/his title
 Results of implementing the
recommendations

Before… …After
Frequency Percentage Frequency Percentage

100%

300 87.1 300

%
71.2
% Improvement
200 200
49
%

100 100 100%

0% 0%

A B C D E F B C A D E F
How Can We Monitor Quality?
In general, how can we monitor quality…?

By observing
variation in
output measures!

1. Assignable variation: we can assess the cause

2. Common variation: variation that may not be possible to
correct (random variation, random noise)
Statistical Process Control (SPC)

Every output measure has a target value and a level of “acceptable” variation
(upper and lower tolerance limits)

SPC uses samples from output measures to estimate the

mean and the variation (standard deviation)

Example
We want water bottles to be filled with 1 L ± 0.05 L

Question:
How do we define the output measures?
 In order to measure variation we need…

The average (mean) of the observations:

N
1
X =
N
x
i =1
i

The standard deviation of the observations:

 (x i − X )2
 = i =1
N
Average & Variation example

Number of chicken slices per pizza: 25, 25, 26, 25, 23, 24, 25, 27

Average: 25
Standard Deviation: 1.12

Number of chicken slices per pizza: 25, 22, 28, 30, 27, 20, 25, 23

Average: 25
Standard Deviation: 3.01

Which pizza would you rather have?

 Accuracy and Consistency

We say that a process is accurate if its mean is close to

the target T.

We say that a process is consistent if its standard deviation

is low.
 Capability Index (Cpk)

Cpk is a process capability index that determines how much a

process can produce. Many people use Process Capability as a
computation tool to estimate the output of a product they are
manufacturing. It helps the manufacturers estimate the potential
production and manage the resources to get the best results.

It’s analogous to the mathematics terms mean and average value as

well as standard deviation. However, it varies in that it employs a
control chart analysis to assess the statistical control of the system.
The assumption that the measurements are regularly distributed is
required by Cpk.
•upper specification limit = USL •lower specification limit = LSL
 Capability Index (Cpk)

It shows how well the performance measure fits the design

specification based on a given tolerance level

A process is k capable if

X + k  UTL and X − k  LTL

UTL − X X − LTL
1 and 1
k k

•upper specification limit = USL •lower specification limit = LSL

 Capability Index (Cpk)

Another way of writing this is to calculate the capability index:

 X − LTL UTL − X 
C pk = min  , 
 k k 

Cpk < 0 The process mean has violated one of the specification boundaries.

Cpk < 1 The process mean is within specification limitations, but a portion of the
manufacturing output has exceeded them (System not capable!)

Cpk >= 1 The process mean is centred and within the specification limits
Example 1: Capability Index (Cpk)

X = 10 and σ = 0.5
LTL = 9
UTL = 11

 X − LTL UTL − X 
C pk = min  , 
 k k 
 10 − 9 11−10  LTL X UTL
C pk = min  or  = 0.667
 3  0.5 3  0.5 
Example
Example
 Risk Based Maintenance (RBM)
Failure Modes & Effects Analysis (FMEA)

Presentation_ID © 2008 Cisco Systems, Inc. All rights reserved. Cisco Confidential 30
Linking Causes to Effects
One to One, One to Many, Many to One, or Many to Many

Cause 1 Effect 1
1:1
1:M
Cause 2 Effect 2

Effect 1
Cause 1
Effect 2
M:1

Cause 1
Effect 1
Cause 2

31
FMEA Process
2
1 Start with For each step,
the process brainstorm
map potential failure
modes and effects

Determine
severity
Determine the
3 potential causes Determine
to each failure likelihood of
mode occurrence

Determine
4 Evaluate current detectability
controls

5
6

Identify actions Determine RPN

32
Calculations
Risk Priority Number

The Risk Priority Number (RPN) identifies the greatest areas of concern.
(1) Severity rating
(2) Occurrence rating
(3) Detection rating

▪ Severity
A rating corresponding to the seriousness of an effect of a potential
failure mode. (scale: 1-10. 1: no effect on the customer, 10: hazardous
effect)
▪ Occurrence
A rating corresponding to the rate at which a first level cause and its
resultant failure mode will occur over the design life of the system,
over the design life of the product, or before any additional process
controls are applied. (scale: 1-10. 1: failure unlikely, 10: failures
certain)
▪ Detection
A rating corresponding to the likelihood that the detection methods or
current controls will detect the potential failure mode before the
product is released for production for design, or for process before it
leaves the production facility. (scale: 1-10. 1: will detect failure, 10:
almost certain not to detect failures)
33
Calculations - Risk Priority Number (RPN)

Severity x Occurrence x Detectability =

Risk Priority Number (RPN)

For a given potential failure mode, how bad

the outcome is multiplied by how likely it
would actually happen multiplied by what
things are in place today to prevent or notice
it before it happens.

34
 FMEA Example
Purchasing Requisition to Purchase Order

Presentation_ID © 2008 Cisco Systems, Inc. All rights reserved. Cisco Confidential 35
 Example
Purchasing Dept.
customer

Complete
Send PR to Incorrect
Focus
Team

Purchase Correct and Receive

Start Requisition
Purchasing PR
Send Back Goods
Dept. Returned
(PR)
Department

No
Purchasing

Yes
Receive Form Complete Send P.O.
PR Correct P.O. To supplier
Supplier

Confirm Complete
Ship
receipt of Commit
Goods
P.O. Process

36
Example
Purchasing Dept.

From the
process map, Brainstorm the
list the process various ways the
steps step could fail

37
Example
Purchasing Dept.

Determine the Determine the

potential effects severity ranking
using the scale

38
Severity Rankings

39
Example
Purchasing Dept.

Determine the Determine how

potential causes likely the failure
would occur due
to this cause

40
Occurrence Rankings

41
Example
Purchasing Dept.

Identify what controls Determine how likely

or measures are the controls in place
currently in place will detect or prevent
the failure mode from
occurring

42
Detectability Rankings

43
Example
Calculate the RPN

Severity Occurrence Detectability RPN

5 x 4 x 3 = 60

44
 Example Occurrence Reduced
Purchasing Dept. from 4 to 2.
PRN cut in half.

FMEA owner & Recalculate

Assign team update the RPN after
Brainstorm specific the document
potential actions actions are
owners as actions are complete
that will lower the complete
RPN

45
FMEA Tips
▪ No absolutes rules for what is a high RPN number.
Rather, FMEA often are viewed on relative scale (i.e.,
highest RPN addressed first)
▪ It is a team effort
▪ Motivate the team members
▪ Ensure cross-functional representation on the team
▪ Treat as a living document, reflect the latest changes
▪ Develop prioritization with the process owners!
▪ Assign an owner to the FMEA; ensure it is periodically
reviewed and updated

48
50