3. m phase
DESCRIPTION
kTRANSCRIPT
![Page 1: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/1.jpg)
Jim McKiernan
Operational Excellence in Biopharmaceuticals3. Introduction to Lean, The Measure Phase
October 2014
![Page 2: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/2.jpg)
IntroductionsCourse Outline and TimingsDate Event Content Timing
16 Sept 11-1 Face to Face
Intro, Business Context, Operational Excellence Overview, LSS Structure, DMAIC, Project selection and initiation
2 hours
2 Oct 11-12 Face to Face
The Define phase, Benefit definition 1 hour
2 Oct 12-13 Face to Face
Measure phase, Process mapping, data types and measurement 1 hour
21 Oct 11-13 Face to Face
Intro to statistics, Lean; 5 lean principles, 8 wastes, Kaizen, 5S 2 hours
27 Oct 11-12 Webinar Analysis Phase, Six Sigma, Root cause analysis 1 hour
10 Nov 11-12 Webinar Data analysis, Design of experiments 1 hour
17 Nov 11-12 Webinar Improve phase
25 Nov 11-13 Face to Face
Change management, team dynamics and team leadership, Control phase, project follow-up and ensuring sustainability
2 hours
28 Nov 9-10 Webinar Test preparation, course review and open questions 1 hour
16 Dec tbc Test Multiple choice, open-book 2 hours
![Page 3: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/3.jpg)
Content
3
Measure Phase
Process Mapping
Data Types
![Page 4: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/4.jpg)
Measure – The Second of the DMAIC Phases
4
Understand and Characterize the Process
Process Maps:SIPOCValue StreamsSwim LanesSpaghetti Diagrams
Initial Ideas:BrainstormingFishbone Diagrams
Graphical Analysis:Charts and GraphsVisualization
Tools:Measurement SystemsCharts and GraphsVisualization
Process
0
5
10
15
20
25
30
35
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb
Month
# R
equi
red
Cha
nges
Might already have some of these
Control Plan
ProceduresImprovement
Activities
Preliminary
Analysis
Measurement
System
Process
Maps
C&E
Matrix
FMEA
Measurement
System Analysis
%R&R, %P/T
Calibration
Requirements
Sampling &
Testing Plans
SPC Plans
Operating
Windows
DOE's
Maintenance
Schedules
Training
Materials
SOP's
Customer
Requirements
Short Term
Capability
Long Term
Capability
Noise Variable
Compensation
Reaction
Plans
Trouble
Shooting
Aids
Initial Control
Plan Assessment
These need to be
updated before
proceeding
![Page 5: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/5.jpg)
Measure – Key Elements
5
Understand the Customer and Demand patterns
Takt Time
Map the As-Is Process
Understand the Process
Collect Process Metrics -> Data Collection Plan
Measure the Process Baseline
Visualize the Process and Data
![Page 6: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/6.jpg)
Measure Phase – Deliverables
6
Process Characterized
Data Collected
Initial Analysis
Quick Wins
Plan for going forward
![Page 7: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/7.jpg)
Process Mapping
7
SIPOC
Value Stream Mapping
Swim Lane Diagrams
Spaghetti Diagrams
Process Mapping succeeds because it allows the team to see the process through the eyes of the customer
![Page 8: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/8.jpg)
Benefits of process mapping
Clear understanding of the ‘as is’ process
Inefficiencies, duplications, time delays are highlighted
Cross functional handovers visible – level of collaboration ?
Increased employee participation & morale
Greater ownership of the process
Smarter ways of thinking & working
Reduced costs
![Page 9: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/9.jpg)
Process Mapping - SIPOC
9
Suppliers Inputs Process Outputs Customers
•Sister Companies•3rd Party Suppliers•Engineering Cos•HR Agencies
•Intermediates•Raw materials•Equipment•Temps
Company X•Intermediates•Finished Goods•Product Info
•Wholesalers•Sister companies•Doctors•Regulators
Suppliers Inputs Process Outputs Customers
•Upstream processing•Warehouse• RM Suppliers•HR Agencies
•Bulk Actives•Excipients•Packaging material•Temps
Fill/Finish•Finished Goods•Batch records•Retain samples
•Warehouse•Stability Testing•QA•Regulators
High Level:
Lower Level:
![Page 10: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/10.jpg)
10
SIPOC example
![Page 11: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/11.jpg)
Process cross functions
Greater opportunities for disconnects and waste creation
![Page 12: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/12.jpg)
Business process improvement
Objective of waste reduction is to create smooth, uninterrupted, flow in the process & reduce non value adding activities
![Page 13: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/13.jpg)
The practicalities of mapping
Large dedicated room for process map, discussion, brainstorming Brown paper rolls A variety of coloured markers Post-its, different colours Blue tack Board/flip chart Documents associated with the process (forms, completion sheets,
procedures etc.) Cross section of workers from the process area & areas affected by the
process (e.g. internal suppliers & customers)
![Page 14: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/14.jpg)
14
Sample Swim Lane Diagram
![Page 15: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/15.jpg)
15
Sample Value Stream Map
15
![Page 16: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/16.jpg)
16
Sample Spaghetti Diagram
16
• Physically walk the process – can be used for materials, documentation etc
• Allows quick and easy visualization of poor layouts and redundant activities
![Page 17: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/17.jpg)
Complete the map
Connect the activities by pencil
Review for flow & completeness
All inputs & outputs accounted for?
Revise activity sequence, add activities as necessary
Invite review from external personnel
Connect activities with arrows
![Page 18: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/18.jpg)
Analyse the map /1
Identify activities which are Value Add and those that are not
Seek to eliminate non-value added work, simplify steps, refine the process
Start again
![Page 19: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/19.jpg)
Analyse the map /2
Measurements
Identify what family of measures should be used to quantify the process e.g.
Quality
Cost
Time
Are these relevant to the customer??
Measure the process performance not an individual employee’s output
![Page 20: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/20.jpg)
Analyse the map /3
Seek opportunities to:
Eliminate causes of errors
Eliminate duplication
Combine operations
Bring checking & inspection to the start of the process
reduce or eliminate paperwork,
e-mails, meetings
![Page 21: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/21.jpg)
Process mapping software
Software is available which is specifically suited to mapping:
Visio
Smartdraw
Essential for large and complex process maps
Excel is also useful for visualization
![Page 22: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/22.jpg)
Types of Data
1. Continuous (Variable) data
2. Discrete (Attribute) data
Working with Data
![Page 23: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/23.jpg)
Types of Measurement Data
Continuous (variable) Data: Continuous scale Standard, defined units
![Page 24: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/24.jpg)
Measurement Data
What is average height of Population ?
![Page 25: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/25.jpg)
Discrete (attribute) Data: Observed or counted Pass or Fail Low information content No scale Need a large sample size
Types of Measurement Data
NO-GO GO
![Page 26: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/26.jpg)
Discrete versus Continuous
Quantity of samples required to understand process
ContinuousDiscrete
$ $
SparseInformation
Rich WithInformation
![Page 27: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/27.jpg)
Categories of Scales
Category of Scale Description Example
Nominal Data Classification No ordering
Eye Colour
Ordinal Data Ordered Differences between values can not be determined
Satisfaction scale 1 .. 5 your degree of satisfaction
Interval Data OrderedDifferences between values can be measuredRatios not meaningfulNo natural start point
Temperature (C,F) – 20o is not twice as hot as 10o
Ratio DataOrderedConstant scale Natural zeroRatios are meaningful
Height, weight, age, length, cost.20cm is twice as long as 10cm
![Page 28: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/28.jpg)
Conclusions
No practical difference between interval and ratio data in terms of quantity of data required
Discrete (ordinal/ nominal) will require much bigger sample size than Continuous data (interval/ratio).
![Page 29: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/29.jpg)
How to collect data
The Process:
Clarify the data collection goals
Develop a data collection procedure
Validate the Measurement System
Collect the Data
Modify type and quantity of data collected, if
necessary
![Page 30: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/30.jpg)
1. Clarify data collection goals
Why are you collecting data ?
What questions will be answered ?
What patterns should be explored ?
How will the data help?
What type of data do I need ?
Should the data be stratified ?
![Page 31: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/31.jpg)
Drill down to find root causes
••
••
•
A B C D E Other
20
40
60
80100%
50%
75%
25%
Business Metric (e.g. downtime)Business Metric
Defect A
y1
Dri
ll d
ow
n
y2 y3
Defect B
y4 y5
![Page 32: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/32.jpg)
Data stratification
Who – Departments, groups, shifts, sections
What – Machines, equipment, products, services
Where – The physical location of defect/problem
When – Day of the week, time of the day etc.
![Page 33: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/33.jpg)
33
How to measure ?
Who should collect the data?
What instrument should be used?
Sample interval or every item?
2. Develop procedures
![Page 34: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/34.jpg)
34
3. Validate measurement system
Measurement system must be validated
What gets measured, gets improved !
![Page 35: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/35.jpg)
Data/measurement accuracy
Finished files are the result of years of scientific study
combined with the experience of years
How many times does the letter F appear in the sentence ?
![Page 36: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/36.jpg)
4. Collect Data
Train the data collectors
Pilot the data collection System
Error Proof the collection system
Present the data
![Page 37: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/37.jpg)
Some questions ?
Do you currently collect / receive any data that helps you drive improvements?
Is there any data that you should be collecting / receiving but currently don’t ?
Do you currently collect / receive any data that does not help you drive improvements?
What data will you collect for your project ?
![Page 38: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/38.jpg)
39
Pareto chart – Why use it ?
Identifies areas that offer the greatest potential for improvements
Shows relative descending frequency in a bar graph.
Focuses effort
Pareto Chart
![Page 39: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/39.jpg)
40
Pareto chart
What does it do?
Applies the Pareto principle: “80/20” rule.
Selects the starting point for problem
solving.
Compares progress before and after
improvement activity.
80%
80%
20%
20%
Causes Problems
Cause
Cause
![Page 40: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/40.jpg)
41
Pareto example
Reasons for failed mortgage applicationEx
ceed
gui
de...
No
ID
Page
mis
sing
No
sign
atur
e
Stat
us c
hang
e
Empl
oym
ent .
..
Too
old
Too
youn
g
Clai
m o
n ho
use
Faile
d m
edic
al
Oth
ers
0
40
80
120
160
200 90
80
70
60
50
40
30
20
10
33%45%
55%
64%72%
80%85% 89% 92%
93% 100%
No.
of r
epor
ted
occu
rren
ces
![Page 41: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/41.jpg)
Contractdisputes
0 -
120 -110 -100 -
90 -80 -70 -60 -50 -40 -30 -20 -10 -
Instructor Content Date mix up
Misc. Failed Recon-called
Debt Bynd.
# of unpaid Bills
Contract disputes50 -
40 -
30 -
20 -
10 -
0 -No PO
No Spec Error on PO
No Proposal.
AFD. Misc. No record
OA
% ofContractDisputes
Pareto chart stratification
![Page 42: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/42.jpg)
43
Run chart – Why use it ?
To allow a team to study a process for
trends over a specified period of time.
0
5
10
15
20
25
30
35
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb
Month
# R
equi
red
Cha
nges
![Page 43: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/43.jpg)
44
Run chart
What does it do? Monitors the processes over time.
Identifies trends or patterns over time
Triggers process improvement efforts
Defines timing and trend of problem
Compares performance before and after process improvement.
0
5
10
15
20
25
30
35
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb
Month
# R
eq
uir
ed
Ch
an
ges
![Page 44: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/44.jpg)
Plot the data
(x axis)Time or sequence
Average
(y a
xis)
Mea
sure
men
t
![Page 45: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/45.jpg)
Run chart essentials
Key success behaviours
Continue to ask questions
Appreciate the ‘aha’s’ revealed through time
Don’t jump to conclusions
![Page 46: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/46.jpg)
Histogram
Why use it ?
Displays the center, variation and shape of the data
Displays data is relative to the specification
Displays process variability (spread)
Benefits
Highlights unusual or unexpected results
Compares before and after change
![Page 47: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/47.jpg)
48
Histogram – What does it do ?
Helps answer the question: “Is the process capable of meeting my customers requirements?”
Displays large amounts of data that are difficult to interpret in tabular form
![Page 48: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/48.jpg)
Histograms
The following data represents measurements for a CTQ with upper and lower specification limits of .14 and .06 respectively. How are you doing at meeting your customer’s expectations?
.13
.11
.11
.13
.12
.14
.16
.16
.14
.12
.10
.09
.17
.13
.13
.10
.10
.16
.12
.13
.15
.15
.14
.13
.12
.13
.12
.11
.14
.15
.13
.10
.12
.15
.11
.13
.11
.14
.11
.15
.15
.12
.13
.13
.12
.14
.13
.13
.16
.12
![Page 49: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/49.jpg)
Constructing a histogram
Step 1: Decide what to measure
Step 2: Gather the data
Step 3: Decide on the class interval
Step 4: Determine the class size
Step 5: Assign the data to each class
Step 6: Plot the data
Step 7: Interpret the chart
![Page 50: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/50.jpg)
Histograms
To build a picture first we have to:
Determine class intervals:
The square root of the number of data points
Calculate the range across all the data:
The range = Max – Min
Estimate the size of each class:
Class size = The range/ (number of classes -1)
![Page 51: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/51.jpg)
Histograms
Using the data from previous slide…
Determine class intervals
The square root of 50 is roughly 7
Calculate the range across all the data
The range = 0.17 – 0.09 = 0.08
Estimate the size of each
Class size = 0.08/6 = 0.013
However, with this data it makes more sense to use nine class intervals
Class size = 0.08/8 = 0.01
![Page 52: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/52.jpg)
53
Histograms
• Fig. 1 - Tally chart • Fig. 2 - Histogram
CTQ Tally
0.17 |
0.16 ||||
0.15 |||| |
0.14 |||| |
0.13 |||| |||| |||
0.12 |||| ||||
0.11 |||| |
0.10 ||||
0.09 |
0.09 0.1 0.11 0.12 0.13 0.14 0.15 0.16 0.170
2
4
6
8
10
12
14
CTQ
Fre
qu
ency
![Page 53: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/53.jpg)
54
Histograms
Histogram for a sample of
sixty recorded mortgage
cycle times
The histogram shows:
Spread
Location
Shape of the distribution2 3 4 5 6 7 8 9 10
0
10
20
Mortgage Cycle Time
Time
Std. Dev. = 1.63Mean = 6.0
N = 61
![Page 54: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/54.jpg)
55
Histograms
240 observations of
mortgage cycle time
Notice how the bumps in the
previous graph disappear
with an increased sample
size2 3 4 5 6 7 8 9 10
0
10
20
30
40
50
60
70
80
Mortgage Cycle Time
Time
Std. Dev. = 1.49Mean = 5.9
N = 241
![Page 55: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/55.jpg)
Normal Type Positively Skew Type
Left-Hand Precipice Type Comb Type
Twin Peak Type
Isolated Peak Type
Types of Histogram Distributions
![Page 56: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/56.jpg)
57
Box plot – Why use it ?
To give you a graphical summary of values
Identify extreme values.
1 2 3 4
1415161718192021222324
Moulding machine
Dim
ensi
on X
![Page 57: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/57.jpg)
58
4321
64
62
60
58
56
54
52
50
Press
We
ight
1
60
59
58
57
56
55
54
53
52
51
Press,
We
ight
1
Box plot example
A box plot is another way to
visualise variation
It marries categoric data and
variable data (In this case the
weights produced on several
machines)
![Page 58: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/58.jpg)
Box plot
A Box Plot: depicts measures of the distribution that allows comparison of various distributions
1 2 3 4
14
15
16
17
18
19
20
21
22
23
24
Filling machine
Fill
Vol
ume
![Page 59: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/59.jpg)
60
Box plot – What does it do ?
A Box Plot consists of:
A rectangular box that represents roughly the middle 50% of the data
Lines or ‘whiskers’ extending from either side that represent the general extent of the data
Marks for outliers (observations far from the rest of the data)
![Page 60: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/60.jpg)
Box plot analysis
*
+
Maximum observation
75th Percentile (Third Quartile)
Median (50th Percentile)
25th Percentile (First Quartile)
Minimum observation
Outlier
Mean
![Page 61: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/61.jpg)
http://www.six-sigma-material.com/
http://asq.org/index.aspx
http://www.lean.org/
Useful Resources
![Page 62: 3. M Phase](https://reader034.vdocuments.net/reader034/viewer/2022050802/55cf903c550346703ba424ed/html5/thumbnails/62.jpg)
Recommended Texts
Statistics without tears
Derek Rowntree
Penguin
Lean Thinking
James Womack & Daniel T Jones
Free Press
Understanding A3 Thinking
Durward K Sobek II & Art Smalley
CRC Press