Quality Control &

Assurance (M 1581)
Process capability
and Xbar-R Chart
Dr. Abdelrahman Elgendy
Department of Mechanical Engineering, Benha University

1/33
 The meaning of Quality
There are many definitions of Quality, Some
defined quality as:
◼ Quality means fitness for purpose.
◼ Quality means delivering what the customers want
◼ Quality means conformance to specifications
◼ Etc.

2
 VOC and VOP
Voice of Customer (VoC)
◼ VoC is also called a Customer
Standard or Specification
requirement.
◼ The customer usually provides a
specification that has a mean or target
value.
◼ The customer also provides tolerance The customer expects the
part/product supplies to follow a
limits to account for process variation: normal distribution
◼ LSL- Lower Specification Limit and
◼ USL- Upper Specification Limit
3
VOC and VOP
◼ Voice of Process (VoP)
◼ VoP is the internal perspective,
focusing on how well the production or
business processes are performing.
◼ VoP aims to ensure that the internal
processes are stable, efficient, and
capable of consistently delivering quality
products.
◼ VoP drives continuous improvement
efforts within the manufacturing or
service processes, leading to better
process control and performance. 4
5
Process variation

Process variation

6
Process Variation

7
 Process capability
◼ Will the production process meet the
requirements?
◼ It is important to produce products within
the specification, the distance between the
upper specification limit (USL) or upper
tolerance (+T) and lower specification limit
(LSL) or lower tolerance (–T), i.e. (USL–
LSL) or 2T must be equal to or greater
than the width of the base of the process
bell, i.e. 6 Ϭ.

8
 Process capability
◼ Importance of Process capability analysis
◼ Predicting how well the process will hold the
tolerances,
◼ Process selection,
◼ Process improvement,
◼ Specifying performance requirements for new
equipment,
◼ Selecting between competing suppliers,
◼ Reducing the variability in a process.
9
 Process capability
◼ How to measure the process capability?
◼ Two-Sided Specification Limits.
◼ Centered Processes (Cp)
◼ Off-Center Processes (Cpk)
◼ One-Sided Specification Limits (Cpl or Cpu)

10
 Process Capability Ratio
◼ In order to manufacture within a specification, the difference
between the USL and the LSL must be greater than the total
process variation.

◼ An estimate of the standard deviation may be obtained from any set of

data

11
 Example
◼ • The ice cream that must be served in an ice cream
parlor has to be between -15 degrees Celsius and -35
degrees Celsius. The process of refrigeration that
keeps the temperature has a standard deviation (SD) of
2 degrees Celsius. And the mean value of this
temperature is -25 degrees Celsius. Using these inputs,
calculate the process capability?

12
Example

13
Process Capability Index Cpk

14
 Example 2
◼ The training room temperature during the training
must be kept between 18 degrees Celsius and 26
degrees Celsius. The process of maintaining this
temperature has a standard deviation of 2 degrees
Celsius. If the mean temperature is 21 degrees Celsius,
calculate the process capability.

15
16
17
18
19
20
Control Charts
Xbar-R Chart

21/33
 In-control and
Out-of-control process
◼ In-control process
◼ Stable, predictable process
◼ Chance causes of variation
◼ Acceptable variation
◼ Out-of-control process
◼ Unpredictable
◼ Assignable causes of variation
◼ Unacceptable variation
22
23
 Variation
◼ Variation can be categorized as the result of:
◼ Chance/Common/Random causes of variation
◼ Causes that we cannot identify

◼ Unavoidable causes

◼ e.g. slight differences in process variables like diameter,

weight, service lime. temperature, machine accuracy

◼ Assignable causes of variation
◼ Causes can be identified and eliminated

◼ e.g. poor employee training, worn tool, machine needing

repair, differences among machines, people, materials, etc

24
Commonly Used Control Charts
◼ Variables data
◼ x-bar and R-charts
◼ x-bar and s-charts
◼ Charts for individuals (x-charts)
◼ Attribute data
◼ For "defectives" (p-chart, np-chart)
◼ For "defects" (c-chart, u-chart)

25
 Constructing Control Charts
◼ Prepare
◼ Choose the quality characteristics
◼ Determine how to collect data, sample size, and sampling frequency
◼ Collect the data
◼ Determine trial control limits
◼ Center line , Compute UCL, LCL
◼ Plot the sample data
◼ Analyze and interpret results
◼ Determine if the process is in control
◼ Eliminate out-of-control points (due to special causes), If any
◼ Recompute control limits as necessary
◼ Go to A.2 26
 Control Charts for Variables
◼ Variables are characteristics that can be measured
◼ dimensions, weight, hardness, operating temp, time, service
life,
◼ Control charts are statistical tools to show whether a
process is in control or not. They show the behavior of
the process over time

27
How to Detect Special Causes
◼ Since both mean and variability may be
changing over time, we need to collect
samples and do calculations that tell us
ത
about centering (𝑋)and spread (R)

28
Data analysis

29
 Data Analysis
◼ The sample mean is an estimate of the population
mean
◼ The sample variance is an estimate of the population
variance
◼ We use the sample data to predict the process
behavior and evaluate it
◼ What percent is scrap?
◼ What percent needing rework?
◼ Is process centered and stable?

30
 X-bar and R Charts
◼ Most commonly used control charts
◼ Construction procedures
◼ Select the quality characteristic
◼ Select the samples: 25 to 50 samples
◼ Sample size should be small; between 3 to 6 (5 is very
common)
◼ Large sample size causes the control limits to be closer to the center
line and the chart becomes very sensitive to small variations in the
average
◼ Small sample size leads to higher inspection costs
◼ How frequently the samples are taken should be decided such that it can
reflect the potential for changes in the process 31
 X-bar and R Charts
Statistical Basis of the Charts

32
33
 X-bar and R Charts
◼ Construction procedures
◼ Plot the charts. X-bar chart on top and R-chart below it
◼ Study the chart to identify any out-of control signals
◼ Eliminate the special causes of the out-of-control signals from
the process (take appropriate corrective actions)
◼ Remove the samples corresponding to the out-of-control
signals from the collected data
◼ Use the remaining samples to calculate the revised control
limits and center lines

34
 X-bar and R Charts
◼ Construction procedures
◼ Revised center lines and control limits represent the process
after having been adjusted and they are used to monitor the
process in the future

◼ 𝑋𝑑 and 𝑅𝑑 , refers to the discarded

sample means and ranges
◼ 𝑘𝑑 refers to the number of
discarded samples 35
X-bar and R Charts

36
 A Process in Good Statistical Control
◼ No point exceed the control limits
◼ Points are approximately normally distributed within
the control limits
◼ Points show no evidence of trends or cycles
◼ Points look quite random with no patterns or runs

37
Typical Out-of-Control Patterns
◼ Point outside control limits
◼ Sudden shift in process average
◼ Cycles /Trends
◼ Hugging the center line
◼ Hugging the control limits
◼ Instability

38
Typical Out-of-Control Patterns

39
 Uses of the Control Charts
◼ Off-line use
◼ To Identify when special causes enter the process and to
characterize the level of common-cause variability
◼ Where to look for improvement opportunities
◼ Help to formulate and assess effect of actions
◼ On-line use
◼ Provide a sound economic basis for operators to make decision
at the machine as to when to adjust the machine (and when to
leave it alone

40
 Benefits of Control Charts
◼ When properly used, control charts can
◼ Be used by operators for ongoing control of a process
◼ Help the process perform consistently, predictably, for quality
and cost
◼ Allow the process to achieve
◼ Higher quality , Lower unit cost and Higher effective capability
◼ Provide a common language for discussing the performance of
the process
◼ Distinguish special from common causes of variation, as a
guide to local action or action on the system
41
 Example
A hard-bake process is used in conjunction with photolithography
in semiconductor manufacturing. We wish to establish statistical
control of the flow width of the resist in this process using and R
charts. Twenty-five samples, each of size five wafers, have been
taken when we think the process is in control. The interval of time
between samples or subgroups is one hour. The flow width
measurement data (in microns) from these samples are shown in
Table 6.1.

42
43
44
45
 The specification limits on flow width are 1.50 ±0.50 microns.

This implies that the “natural” tolerance limits in the process (three-sigma above and
below the mean) are inside the lower and upper specification limits. Consequently, a
moderately small number of nonconforming wafers will be produced.

n-gl.com 46