Professional Documents
Culture Documents
Basic Statistical Thinking
Basic Statistical Thinking
Productionism
Quality is conformance to Standards
Consumerism
Quality is meeting a Customer’s delight
improvement
( Lt Col Dr Yasir)
Quality
Suppliers Environment
Cost
Machines
Responsiveness
Methods
Material
Measurement
Note that customers and suppliers should beconsidered part of
the process definition (CUSTOMER FOCUS)
11/15/23 Dr Tasweer Hussain Syed 22
Tools for describing processes
Tools for describing processes include
process flow diagram
cause and effect diagram
Appropriate level of detail in process
description is critical
It is more difficult to think of non-
manufacturing work as a process because
not tangible
usually not well documented (no blueprint)
possible
and customers
process may have many customers and many
suppliers
process may have many inputs and many
outputs
organizing,
presenting,
analyzing, and
interpreting
Data + Variable
= Meaningful Information
40 Data
40 Shoes
20 Pairs of Shoes
Quantitative Quantitative
Discrete
Continuous
Nominal
Ordinal
Interval
Ratio
NOMINAL, CATEGORICAL
E.g. Apples and pears, gender, eye colour,
ethnicity.
Data that are classified into categories and
cannot be arranged in any particular
order. Nominal=Categorical=Dichotomous
understanding DATA:
DATA = Recorded Information
e.g., Sales, Productivity, Quality,
Costs, Return, …
5. Hypothesis testing:
Data decide between two possibilities
Does “it” really work? [or is “it” just randomly
better?]
Is financial statement correct? [or is error
material?]
Whiter, brighter wash?
Makes use of
Statistics – all the basic activities, and
Prediction, classification, clustering
Computer science – efficient algorithms (instructions) for
Collecting, maintaining, organizing, analyzing data
Optimization – calculations to achieve a goal
Maximize or minimize (e.g. sales or costs)
Twentysomethings
Young
Mobile College Campuses
Adults Military Proximity
.
.
. .
.
.
26 5 18 20 13 13
13 7 9 19 19 22
33 10 9 5 18 9
10 3 10 18 10 7
13 13 17 17 17 17
11/15/23 Dr Tasweer Hussain Syed 50
FREQUENCY TALLY
26 1
13 5
33 1
10 4
20 1
19 2
5 2
18 3
17 4
7 2
3 1
9 3
22 1
11/15/23 Dr Tasweer Hussain Syed 51
11/15/23 Dr Tasweer Hussain Syed 52
Two variant curves.
(a) The values are widely dispersed away from their mean.
(b) The values are tightly grouped about their mean.
xi x 2
n
Where = standard deviation of values in distribution
1,264
42.13
30
Step 5 TAKE THE SQUARE ROOT OF THE
QUOTIENT FROM STEP 4.
2 4 6 8 10 12 14 16 18 20 22 24 26
3 3
.813 of
area
.747 of
area
1,250 1,400
11/15/23 Dr Tasweer Hussain Syed 65
MEDIAN / MODE
No. of persons Monthly earning
in Rupees
25
1400
15
1500
12
2400
18
3600
11/15/23 20
Dr Tasweer Hussain Syed 66
INTERPRETING MEDIAN
FIFTY PERCENT OR MORE, OF THE DATA
WILL BE GREATER THAN OR EQUAL TO
THE MEDIAN.
C. Negatively Skewed
11/15/23 Dr Tasweer Hussain Syed 68
MEASURES OF
VARIATION
OR
2000
Number of Troops
1500
1000
500
0
6.125 6.375 6.625 6.875 7.125 7.375 7.625
34% 34%
14% 14%
2% 2%
APPROXIMATE PERCENTAGES OF DIFFERENT MEASUREMENTS
WITHIN THE NORMAL DISTRIBUTION CURVE
11/15/23 Dr Tasweer Hussain Syed 78
FUNDAMENTAL PRINCIPLES
34% 34%
14% 14%
2% 2%
CAUSES OF VARIATION
Materials Machines
Dimension
PROCESS/CONTROL MANAGEMENT
Ensuring Accuracy, Precision, Consistency in all processes
MIXED
PRODUCT OK
Product
Inputs
Measu-
PROCESSES rement
REPAIR
REPAIR
Recurrence Defective
Prevention Product
Program Systematic C/A
Systematic
Investigation
Investigationfor
for
Root-causes
Root-causes
11/15/23 Dr Tasweer Hussain Syed 87
11/15/23 Dr Tasweer Hussain Syed 88
PROBLEM SOLVING TOOLS
SPEC SPEC
HISTOGRAM
.5620
.5619
.5618
.5617
.5622
.5621
.5620
.5619
.5618
.5617
.5622
.5621
.5623
.5623
28 51 40 52 43
11/15/23 Dr Tasweer Hussain Syed 93
GUIDELINES FOR DETERMINING THE
NUMBER OF INTERVALS
Fewer than 50 5 to 7
50 to 100 6 to 10
101 to 150 7 to 12
12
10
Frequency
0
17 22 27 32 37 42 47 52 57 62
SL SU
Fig: a
SL Fig: b
SU
11/15/23 Dr Tasweer Hussain Syed 100
SL Fig: c SU SL Fig: d SU
Shape:
The frequency is low near the middle of the range of
data, and there is a peak on either side.
Note:
This shape occurs when two distributions with widely
different mean values are mixed.
11/15/23 Dr Tasweer Hussain Syed 102
A GOOD SITUATION: PROCESS SPREAD
IS NARROW AND IS CENTERED
BETWEEN SPECIFICATIONS.
30
Lower specification
Upper specification
Frequency
20
10
30
Lower specification
Upper specification
Frequency
20
10
0
140 150 160 170 180 190 200 210
11/15/23
Tensile Strength (pounds)
Dr Tasweer Hussain Syed 104
PROCESS IS OFF CENTER
30
Lower specification
Upper specification
Frequency
20
10
40 50 60 70 80 90 100 110
specification
Upper
30
Lower specification
Frequency
20
10
Length (inches)
11/15/23 Dr Tasweer Hussain Syed 106
USE OF HISTOGRAMS