UNIT 4 NOTES

ME317, SIX SIGMA

CONTROL CHARTS USED FOR ATTRIBUTES AND VARIABLES

There are two main types of control charts used in statistical process control: attribute control
charts and variable control charts. Each type of chart is designed to monitor a different type of
process characteristic, and they use different statistical techniques to do so.

1. Attribute Control Charts:

Attribute control charts are used to monitor the proportion or rate of defective items in a process,
which are also known as "attributes". They are used when the process output can be classified into
discrete categories, such as pass/fail, good/bad, or conforming/non-conforming. The following are
the main types of attribute control charts:

• p-Chart: The p-Chart is used to monitor the proportion of defective items in a process over
time. It is based on the binomial distribution, and assumes that the probability of a
defective item is constant over time. The chart plots the proportion of defective items in a
sample against the sample number or time. The center line represents the expected
proportion of defects, and the control limits are based on the binomial distribution.
• np-Chart: The np-Chart is similar to the p-Chart, but is used when the sample size is
constant. It plots the number of defective items in a sample against the sample number or
time. The center line represents the expected number of defects, and the control limits are
based on the Poisson distribution.
• c-Chart: The c-Chart is used to monitor the number of defects per unit of output, when the
sample size varies. It plots the number of defects in a sample against the sample number
or time. The center line represents the expected number of defects per unit, and the control
limits are based on the Poisson distribution.
• u-Chart: The u-Chart is used to monitor the number of defects per unit of output, when
the sample size is constant. It plots the number of defects in a sample divided by the sample
size against the sample number or time. The center line represents the expected number
of defects per unit, and the control limits are based on the Poisson distribution.
2. Variable Control Charts:

Variable control charts are used to monitor process characteristics that can be measured on a
continuous scale, such as length, weight, temperature, or time. They are used when the process
output can be measured quantitatively, and are based on statistical distributions. The following are
the main types of variable control charts:

• X-bar and R-Chart: The X-bar and R-Chart are used to monitor the process mean and
variability, when the sample size is constant. The X-bar chart plots the sample mean against
the sample number or time, while the R-chart plots the range of the sample. The center
line of the X-bar chart represents the expected process mean, while the control limits are
based on the normal distribution. The center line of the R-chart represents the expected
range of the sample, while the control limits are based on the distribution of the range.
• X-bar and s-Chart: The X-bar and s-Chart are used to monitor the process mean and
variability, when the sample size varies. The X-bar chart plots the sample mean against the
sample number or time, while the s-chart plots the sample standard deviation. The center
 line of the X-bar chart represents the expected process mean, while the control limits are
based on the t-distribution. The center line of the s-chart represents the expected process
standard deviation, while the control limits are based on the chi-square distribution.
• Individuals and Moving Range (I-MR) Chart: The Individuals and Moving Range (I-MR)
Chart is used to monitor the process mean and variability, when the sample size is 1. The
chart plots the individual data points against the sample number or time, and the moving
range of the data points. The

1. Attribute Control Charts:

Attribute control charts are used to monitor the proportion or rate of defective items in a process.
This type of chart is used when the process output can be classified into discrete categories, such
as pass/fail, good/bad, or conforming/non-conforming. The following are the main types of
attribute control charts:

• p-Chart: The p-Chart is used to monitor the proportion of defective items in a process over
time. It is based on the binomial distribution, and assumes that the probability of a
defective item is constant over time. The chart plots the proportion of defective items in a
sample against the sample number or time. The center line represents the expected
proportion of defects, and the control limits are based on the binomial distribution.

For example, if a manufacturing process is producing 100 widgets per hour, and the p-chart
indicates that 5% of the widgets are defective, then the manufacturer can use the chart to
determine if the proportion of defective widgets is increasing or decreasing over time.

• np-Chart: The np-Chart is similar to the p-Chart, but is used when the sample size is
constant. It plots the number of defective items in a sample against the sample number or
time. The center line represents the expected number of defects, and the control limits are
based on the Poisson distribution.

This chart is useful when the process output is in discrete categories but the sample size remains
constant over time. For example, a restaurant may use an np-chart to monitor the number of
customer complaints received per day.

• c-Chart: The c-Chart is used to monitor the number of defects per unit of output, when the
sample size varies. It plots the number of defects in a sample against the sample number
or time. The center line represents the expected number of defects per unit, and the control
limits are based on the Poisson distribution.

This chart is useful when the process output is in discrete categories, but the sample size varies
over time. For example, a manufacturing process may use a c-chart to monitor the number of
defects per batch of product produced.

• u-Chart: The u-Chart is used to monitor the number of defects per unit of output, when
the sample size is constant. It plots the number of defects in a sample divided by the sample
size against the sample number or time. The center line represents the expected number
of defects per unit, and the control limits are based on the Poisson distribution.
This chart is similar to the c-chart, but is used when the sample size remains constant over time.
For example, a quality control team may use a u-chart to monitor the number of defects per unit
of a product produced.

2. Variable Control Charts:

Variable control charts are used to monitor process characteristics that can be measured on a
continuous scale, such as length, weight, temperature, or time. They are based on statistical
distributions, and are used when the process output can be measured quantitatively. The following
are the main types of variable control charts:

• X-bar and R-Chart: The X-bar and R-Chart are used to monitor the process mean and
variability, when the sample size is constant. The X-bar chart plots the sample mean against
the sample number or time, while the R-chart plots the range of the sample. The center
line of the X-bar chart represents the expected process mean, while the control limits are
based on the normal distribution. The center line of the R-chart represents the expected
range of the sample, while the control limits are based on the distribution of the range.

For example, a company may use an X-bar and R-chart to monitor the average length and
variability of bolts produced on

Control charts are statistical tools that are used to monitor a process over time. They can help to
identify and correct any issues with the process before they become major problems. Interpreting
control charts involves analyzing the data points on the chart and looking for patterns or trends
that may indicate that the process is out of control. The following are the steps involved in
interpreting control charts:

1. Determine the Control Limits: The first step in interpreting a control chart is to determine
the control limits. These are the upper and lower bounds of the data that indicate when a
process is considered out of control. The control limits are calculated based on the data
collected from the process, and can be derived using various statistical methods.
2. Look for Points Outside the Control Limits: The next step is to look for any data points that
fall outside the control limits. These points are called "outliers" and indicate that the process
is not performing as expected. Outliers can be caused by factors such as measurement
error, equipment malfunction, or operator error. If an outlier is detected, it should be
investigated to determine the cause of the problem.
3. Look for Patterns in the Data: In addition to outliers, control charts can also show patterns
or trends in the data. These patterns can indicate that the process is not performing as
expected, even if all data points fall within the control limits. The following are the common
patterns observed in control charts:
• Trend: A trend is a gradual change in the process mean over time. It can be indicated by a
series of data points that increase or decrease over time. A trend can indicate that the
process is changing, and may require corrective action.
• Cycle: A cycle is a pattern that repeats itself over time. It can be indicated by a series of
data points that fluctuate between upper and lower limits. A cycle can indicate that the
process is affected by seasonal or other periodic factors, and may require adjustment.
 • Run: A run is a sequence of data points that fall on one side of the center line. A run can
indicate that the process is consistently biased in one direction, and may require
adjustment.
• Stratification: Stratification is a pattern that occurs when the data is grouped into
subcategories, such as shifts or operators. It can indicate that the process is affected by
factors that vary between subcategories, and may require investigation.
4. Determine if the Process is in Control: The final step in interpreting a control chart is to
determine if the process is in control. If all data points fall within the control limits and there
are no patterns or trends in the data, then the process is considered to be in control. If
outliers or patterns are detected, then the process may be out of control and require
corrective action.

In summary, interpreting control charts involves analyzing the data points, looking for outliers,
patterns and trends in the data, determining if the process is in control or out of control, and taking
corrective action if necessary. Control charts provide a valuable tool for monitoring and improving
the quality of a process, and can help to ensure that products and services are delivered
consistently and reliably.
Process capability index is a statistical measure used to assess the ability of a process to
consistently produce products or services within a set of predetermined specifications. It measures
the process performance in terms of its ability to meet the customer requirements and
expectations. Process capability index is often used in manufacturing and service industries to
evaluate the quality of their processes and identify areas for improvement.

The two commonly used process capability indices are Cpk and Cp.

• Cp is a measure of the potential capability of a process. It is calculated by dividing the

allowable tolerance width by the process variation. A process with a high Cp value indicates
that it is capable of producing products or services within the desired specifications.
• Cpk is a measure of the actual capability of a process. It takes into account the deviation of
the process mean from the target value, and it is calculated by taking the minimum value
of the difference between the upper and lower specification limits divided by the process
variation. A process with a high Cpk value indicates that it is capable of producing products
or services that meet the desired specifications consistently, even in the presence of
process mean shifts.

Both Cp and Cpk values range from 0 to 1. A value of 1 indicates that the process is perfectly
capable of producing products or services within the desired specifications, while a value of 0
indicates that the process is not capable at all.

Process capability index is an important tool for quality control and continuous improvement. By
monitoring and analyzing the process capability indices, organizations can identify areas where
improvements are needed and implement corrective actions to improve process performance and
meet customer expectations.

Process capability and performance indices are estimated using statistical methods to analyze
the data collected from a process. The data may be collected from a sample of products or services
produced by the process, or from the entire population of products or services produced by the
process.
The following are the steps involved in estimating process capability and performance indices:

1. Define the specifications: The first step is to define the upper and lower specification limits
for the process. The specifications are usually based on the customer requirements and
expectations.
2. Collect data: The next step is to collect data on the process. The data may include
measurements of process variables, such as length, width, height, weight, or time.
3. Calculate process mean and standard deviation: The process mean and standard deviation
are calculated from the data collected. The mean represents the average value of the
process, while the standard deviation represents the variability of the process.
4. Calculate capability indices: The capability indices, Cp and Cpk, are calculated using the
following formulas:

Cp = (USL - LSL) / 6σ

Cpk = min[(USL - Mean) / 3σ, (Mean - LSL) / 3σ]

where USL is the upper specification limit, LSL is the lower specification limit, and σ is the standard
deviation of the process.

Cp measures the potential capability of the process to meet the specifications, while Cpk measures
the actual capability of the process to meet the specifications.

5. Interpret the results: The results of the capability indices can be interpreted as follows:
• A Cp value of 1 or greater indicates that the process is capable of meeting the
specifications.
• A Cpk value of 1 or greater indicates that the process is capable of meeting the
specifications, even in the presence of mean shifts.
• A Cp or Cpk value less than 1 indicates that the process is not capable of meeting the
specifications.

Process capability and performance indices are important tools for quality control and continuous
improvement. They provide valuable information about the performance of a process and help
organizations identify areas where improvements are needed to meet customer requirements and
expectations.

MORE DETAILED EXPLANATION OF HOW PROCESS CAPABILITY AND PERFORMANCE

INDICES ARE ESTIMATED:

1. Define the specifications: The first step in estimating process capability and performance
indices is to define the upper and lower specification limits for the process. These are the
acceptable range of values or conditions that the products or services produced by the
process must meet in order to satisfy customer requirements. The specifications may be
based on factors such as safety, reliability, functionality, or appearance, depending on the
nature of the product or service.
2. Collect data: Once the specifications have been defined, data on the process must be
collected. This can be done by taking measurements of the relevant variables using
appropriate tools and equipment. For example, if the process involves manufacturing
 components of a certain size, measurements of length, width, and thickness may be taken.
If the process involves providing a service, measurements of time taken to complete the
service may be recorded.
3. Calculate process mean and standard deviation: The data collected is used to calculate the
process mean and standard deviation. The mean represents the average value of the
process, while the standard deviation represents the variability or spread of the process
data around the mean. The mean and standard deviation can be calculated using statistical
software or formulas.
4. Calculate capability indices: Once the process mean and standard deviation have been
calculated, the capability indices can be estimated. The two commonly used indices are Cp
and Cpk. Cp measures the potential capability of the process to meet the specifications,
while Cpk measures the actual capability of the process to meet the specifications, taking
into account the deviation of the process mean from the target value.

Cp is calculated using the formula:

Cp = (USL - LSL) / 6σ

where USL is the upper specification limit, LSL is the lower specification limit, and σ is the standard
deviation of the process. A Cp value of 1 or greater indicates that the process is capable of meeting
the specifications.

Cpk is calculated using the formula:

Cpk = min[(USL - Mean) / 3σ, (Mean - LSL) / 3σ]

where Mean is the process mean. Cpk takes into account both the variability of the process and
the deviation of the process mean from the target value. A Cpk value of 1 or greater indicates that
the process is capable of meeting the specifications, even in the presence of mean shifts.

5. Interpret the results: Once the capability indices have been calculated, they can be
interpreted to assess the performance of the process. A Cp or Cpk value less than 1
indicates that the process is not capable of meeting the specifications, and improvements
may be needed to bring the process into compliance. A Cp or Cpk value of 1 or greater
indicates that the process is capable of meeting the specifications, and the closer the value
is to 1, the more capable the process is.

In summary, process capability and performance indices are estimated by defining the
specifications, collecting data on the process, calculating the process mean and standard deviation,
and then calculating the capability indices. The results can be used to assess the performance of
the process and identify areas for improvement.

POINT ESTIMATE OF CAPABILITY AND PERFORMANCE INDICES

A point estimate of capability and performance indices is a single numerical value that is used to
estimate the process capability or performance based on a sample of data collected from the
process.
Point estimates are used when the entire population of products or services produced by the
process is not available for analysis, and only a sample of data is collected. In this case, the point
estimate is used as an approximation of the true process capability or performance.

The point estimate for the capability index Cp is calculated using the formula:

Cp = (USL - LSL) / 6σ

where USL is the upper specification limit, LSL is the lower specification limit, and σ is the standard
deviation of the sample.

Similarly, the point estimate for the capability index Cpk is calculated using the formula:

Cpk = min[(USL - Xbar) / 3s, (Xbar - LSL) / 3s]

where Xbar is the sample mean, s is the sample standard deviation, USL is the upper specification
limit, and LSL is the lower specification limit.

It is important to note that point estimates are subject to sampling error, which is the difference
between the point estimate and the true process capability or performance. The larger the sample
size, the smaller the sampling error and the more accurate the point estimate.

To reduce the sampling error, it is recommended to use multiple samples and calculate the average
of the point estimates. This is known as the average capability or performance index.

In summary, a point estimate of capability or performance index is a single numerical value

calculated from a sample of data collected from a process. It is an approximation of the true
capability or performance, and is subject to sampling error. To reduce the sampling error, multiple
samples can be used to calculate the average capability or performance index.

CONFIDENCE INTERVAL FOR CAPABILITY AND PERFORMANCE INDICES

A confidence interval for capability and performance indices is a range of values that is used to
estimate the true process capability or performance with a certain degree of confidence. Unlike a
point estimate, which provides only a single value, a confidence interval provides a range of values
within which the true process capability or performance is expected to lie.

The confidence interval is calculated using statistical methods, and is based on the sample data
collected from the process. The level of confidence is typically expressed as a percentage, such as
95% or 99%, and indicates the probability that the true process capability or performance falls
within the confidence interval.

To calculate the confidence interval for capability index Cp, the following steps can be taken:

1. Calculate the point estimate of Cp using the formula:

Cp = (USL - LSL) / 6σ
where USL is the upper specification limit, LSL is the lower specification limit, and σ is the standard
deviation of the sample.

2. Calculate the standard error of the point estimate using the formula:

SE = σ / sqrt(n)

where n is the sample size.

3. Calculate the margin of error using the formula:

ME = z * SE

where z is the critical value of the standard normal distribution corresponding to the desired level
of confidence.

4. Calculate the confidence interval using the formula:

CI = (Cp - ME, Cp + ME)

where CI is the confidence interval.

Similarly, to calculate the confidence interval for capability index Cpk, the following steps can be
taken:

1. Calculate the point estimate of Cpk using the formula:

Cpk = min[(USL - Xbar) / 3s, (Xbar - LSL) / 3s]

where Xbar is the sample mean, s is the sample standard deviation, USL is the upper specification
limit, and LSL is the lower specification limit.

2. Calculate the standard error of the point estimate using the formula:

SE = sqrt[(1 - Cpk^2) / n]

where n is the sample size.

3. Calculate the margin of error using the formula:

ME = z * SE

where z is the critical value of the standard normal distribution corresponding to the desired level
of confidence.

4. Calculate the confidence interval using the formula:

CI = (Cpk - ME, Cpk + ME)

where CI is the confidence interval.

It is important to note that the width of the confidence interval depends on the sample size and
the level of confidence. The larger the sample size, the narrower the confidence interval, and the
higher the level of confidence, the wider the confidence interval.

In summary, a confidence interval for capability and performance indices is a range of values within
which the true process capability or performance is expected to lie with a certain degree of
confidence. It is calculated using statistical methods based on the sample data collected from the
process, and is expressed as a range of values with an associated level of confidence.