Unit - 2

STATISTICAL PROCESS CONTROL IN INDUSTRY

 UNIT II
STATISTICAL PROCESS CONTROL IN INDUSTRY

Statistical process control – quality control vs

process control – control charts – applications
– problems – seven tools of quality – seven tools
of management – implementation in electrical
industry.
Process control
 Process control
• Quality of a product or service depends on the quality of process
employed

• Process should produce the products truly as per design

• The objective of process control is to control the quality of

process and ensure that the deliverables are produced as
planned.

• The process characteristics can be measured analyzed using

statistical tools and controlled.
Statistical process control(SPC)
 Statistical process control(SPC)
The goals of SPC:
• Collection of data of performance of products and service
deliverables.
• Finding out variations.

• Analyzing through brain stroming & determining the

causes and eliminating the causes.
• Improving performance of processes continuously.
Process control
 Process control
• The quality of the products or service is planned and expressed in
the form of specifications or company standards. Ex: IS,ISI..,
• Meeting the specifications by closely monitoring the process.
• Each process will have specifications so as to meet the end product
quality standards.
• Requires measurements/examinations during the execution of task
in the processes.
• The intermediate product should be examined to see the
performance of the products.
• Its not performing the as per the specifications the processes are to
be adjusted.
• After the adjustment of the process, performance are to be
measured. End of the class
Process Requirements
 Requirements
• Process flow chart.
• Specifications for the processes.
• Documented procedures for carrying out the tasks and for
measurements.
• Criteria for acceptance or rejection of the product made.
• Data collection sheets(from QC)
• Instructions for recording data use of the required tools for
analysis carrying out analysis.
• Methodology for adjusting the process based on actual results.
• Training for the process owners in all the above points.
 7 tools
• Process flow chart
• Cause and effect diagram
• Check sheet
• Scatter diagram
• Pareto char.
• Histogram
• Control chart
 Control Chart
A statistical tool used to distinguish between
process variation resulting from common
causes and variation resulting from special
causes.
 Control Chart
Essential features of a control chart

Upper Control Limit

Variable

Central Line
Values

Lower Control Limit

Time
 Control Chart Types

Control charts

Variables Attributes
chart chart

R X P,np C,u
chart Chart Chart Chart
 Control chart for variables
• Quality characteristics which can measured and expressed
in specific units of measurements is called variables.
• Control chart based upon measurements of quality
characteristics for variables.
Types
o X or average chart- monitor the centering of the process to control
its accuracy
o R or range chart- monitor the dispersion or precision of the process
o S or standard deviation chart– shows variation of the process
 Construction of X and R chart
• Select the characteristics for applying control chart.
• Select the appropriate type of control chart
• Collect the data.
• Choose the rational sub group, ie,. Sample
• Calculate the average(X) and range (R) for each sample.
Sub group average X = x1 + x2 +x3 +x4 +x5 / 5
Sub group range R = Max Value – Min value
• Calculate the average of the averages(X) and average of
range(R)
• Calculate the averages of the average and range

x 
X
N

R 
R
N
 X Chart
Control Limits
Second Approach ,is to use the sample range as a measure of process
variability, CL  x
the appropriatex formulas for control limits are calculated as,
UCL  x  A R Factor or
x 2
LCL  x  A R
x 2 constant
From
Sub group average X = x1 + x2 +x3 +x4 +x5 / 5 Tables
Sub group range R = Max Value – Min value
 R Chart Control Limits

R-charts used to monitor process

dispersion
CL R  R
UCLR  D4R Statistical factor
LCLR  D3R
From Tables
 Cont…
• Plot CL, UCL, LCL on the chart
• Plot individual X and R values on the chart
• Check whether the process is in control or not.
• Revise the control limits if the points are out of control, by
removing the out of control points
 Control chart for
attributes
• Limitations of variable chart are
o X and R chart cannot be used for quality characteristics that are attributes- like incorrect
colours, blow holes, cracks… are not measurable
o X and R chart can be used for one measurable characteristics at a time.- automatic turning
may have 50 dimensions so need 50 charts..
In order to avoid the above limitations control chart for attributes are used.
An attributes refers to those quality characteristic that conform to specifications or do not
conform to specifications.
 Types of attributes
• Attributes where measurement are not possible.
• Attributes where measurements can be made but are not made because of
economy of times, cost or need.
Non conformity Vs Non conforming item (Defect Vs Defective)
• An item is said to be a non conforming item or defective if fail to conform to the
specifications in any of the characteristics. It refers to entire item.
• Each characteristics that does not meet the specification is a non conformity or defect.
It refers to quality characteristics.
Eg: a paint bubble on the surface of a newly manufactured car may be an conformity, yet
the car was shipped as conforming.
 Type of attributes
• p- chart- for fraction rejected as non confirming to specifications
• np- chart- for number of non conforming items.
• c- chart- for number of non conformities.
• u- chart- for number of non conformities per unit.
 p Chart for Attributes
• P-chart is used to monitor the proportion of defectives
generated by a process
• Type of attributes control chart
o Nominally scaled categorical data

e.g., good-bad
•
• Shows % of nonconforming items
• Example: Count # defective chairs & divide by total chairs
inspected; Plot
o Chair is either defective or not defective
 p Chart-Control Limits

• p = np / n where p = Fraction of Defective

np = no of Defectives
n = No of items
inspected in sub group
p= Avg Fraction Defective
p (1  p )
= ∑np/ ∑n = CL
UCLp  p  z
n z = 3 for 99.7%
p (1  p ) limits
LCL p  p  z
n
 Purpose of the p Chart
Identify and correct causes of bad quality.
Identity and correct thee erratic causes of quality improvement
The average proportion of defective articles submitted for
inspection, over a period of time.
To suggest where it is necessary to use X and R chart to
diagnose quality problems
Determine average Quality Level.
 np CHART
p and np are quiet same
Whenever subgroup size is variable,p chart is used. If sub group size is constant, then
np is used.
FORMULA: Central Line CLnp = n p
Upper Control Limit, UCLnp = n p +3√ (n p (1- p ))
Lower Control Limit, LCLnp = n p -3 √ (n p (1- p ))
Where p = ∑ np / ∑n =Average Fraction Defective
n = Number of items inspected in subgroup.
When and why np chart is prefered over p chart?
• When subgroup size is constant, the np chart is preferd over p chart
• Reason – (i) the np chart saves one calculation for each subgroup, the division of number
of rejects by subgroup size to get p.
• (ii) some people may understand the np chart more readily
 c Chart(c for count)
• When the goal is to control the number of defects per unit, a c-chart is
used
• Type of attributes control chart
o Discrete quantitative data

• Shows number of nonconformities (defects) in a unit

o Unit may be chair, steel sheet, car etc.
o Size of unit must be constant

• Example: Count no of defects (scratches, chips etc.) in each chair of a

sample of 100 chairs; Plot
• The p and np chart control the fraction defective in the product
whereas the c chart controls the number of defects in the product. In
other words , the control chart for defects is called as c chart.
 c Chart
Control Limits
• c- no. of defects
• n- no. of samples

c 
 c
n
UCLc  c  3 c
Use 3 for
LCLc  c  3 c 99.7% limits
CLc  c
 u Chart
• u chart is used in case of no. of defects observed in
sample when the sub group size varies from sample to
sample u= no. of defects in a
• u chart control no. of units in a
the no. of defects per unit
sample(c)
sample(n)
UCLu  u  3 u n
LCLu  u  3 u n
CLu  u
 Use of p,np,c,u chart
Type Use
p Chart Finding fraction defectives as non-conforming
specification
np Chart For defectives

c Chart To control no. of, defects per unit when

subgroup size is normal
u Chart To control average no. of, defects per unit when
subgroup size is not constant(i.e)subgroup size
varies from sample to sample
 Acceptance sampling
The process of randomly inspecting a sample of goods and
deciding whether to accept the entire lot based on the results.

Acceptance sampling is useful in following cases:

• When testing method is destructive
• When it is not affordable to go for 100% inspection(due to time or cost
involved)
• When the vendor has an excellent record of performance in supplying
quality products
Advantages of Acceptance sampling
Advantages
• Less expensive because there is less inspection
• Reduce level of damage to products, as there is less inspection
• Number of people required for inspection is less
• Inspection error is usually very less
• When a lot is rejected based on acceptance sampling, the vendor is tempted to improve
his process capability
Disadvantages
• There are possibilities for accepting bad lots and rejecting good
lots
• Acceptance sampling requires proper planning and
documentation, which is not required for 100% inspection
 Process Capability
• Tolerances or Specifications
indicate a range of values within which individual units of output must fall in
order to be acceptable.
The design team decides tolerances based on customer requirements
• Control limits
Statistical limits up to which sample statistics such as means and ranges can
vary due to randomness of process
• Process variability
Measure that indicates natural or inherent(i.e. random)
variability in process.
measure in terms of process SD
 Process Capability-
Calculation
Process Capability(Cp) = Specification width
process width

Cp = upper specification limit-lower

 specification limit
6
 1. Process flow chart:
• Diagrammatic view of the various steps in sequential order
that form an overall process in an organization.
• Steps:
o Define the process.
o List the steps involved.
o Sequence of process.
o Analyse the flow chart.
 Analysis:
Points to checked at the micro level
o Are all the steps are necessary?
o If any step missing?
o Which are the points delay?
o Can the process be simplified?
Point to be checked for every input
o Are there specified, clear, and agreed standards?
o Is there adequate feedback?
 Process steps:
• Is the step necessary.
• Are facilities/ equipment adequate?
• Do the operators have necessary skills.
• Possible to measure the performance of product.
• What impact a detect in the process will have.
Output:
• Are there specified, clear and agreed standards.
• Is there adequate feedback.
Advantages of Process flow chart:
• Employee who work in the process understand all the process.
• Can check against poor quality.
• To analyze and eliminate unnecessary process.
• Improves the communication between departments.
• It is very useful for training new employees.
• Understand the employees to realize how they fit in the over
all process.
 2.CAUSE AND EFFECT DIAGRAM(Fishbone
Diagram).
• It is a graphical-tabular chart to list and analyze the potential
causes.
• The objective of the diagram is to find out ways to reduce the
cycle time.
• Ishikawa diagram after the man who devoloped it in 1943.
 When do we use it?
• To analyze cause and effect relationship
• To facilitate the search for solutions of related problems
• To standardize existing and proposed operations, and
• To educate and train personnel in decision making and
corrective action activities
 How do we construct it?
• Define the effect clearly and concisely
• Mark the short description of the effect in a box. Then draw a line
from this box towards left. skeleton diagram
• List down all the possible minor and major causes through a
brainstorming session.
• Mark the major causes on the branches and minor on the sub
branches of the CE diagrams.
• Look for possible solutions for these causes.
• Introduce the changes.
 3. Check sheet(data collection sheet)-tally sheet
• It is a systematic data gathering and registering to get a clear view of the facts.
• Used to track of how often something occurs.
• Used to indicate the frequency of a certain occurance.
How do we construct it?
• Formulate the objective
• Decide which data is necessary.
• Determine who and how data will be analyzed.
• Draw format to record data.
• Collect and record data problem-wise by putting tally lines
• Count by tallying on the list & Represents the numbers by 1,2,34..,
• Mark on the list the total numbers of facts, which were noticed.
 Types of check sheets
• Process distribution check sheet
o To collect process variability

• Defective item check sheet

o Specify the variety of defects and with their frquency of occurrence

• Defect location check sheet

o Identify where the defects occur on the prodeuct

• Defect factor check sheet

o Monitor the input parameter
3. Check sheet(data collection
sheet)-tally sheet
 4.Histogram
• It is a bar chart/diagram showing a distribution of variable quantities or
characteristics.
• It display frequency distribution of the numerical data.
• Data displayed as a series of rectangles of equal width and varying heights.
When do we use it?
• Used to show clearly where the most frequent occurring values are located
and the data is distributed.
• It is a tool for determining the max. process results.
• It enables the analyst to quickly visualize the feature of a complete set of data.
 Cont..
How do we construct?
• After the data collection, count the number of data values collected.
• Determine the range of the data.
Range = highest value- lowest value
• Divide the data values in groups or classes and count the number of values in each class. Ex:

No. of values No. of classes No. of values No. of classes

Less than 50 5-7 100-250 7-12
50-100 6-10 More than 250 10-20
• Now determine the width of the class.
width of the class=Range/No. of classes selected from the table.
• draw frequency table for all values.
• Construct histogram based on the frequency table. Class limit in horizontal axis and frequency in vertical axis.
• Finally write the title and no. of values on the diagram.
5. Pareto diagram(80/20 rule and as
ABC analysis)
• A Pareto is a diagnostic tool commonly used for separating the vital
few causes that account for a dominate share of quality cost.
• It is based on pareto principle, which states that a few of the defects
accounts for most of the effect.
• Only 20% of problem account for 80% of the effects.
When do we use it?
• Situation where one need to prioritise based on its relative
importance.
• Used as a risk assesment technique from activity level to system level.
 How can we construct?
• Obtain data, using a check sheet or brainstorm.
• Arrange the data in descending order starting from the largest category to smallest.
• Calculate the total and percentage of the total that each category represents.
• Compute the cumulative percentage.
• Draw a bar chart with two` vertical axes. Along the left vertical axis, mark the measurd values
for each cause, starting from zero till the total number of causes. The right vertical axis should
have the same height and should go from 0 to 100%. This axis displays the cumulative
percentages. List the different kind of causes along the horizontal axis, from left to right in
descending order of frequency or costs.
• Draw a bar above each item whose height represents the number for that cause.
• Plot a cumulative percentage line.
• Now draw a horizontal line from 80% (on the right vertical axis) to the left till the point of
intersection with the cumulative line, and then draw a vertical line from this intersection
downwards till the horizontal axis. Left from this intersection point are the 20% of causes(the
most essential bottlenecks) which causes 80% of the damages
 6. Scatter diagram
• Graphical device to depict the relationship between two variables.
When do we use it?
• To display what happens to one variable when another variable is changed.
• Used to understand, why particular variations ouur and how they can be controlled.
How do we construct it?
• Select variables which are related to each other.
• Collect the data.
• Create the scatter diagram
o X- axis: cause or independent variable
o Y- axis: effect or dependent variable

• Examine the shape of cloud of points

• Determine the type and strength of the mutual relationships.
Scatter dia..
 7. Control chart
• Invented by walter A. Shewart
• Widely used tool in the statistical process control(SPC)
• It is Graphical that displays data taken over time.
• It is used to check process being controlled.
• It consists of 3 horizontal lines remains const. over time
o Center
o Lower control limit(LCL)
o Upper control limit(UCL)
 Types of control chart
• For variables- time, length, temp, weight, pressure
• For charecteristics- quantifiable data such as number of
defects, typing errors ina report
 How do we use it?
• Identify when the process id gone out of statistical control-
corrective action to be taken
• Used to show sources of variations and to eliminate them.
• Used to evaluate process stability and to adjust the process
How do we construct?
 Quality Control
Tool

Graphs
 Types of graph
Bar graph Band chart

Pie chart

Line graph
Gantt chart
 their uses
Graph type Characteristics & Application

Bar graph Constructed with parallel bars of

identical width but differing length. This
is used to compare size of two quantities

Line graph Reveals overall trend of data. This is used

to look at direction of changes over a time
frame
Pie chart This is used when breaking up of
components of quantity is to be changed

Band chart Placing two or more band graphs in

 Statistical Quality
Control
Definition
Statistical quality control (SQC) is the term used to
describe the set of statistical tools used by quality professionals.

Statistical quality control can be divided into three broad

categories:
• Descriptive statistics
• Statistical process control (SPC)
• Acceptance sampling
 Descriptive Statistics
Descriptive Statistics is the branch of statistics
which most people are familiar.

It characterizes and summarizes the most

prominent features of a given set of data
(means, medians, standard deviations,
percentiles, graphs, tables and charts. 11/2/21 63
 WHAT IS THE MEAN?
ORDERED DATA SET
The mean is simply the average value of the -5
data. -3

mean = x = ånxi = -
-1
2 = -.17 -1
12 0
0
n=12 0
0
0
1
3
-6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 4
Mean åxi = -2
11/2/21 64
WHAT IS THE MEDIAN?
If we rank order (descending or ascending) the data set ORDERED DATA SET
,we find the value half way (50%) through the data points -5
and is called the median value. -3
-1 50% of
data
-1 points
0
0
Median 0
0
0
1
3
-6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 4
Median
Value
11/2/21 65
 WHAT IS THE MODE?
If we rank order (descending or ascending) the data set
ORDERED DATA SET
We find that a single value occurs more often than any -5
.other. This is called the mode. -3
-1
-1
0
0
Mode Mode 0
0
0
1
3
-6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 4
11/2/21 66
 WHAT IS THE RANGE?
The range is a very common metric . ORDERED DATA SET
-5
To calculate the range simply subtract the minimum value -3
in the sample from the maximum value. -1
-1
Range = x MAX - x MIN = 4 - (-5) = 9 0
0
0 Range
0
0
1
3
-6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 4
Range
Min Max
11/2/21 67
 WHAT IS THE VARIANCE/STANDARD DEVIATION?
The variance (s2) is a very robust metric .
The standard deviation(s) is the square root of the variance and is the most
commonly used measure of dispersion.  Xi  X  2
åXi - 2 -.17
DATA SET X -Xi
X = = = -5 -5-(-.17)=-4.83 (-4.83)2=23.32
n 12 -3
-1 -3-(-.17)=-2.83 (-2.83)2=8.01

= å(n - 1 ) = 12 - 1 = 56
2
2 X -X i 6167
.
-1
0
-1-(-.17)=-.83 (-.83)2=.69
s . 0 -1-(-.17)=-.83 (-.83)2=.69
0 0-(-.17)=.17 (.17)2=.03
0
0 0-(-.17)=.17 (.17)2=.03
1
3 0-(-.17)=.17 (.17)2=.03
-6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 4 0-(-.17)=.17 (.17)2=.03
0-(-.17)=.17 (.17)2=.03
1-(-.17)=1.17 (1.17)2=1.37
11/2/21 68
 Types of Measures
• Measures where the metric is composed of a classification in one of
two (or more) categories is called Attribute data.

_ Good/Bad
o Yes/No

• Measures where the metric consists of a number which indicates a

precise value is called Variable data.
o Time
o Miles/Hr
11/2/21 69
New- Seven Management Tools
 The New Seven Quality Tools
The New Seven Quality Tools are used in following areas

o Product quality improvement

o Cost reduction
o New product development
o Policy development
 The New Seven Quality Tools
• Affinity Diagrams
• Interrelationship Diagrams
• Tree Diagrams
• Matrix Diagrams
• Matrix Data Analysis
• Process Decision Program Charts
• Arrow Diagrams
 Affinity Diagrams
• Organizes a large amount of verbal data related to a
broad problem or subject
o Ideas, opinions, facts

• Usage example: Establishing a new QC policy

• Steps:
o Gather a large number of ideas

o Put individual ideas on cards or sticky notes

o As a team, group the ideas according to natural “affinity” or relationship to each other

o These natural groups become “strategic factors”

 Affinity Diagram Example
Your team has been brainstorming to develop a list of ideas to incorporate
into the vision. They have come up with the following list. Develop an
affinity diagram and name each strategic factor.
•Low product maintenance
•Satisfied employees •Low production costs

•Courteous order entry •Innovative product features

•Low prices
•Quick delivery
•Growth in shareholder value
•Teamwork
•Responsive technical support
•Personal employee growth
•High return on investment
•Constant technology innovation
•High quality
•Motivated employees
•Unique products
•Small, lightweight designs
Affinity Diagram Example Cont.
 Interrelationship Diagrams
• Identifies and explores causal relationships among related concepts or ideas.
Can address problems with a complex network of causes and effects.
o Identifies key drivers and bottlenecks
• Usage examples: design steps to counter market complaints, or reform
administrative departments
• Steps:
o Write each concept or idea on a piece of paper in a circular pattern (allow
room between concepts)
o Number them to make comparison process easier to track
o Use pairwise comparisons (1-2, 1-3, 1-4…2-3, 2,4…3,4)
• If there is a relationship draw arrow to effect
• If there is no relationship leave blank
• The can be no 2-way relationships
 Interrelationship Diagrams Cont.
• Steps (Cont.)
o Analyze the diagram
• Count the arrows (# out (cause) - # in(final effect (critical)))
o Highest out are primary drivers
• Resources here can produce pronounced change
o Lowest are key bottlenecks
• Affected by many other options
• May be inhibiting other options from proceeding as
required
• Highlight primary drivers and key bottlenecks
• Note: examine only cause and effect relationships. Likely will have
arrows on only 50% of relationships.
 Interrelationship Diagram Example
• Use the strategic factors derived from your affinity
diagram to develop an interrelationship diagram. If
you were unsuccessful in developing your own
strategic factors use the following:
o Customer Value
o Work Environment
o Customer Service
o Technology
o Product Innovation
Interrelationship Diagram Example Cont.
 Tree Diagrams
• Expands a purpose into the tasks required to
accomplish it.
• Usage examples: deploy a quality plan, or develop
objectives, policies and implementation steps.
• Steps:
Tree Diagrams (cont.)
o Work from left to right
o Start with the purpose to be accomplished
o Generate the high level targets or goals that must be completed to
accomplish the purpose
o Link each goal to the purpose (these are the first branches of the
tree)
o Expand on each target to identify and define subordinate tasks to
accomplish each target
o Link each to their target
o Continue expansion process until final level is implementable.
o Review logic of completed tree (perhaps with larger group)
Tree Diagram Example Cont.
Key Strategic Factor Goals Strategies
 Matrix Diagrams
• S.M.A.R.T. Plan Matrices
o Technique for structuring the task details when planning the implementation of a
project.
o May use the final output of a tree diagram
o For each implementable task:
• Specific (activity or task)
• Measurable (outcome or process)
• Assignment (who will perform)
• Resources (what is needed)
• Time (anticipated duration)
• Predecessors (what must be done first)
o Consensus should be reached among all parties on the SMART matrix
• Correlation Matrices
o Shows the relationship between one list of variables and another. Relationships are
usually based on experience.
o Such a matrix forms the body of a “house of quality”
Types of matrix diagram
 5. Matrix Data Analysis
•  The purpose of matrix data analysis diagram is to present
numerical data about two sets of factors in a matrix form and analyze
it to get numerical output.
• Arranges a large array of numbers so that they may be visualized and
comprehended easily
• Usage example: evaluate the desired quality level from the results of a
market survey
• Steps:
o Begin with numerical matrix relating goals or requirements to actions or performance
o Assign weights to each goal or requirement
• Subjective
• Objective (principle component analysis)
 Matrix Data Analysis Example
Requirement Importance Best Competitor Own Weighted Gap
Weight Evaluation Evaluation

Price .2 6 7

Speed of Delivery .3 7 6

Reliability .4 5 6
Customizability .1 8 7
 Matrix Diagram S.M.A.R.T. Plan Example
Specific Measurable Assignment Resources Time Predecessors
(Weeks)
A Evaluate Deliverable Steve 8 hours 1 -
Needs
B Schedule Deliverable Doug 4 hours 1 -
Training
C Evaluate Deliverable Morgan 10 hrs, copies 2 A
Software of software

D Training Deliverable Doug 20 hrs, 3 C

Materials software
manuals
E Purchase Deliverable Ted 2 hours 2 C
F Install # systems Ted 50 hours 2 E
G Train Users # trained Doug 20 hours 1 B, D, F
Types of matrix diagram
  = Strong relationship

Correlation Matrix  = Medium relationship

 = Weak relationship
Example
Actions Improve
Improve Work Develop New
Manufacturing
Goals Environment Products
Technology

Cost Effectiveness  
High Quality  
Shareholder Value  