Scribd
Upload
59 views46 pages

Statistical Process Control: Rajiv Gupta BITS Pilani

This document provides an overview of a lecture on statistical process control. It covers the normal distribution and key metrics like measures of central tendency, variation, and control charts. The lecture consists of 5 modules that cover normal distribution, measures of central tendency and dispersion, sources of variation, control charts, and variables control charts specifically. It provides examples of calculating averages, ranges, standard deviations and control limits for an X-bar and R chart using sample data. The goal is to understand process variability and use control charts to monitor processes and identify assignable causes.

Uploaded by

mohimran2002
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
59 views46 pages

Statistical Process Control: Rajiv Gupta BITS Pilani

This document provides an overview of a lecture on statistical process control. It covers the normal distribution and key metrics like measures of central tendency, variation, and control charts. The lecture consists of 5 modules that cover normal distribution, measures of central tendency and dispersion, sources of variation, control charts, and variables control charts specifically. It provides examples of calculating averages, ranges, standard deviations and control limits for an X-bar and R chart using sample data. The goal is to understand process variability and use control charts to monitor processes and identify assignable causes.

Uploaded by

mohimran2002
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 46

Statistical Process Control

Lecture 1
Rajiv Gupta
BITS Pilani

Agenda
Module 1
Normal Distribution

Module 2
Measures of central tendency and dispersion

Module 3
Variation

Module 4
Control Charts

Module 5
Variables Control Charts
2

Begin Module 1
Normal Distribution

Variability in a Process and Process


Output
Processes have inherent variability due to differences in
-- incoming materials, energy (e.g. fluctuations electrical
voltage and hydraulic pressure) and information
-- operating conditions of machines/ equipment/tools
(e.g. work positioning, holding; wear)
-- operators and their work practices
-- environment (e.g. temperature, humidity, light levels)
The process variability will affect its output, i.e. product
characteristics and quality. Thus, process variability (related to
CTQs) must be reduced.
The product and its production process must be designed to
minimize variability in its production process.

The Normal Distribution


The Normal distribution is one of the most common
statistical distributions that has been used to model
the variability of the data
It is a bell-shaped, symmetric, continuous
distribution that has been shown to fit data from a
variety of data such as heights of people
The Normal distribution is completely specified by
two parameters, the mean and the standard
deviation
5

Normal Distribution

Standardized Normal Variable

See next two slides

Normal
Distribution
Table

Normal
Distribution
Table
(contd.)

Parameters of Normal Distribution

10

Location and Dispersion


(outputs of a process of shooting arrows)

11

End of Module 1

12

Begin Module 2
Measures of central tendency and dispersion

13

Population and Samples


Population
Population is a well-defined collection of individuals or
objects of a particular type that is of interest to you

Sample
A sample is a subset of the population that is selected in a
prescribed manner

Often a sample is taken to draw inferences about the


population because it may be impractical to poll the
entire population

14

Measures of Central Tendency


A measure of central tendency of a distribution is a
numeric value that describes the central position of
the data or how the data tend to accumulate in the
center.
Three common measures are: 1) the average
(arithmetic mean), 2) the median (value that divides
the data equally), and 3) the mode (the value that
occurs with the greatest frequency).

15

Average (or Mean)

16

Median and Mode


Data set 1 4 4 8 20 22 23 40 50
Average/Mean = 172/9 = 19.1
Median is 20
Mode is 4

17

Measures of Dispersion
A measure of dispersion describes how the
data are spread out or scattered on each side
of the central value.
Two common measure of dispersion are: 1)
Range (difference between largest and
smallest values of observations), and 2)
standard deviation.

18

Standard Deviation (s)


s is an estimate of . Note: is the population
standard deviation.

19

Data set 1 4 4 8 20 22 23 40 50
Xbar = 19.1
S = (((1-19.1)2+(4-19.1)2+(4-19.1)2+(8-19.1)2+(20-19.1)2+(22-19.1)2
+(23-19.1)2+(40-19.1)2+(50-19.1)2)/8)
= (2322.889/8)
= 17.04

Range R = 50 -1 = 49

20

End of Module 2

21

Begin Module 3
Variation

22

Variations
No two products produced from the same
process are ever made exactly alike. (Precise
instruments will show differences).
Variation: Quantitative change in value
between items or over time caused by
common and special (assignable) causes.
Variations in a process can be caused by
Common and Assignable Causes.
23

Common Causes
Common causes are normal or natural
random variations in a process when the
process is running normally or is under
control.
There are no other causes other than
natural/inherent random variations due to: 1)
Operator, 2) material, 3) equipment, 4)
method, 5) tooling, and 6) environment.
24

Assignable Causes
Assignable (or Special) causes of variations
arise because of special circumstances that
are not an inherent part of the process and
can be eliminated.
They are called assignable because someone
has made a mistake (or something that is
outside the normal random variation in the
process has occurred). Examples are new
material, new operator, etc.
25

Variations
Categories of variations:
- Within-a-piece variation (e.g. hardness of the
material is not uniform across different locations
on a component).
- Piece-to-piece variation (e.g. pieces vary from
each other when measured at the same location)
- Time-to-time variation (e.g. measured value
changes when measured at a different time).
26

Repeatability & Reproducibility


Repeatability refers to the degree of variability
between repeated measurements with the same
instruments, operator, location, time and method
(e.g. the operator make a succession of
measurements).
Reproducibility refers to the degree of variability
between measurements when different methods,
people, places, or times are used (e.g. measurement
made in different laboratories).

27

End of Module 3

28

Begin Module 4
Control Charts

29

Process Output-- Run Chart & Control


Charts
Run chart shows process output (e.g. X-bar plotted
as a function of subgroup number).
Control Chart Run chart with control limits (i.e. LCL,
UCL)
Note: Control limits are not the same as specification
limits (i.e. LSL, USL)
Two types of control charts
- Variables Control Charts
- Attribute Control Charts
30

Types of Control Charts


Variables Control Charts
Used to monitor measurable quality
characteristics of a process. This may include
temperature, viscosity, weight, etc.,
Xbar and R Charts or Xbar and S Charts

Attribute Control Charts


Attribute value can have only two values pass/fail,
conforming/non-conforming, present/absent etc.,
P Charts and np Charts
31

Process Outputs
Processes produce products.
Products have characteristics that are related to
quality (CTQ).
Product characteristics vary due to variations in
various components of the production system (e.g.
tools, temperature, raw material, operators, etc.).
Output products are measured (during inspections)
by pulling samples from the process.
Sample = Some products are selected for inspection.

32

Samples
Selected products are samples from the population of output
products.
Samples can be obtained in subgroups of size n.
Accuracy in estimating parameters will increase with increase
in subgroup size (n).
Cost of sampling (and inspection) will increase with increase
in subgroup size (n).
Items in each subgroup can be inspected for:
- Measurements of characteristics
- Defects (e.g. scratches, distortions, wrinkles)
- Items can be accepted or rejected (defectives or
non-conforming parts).
33

Output Measurements
The characteristics of products (e.g. length)
are assumed to follow certain known
distributions (e.g. normal).
From the measurements (or inspection) of the
samples, estimates of the process distribution
parameters (e.g. mean, standard deviation,
proportion defectives) are computed.

34

End of Module 4

35

Begin Module 5
Variables Control Charts

36

Variables Data
From the measurements of a given characteristic of
each item in each subgroup-- X-bar (X), Range (R)
and Standard Deviation (s) can be computed.
Run charts for X, R and S can be developed.
Upper and lower control limits (UCL and LCL) are
computed using formulas.
When upper and lower control limits are added to a
run chart, the chart becomes a control chart.

37

Variables Data
Points falling outside the control limits can be
investigated, assignable causes can be
identified, process can be fixed, and natural
variability of the process can be computed.
The following pair of charts are used for SPC
X-bar and R
X-bar and S
X and RM
38

X versus X bar
Variables Control Charts can be constructed
for individual observations as well as for
sample means
Sample means provide a more appropriate
measure of the status of the process
Variations in the individual readings may mask
underlying patterns and shifts in the process
output
39

Variables Data
R and S both provide estimates of dispersion.
- R is easier to compute than S.
- Range (R) can be easily computed by obtaining
difference between the largest and the smallest
values in each subgroup.
- S provides a better estimate of dispersion than R.
- R can be easily computed by shop floor personnel.
- R is generally used when n = 2 to 9,
and S is generally used when n >= 10.
(Typically, n = 10 to 20).
40

Example Xbar and R Chart

S1
S2
S3
S4
S5
S6
S7
S8
S9
S 10
S 11
S 12
S 13
S 14
S 15
S 16
S 17
S 18
S 19
S 20
S 21
S 22
S 23
S 24
S 25

Pc 1

Pc 2

Pc 3

Pc 4

Pc 5

Xbar

0.65
0.75
0.75
0.6
0.7
0.6
0.75
0.6
0.65
0.6
0.8
0.85
0.7
0.65
0.9
0.75
0.75
0.75
0.65
0.6
0.5
0.6
0.8
0.65
0.65

0.7
0.85
0.8
0.7
0.75
0.75
0.8
0.7
0.8
0.7
0.75
0.75
0.7
0.7
0.8
0.8
0.7
0.7
0.65
0.6
0.55
0.8
0.65
0.6
0.7

0.65
0.75
0.8
0.7
0.65
0.75
0.65
0.8
0.85
0.6
0.9
0.85
0.75
0.85
0.8
0.75
0.85
0.6
0.85
0.65
0.65
0.65
0.75
0.65
0.7

0.65
0.85
0.7
0.75
0.85
0.85
0.75
0.75
0.85
0.8
0.5
0.65
0.75
0.75
0.75
0.8
0.7
0.7
0.65
0.6
0.8
0.65
0.65
0.6
0.6

0.85
0.65
0.75
0.65
0.8
0.7
0.7
0.75
0.75
0.65
0.8
0.7
0.7
0.6
0.85
0.65
0.8
0.6
0.7
0.65
0.8
0.75
0.65
0.7
0.65

0.70
0.77
0.76
0.68
0.75
0.73
0.73
0.72
0.78
0.67
0.75
0.76
0.72
0.71
0.82
0.75
0.76
0.67
0.70
0.62
0.66
0.69
0.70
0.64
0.66

0.2
0.2
0.1
0.15
0.2
0.25
0.15
0.2
0.2
0.2
0.4
0.2
0.05
0.25
0.15
0.15
0.15
0.15
0.2
0.05
0.3
0.2
0.15
0.1
0.1

0.72

0.18

Average

41

X-bar and R
Charts

42

Constants for computing control limits of variables charts

43

Calculation of UCL and LCL


Average Xbar = 0.72 Average R = 0.18

For the Xbar Chart,


Center line = Average Xbar = 0.72
UCL = Average Xbar + A2*Average R
LCL = Average Xbar A2*Average R
For n = 5, A2 = 0.58
UCL = 0.72 + 0.58*0.18 = 0.82
LCL = 0.72 0.58*0.18 = 0.61

For the R Chart,


Center line = Average R = 0.18
UCL = D4*Average R
= 2.11*0.18 = 0.38
LCL = D3*Average R
= 0* 0.18 = 0
44

Example of Control Charts: X-bar


and R Charts

45

End of Module 5

46

You might also like