OPMT 1005 – Process Improvement & Lean

Design
Week Three – Data & Statistics

2023-05-23
Land Acknowledgement

We would like to begin this Orienta4on by respec8ully acknowledging that we live,

work, and learn on the tradi4onal, ancestral, and unceded territory of the
Sḵwx̱wú7mesh (Squamish), səlilwətaɬ (Tsleil-Waututh), and xʷməθkʷəy̓əm
(Musqueam) Na4ons.

We take responsibility for working to understand and challenge the constructs of

sePler colonialism as they exist on this land and the spaces we par4cipate in and
strive to uphold the ideals of truth and reconcilia4on.

2 2023-05-23 vcc.ca
Week Three – Learning Objec9ves

§ Understand the theory behind some basic statistical terms.

§ Define the continuous variables and where they fits among various types of
data.

§ Understand and demonstrate the use of various data measures such as

measures of central tendency, measures of dispersion and measures of position.

§ Understand and demonstrate the use of data distribution curves.

§ Differentiate between a process under statistical control and a capable process.

§ Differentiate between specification and control limits of a process.

3 2023-05-23 vcc.ca
Basic Statistics – Definitions

Statistics:
A group of methods used to collect, analyze, present, and interpret data
and to make decisions.

Descriptive Statistics:
Methods of organizing, displaying and describing data by using tables,
graphs, etc.

Inferential Statistics:
Methods that use sample results to help make decisions or predictions
about a population.

4 2023-05-23 vcc.ca
Basic Statistics – Definitions

Population:
All elements (individuals, items or objects) whose characteristics are being studied.
It is also called Target population.

Sample:
A portion of the population selected for the study.

Representative Sample:
A sample that represents the characteristics of the population as closely as
possible.

Random Sample:
A sample drawn in such a way that each element of the population has a chance of
being selected.
5 2023-05-23 vcc.ca
Population & Sample

Illustration Source: Introductory Statistics 6th Edition; Prem S. Mann; John Wiley & Sons

6 2023-05-23 vcc.ca
Types of Data

Data

Qualita've (Categorical) Quantitative (Countable)

Continuous Data
Length / Weight / Height / Time / etc.
§ Statistical Parameter?
§ Relevant Statistical Model?
§ Statistical Graphs?
Discrete/Attribute Data
# of Houses / Cars / Accidents / etc.

vcc.ca
Types of Data

Data

Qualita've (Categorical) Quantitative (Countable)

§ Mean, Median, Mode, Range, Continuous Data

Standard Deviation, etc. Length / Weight / Height / Time / etc.
§ The Normal Distribution
§ Histograms, Box Plot, Time
Series, etc.
Discrete/Attribute Data
# of Houses / Cars / Accidents / etc.

vcc.ca
Types of Data

Data

Qualitative (Categorical) Quantitative (Countable)

Continuous Data
§ DPU, PPM, Percentage, Proportion, etc. Length / Weight / Height / Time / etc.
§ The Binomial Distribution
§ The Poisson Distribution
§ Pie Charts, Bar Graphs, Time Series, etc.
Discrete/Attribute Data
# of Houses / Cars / Accidents / etc.

vcc.ca
Continuous Random Variables

IllustraBon Source: Introductory StaBsBcs 6th EdiBon; Prem S. Mann; John Wiley & Sons

vcc.ca
Example of Raw Data

21 19 24 25 29 34 26 27 37 33

18 20 19 22 19 19 25 22 25 23

25 19 31 19 23 18 23 19 23 26

22 28 21 20 22 22 21 20 19 21

25 23 18 37 27 23 21 25 21 24

11 2023-05-23 vcc.ca
Example of Frequency Distribution

Illustration Source: Introductory Statistics 6th Edition; Prem S. Mann; John Wiley & Sons

12 vcc.ca
Relative Frequency & Percentage
Distributions

Illustration Source: Introductory Statistics 6th Edition; Prem S. Mann; John Wiley & Sons

13 vcc.ca
Example & Bar Graph of Frequency
Distribution

Illustration Source: Introductory Statistics 6th Edition; Prem S. Mann; John Wiley & Sons

14 vcc.ca
Frequency Distribu9on Curve

Illustration Source: Introductory Statistics 6th Edition; Prem S. Mann; John Wiley & Sons

15 vcc.ca
Histograms

Histograms are useful for large data sets.

The data below represents the thickness on semiconductor wafers

Example Source: Introduction to Statistical Quality Control, 6th Edition; Douglas C. Montgomery 2009; John Wiley & Sons, Inc.

16 vcc.ca
Histograms

Example Source: Introduction to Statistical Quality Control, 6th Edition; Douglas C. Montgomery 2009; John Wiley & Sons, Inc.

17 vcc.ca
Symmetric and Non-symmetric
Histograms

Illustration Source: Introductory Statistics 6th Edition; Prem S. Mann; John Wiley & Sons

18 vcc.ca
Stem-and-Leaf Display

If we collect data regarding the monthly rents paid by a sample of 30

households selected from a small city, we may get the following data:

880 1081 721 1075 1023 775 1235 750 965 960

1210 985 1231 932 850 825 1000 915 1191 1035

1151 630 1175 952 1100 1140 750 1140 1370 1280

19 vcc.ca
Stem-and-Leaf Display

6 30
7 75 50 21 50
8 80 25 50
9 32 52 15 60 85 65
10 23 81 35 75 00
11 91 51 40 75 40 00
12 10 31 35 80
13 70
Illustration Source: Introductory Statistics 6th Edition; Prem S. Mann; John Wiley & Sons

20 vcc.ca
Measures of Central Tendency

Mean:
For population data: µ = Ʃx/N
For sample data: x = Ʃx/n

Median:
Is the value of the middle term in a data set that has been ranked in an
increasing order

Mode:
The value that occurs with the highest frequency in a data set

21 vcc.ca
Mean, Median & Mode
Symmetric Frequency Distribution Curve

Illustration Source: Introductory Statistics 6th Edition; Prem S. Mann; John Wiley & Sons

22 vcc.ca
Mean, Median & Mode
Skewed Frequency Distribu9on Curve

Illustration Source: Introductory Statistics 6th Edition; Prem S. Mann; John Wiley & Sons

23 vcc.ca
Raw Data
Top Diameter (mm) Sample 1 Sample 2 Sample 3
Shot 1 13.912 13.908 13.906
Shot 2 13.909 13.909 13.895
Shot 3 13.910 13.907 13.909
Shot 4 13.912 13.897 13.911
Shot 5 13.913 13.910 13.898
Shot 6 13.911 13.910 13.911
Shot 7 13.910 13.906 13.911
Shot 8 13.907 13.910 13.912
Shot 9 13.913 13.910 13.898
Shot 10 13.911 13.913 13.909
Shot 11 13.911 13.914 13.912
Shot 12 13.907 13.910 13.906
Shot 13 13.913 13.904 13.892
Shot 14 13.906 13.915 13.903
Shot 15 13.913 13.911 13.907

24 2023-05-24 vcc.ca
Measures of Central Tendency

Mean
Median
Mode

25 2023-05-24 vcc.ca
Measures of Central Tendency

Median
Mean Mode

Data skewed to the left

vcc.ca
Measures of Dispersion

Range:
R = Largest value – smallest value

Variance & Standard Deviation:

§ The value of standard deviation tells how closely the values of a data set are clustered
around the mean.
§ In general, a lower value of the standard deviation for a data set indicates that the values
of that data set are spread over a relatively smaller range around the mean.
§ In contrast, a larger value of the standard deviation for a data set indicates that the values
of that data set are spread over a relatively larger range around the mean.

27 vcc.ca
Use of Standard Devia9on

For a bell-shaped
distribution:

1. 68.2% of the observations lie

within one standard deviation of
the mean

2. 95.4% of the observations lie

within two standard deviation of
the mean

3. 99.7% of the observations lie

within three standard deviation of
the mean

28 vcc.ca
Illustration of the Empirical Rule

Illustration Source: Introductory Statistics 6th Edition; Prem S. Mann; John Wiley & Sons

29 vcc.ca
Normal Distribution with Mean μ
& Standard Deviation σ

IllustraBon Source: Introductory StaBsBcs 6th EdiBon; Prem S. Mann; John Wiley & Sons

30 vcc.ca
Normal Distribution with Mean μ
& Standard Deviation σ

Illustration Source: Introductory Statistics 6th Edition; Prem S. Mann; John Wiley & Sons

31 vcc.ca
Measures of Position

§ Quartiles
§ Percentiles
§ Box Plot

32 vcc.ca
Box Plots

Illustration Source: Introductory Statistics 6th Edition; Prem S. Mann; John Wiley & Sons

33 vcc.ca
 Measures of Position
Raw Data

Sample Day 1 Day 2 Day 3 Day 4 Day 5 Day 6 Day 7 Day 8 Day 9 Day 10 Day 11 Day 12 Day 13 Day 14

1 56.0044 56.0177 55.993 56.0099 56.0292 56.0257 56.05 55.9804 55.9824 56.0008 56.0345 56.0025 56.0032 55.9401
2 56.0038 55.9804 55.9967 56.0014 56.0078 55.9554 56.0305 56.0162 55.9664 55.9944 56.008 55.9981 55.9875 55.9965
3 55.9922 56.0019 55.9857 55.9943 56.0172 56.0002 55.9689 56.0122 55.9759 55.9944 56.0266 56.0035 55.9882 56.0007
4 55.9771 55.9924 55.9892 56.0014 55.9973 55.9922 56.0257 55.9919 55.9691 56.013 56.0088 55.9907 55.9948 55.9898
5 56.0205 56.0032 55.9885 56.0061 55.9945 56.032 55.9667 55.9866 56.0259 55.926 56.0368 55.9534 55.9883 55.9856
6 56.0058 55.9714 55.9884 56.0067 56.0085 55.9784 55.9948 55.9919 56.0313 55.9925 56.0043 55.9937 56.0086 55.9908
7 56.0065 55.9935 55.9896 56.0148 56.0217 56.0053 56.0334 55.9781 55.9948 55.9949 56.0114 55.9585 55.9585 55.9396
8 56.0126 56.0076 55.9924 55.9959 56.0217 56.0068 55.9729 55.9933 55.9933 55.9832 56.0151 56.0065 55.994 55.9752
9 56.0186 56.0116 55.9945 56.0031 56.0134 56.0186 55.9822 55.9727 56.0034 55.9909 56.0088 56.0037 55.9954 56.0076
10 56.0087 55.9941 55.9845 56.0007 56.0209 56.0288 56.0246 55.9727 55.9685 56.0015 55.9964 56.0012 55.9741 55.9363
11 56.0063 56.0153 56.0321 56.0047 56.0017 55.9809 56.0084 56.0009 55.9817 56.0097 55.9951 55.9789 55.9758 56.0046
12 55.9865 56.0148 55.9839 56.008 55.9935 56.0025 55.988 55.9973 56.0138 55.9307 56.0271 56.0162 56.0034 55.9417
13 56.0071 56.0098 56.0068 56.0034 56.0089 56.0273 56.031 55.9918 55.9736 55.9896 55.99 55.9726 56.0062 55.9908

vcc.ca
Measures of Position
Boxplot

vcc.ca
Measures of Position
Individual Plot

vcc.ca
Statistical Process Control (SPC)

Objective of SPC:
§ The first objective of SPC is to get the process in control, which means the identification
and elimination of special causes of variation.

A Process in Control:
§ This process will consistently produce parts within tolerance limits.
§ This is done by eliminating all of the special causes of variation that may exist.
§ After a process is in control (stable) and producing consistently within its natural balance, it
can be then compared to the engineering tolerance limits.

A Capable Process:
§ The capability of a process is directly related to the ability of the process to produce parts
consistently within the drawing tolerance limits.
§ Capability can not be studied until the process is in control.

37 2023-05-23 vcc.ca
Statistical Process Control (SPC)

Random Variation:
§ This is the variation that has several common causes.
§ Common causes of variation are things like vibration, heat, and humidity that either
individually or collectively cause variation

Non-random Variation:
§ This is the variation that is not built into the process.
§ Causes of variation like this are called special causes. Special causes of variation are those
that can be directly identified and corrected.
§ Examples of special causes are tool wear, loose work-holding device, fatigue and power
surge

38 2023-05-23 vcc.ca
Specification & Control Limits

§ SpecificaGon limits are defined by the customer.

§ Customer can be external or internal.

§ SpecificaGon limits are different than (and independent of) the control
limits.

§ SpecificaGon limits define the “Voice of Customer”.

§ Control Limits define the “Voice of Process”.

39 2023-05-23 vcc.ca
Specification & Control Limits

Target

LSL LCL UCL USL

Lower Lower Upper Upper
Specification Control Control SpecificaGon
Limit Limit Limit Limit

Width of distribution
(Voice of Process)

Tolerance = USL - LSL

(Voice of Customer)

40 vcc.ca
 Defect Probability

Target
LSL USL
Lower Upper
Specification Specification
LCL Limit Limit UCL
Lower Upper
Control Control
Limit Limit

Defects Defects

41 vcc.ca
Capability Indices

These indices reflect how the process could perform relative to the specification limits, if the
shifts and drifts of the process could be eliminated.

§ Cp relates the process spread (the 6-σ variation) to the specification spread. In other
words, Cp relates how the process is performing to how it should be performing.
§ Cp does not consider the location of the process mean relative to the specification interval.
§ Cp tells you what capability your process could achieve if centered.
§ CPU and CPL - relates the process spread (the 3σ variation) to a single-sided specification
spread (μ-LSL or USL- μ).
§ CPL and CPU consider both process mean and process spread.
§ Cpk - minimum of CPU and CPL.
§ Cpk incorporates information about both the process spread and the process mean, so it is
a measure of how the process is actually performing.

42 2023-05-23 vcc.ca
Capability Indices

43 vcc.ca
Process Capability

Always remember:
§ Cp = Capability of the Process
§ Capability of the Process = (Voice of the Customer / Voice of the Process)
§ Capability of the Process = (Tolerance / Width of Distribution)
§ Cpk considers the location of the process mean while Cp does not
§ If Cp and Cpk are approximately equal, then the process is centered between the
specification limits
§ If Cp is greater than Cpk, then the process is not centered

Cp / Cpk (Short Term) Cp / Cpk (Long Term)

Red (Bad) < 1.33 < 1.0

Yellow (Marginal) 1.33 – 1.67 1.0 – 1.33

Green (Good) > 1.67 > 1.33

44 2023-05-23 vcc.ca
Process Capability

What do you think of this process?

45 2023-05-24 vcc.ca
Process Capability

What do you think of this process?

46 2023-05-24 vcc.ca
Process Capability

What do you think of this process?

47 2023-05-24 vcc.ca
Process Capability

What do you think of this process?

48 2023-05-24 vcc.ca
Discussion Questions

For the following 50 data points representing a diameter of a machined product

measured in millimeters, calculate:
§ Mean, Median, Mode, and Range
§ Plot the data on a histogram
§ Does the histogram represent a normal distribution? Yes or no? Why?
§ Is the histogram symmetrical? Is it skewed to the right? Or is it skewed to the left??
§ If the customer specifications are 13.915mm as a target value for diameter with an
allowance of 13.910mm as a minimum acceptable diameter to 13.920mm as a maximum
acceptable diameter (i.e. 13.915mm +/- 0.005mm), how would you comment on this batch
of machined parts?
§ What is the percentage of non-acceptable parts to the total batch quantity?
§ If the standard deviation of this group of data is 0.008mm, calculate the Cp and Cpk and
comment on the capability of this machining process?
§ Raw data in the next slide

49 vcc.ca
Discussion Questions

Raw Data:
50 data points representing a diameter of a machined product measured in
millimeters

13.890 13.899 13.902 13.904 13.907 13.908 13.909 13.911 13.914 13.919
13.891 13.900 13.902 13.904 13.907 13.908 13.909 13.911 13.915 13.920
13.895 13.901 13.903 13.906 13.907 13.908 13.910 13.912 13.916 13.923
13.897 13.901 13.903 13.906 13.907 13.908 13.910 13.913 13.917 13.924
13.898 13.902 13.904 13.907 13.907 13.909 13.910 13.913 13.918 13.925

§ This exercise will be performed and submitted in groups

§ Please answer the questions in the forum – deadline to submit your answers and share
your thoughts is next week before the start of the next session

50 2023-05-24 vcc.ca