Procedure of control chart
.
Define the Metric or Quality Characteristic
• Identify the process variable or quality characteristic that
needs to be controlled. This could be anything from the
diameter of a manufactured part, the time to service a
customer, or the temperature of a chemical process.
Collect Data
• Gather data from the process. This typically involves taking
measurements from a sample of output from the process. The
sample size and the frequency of sampling depend on the process
dynamics and the practical aspects of data collection.
• Calculate Process Statistics
• Compute the mean (average) and standard deviation (or range) of
the collected data. These statistics are used to set the control limits
of the chart.
Determine Control Limits
• Control limits are set using the process statistics. Typically, these
limits are set at ±3 standard deviations from the process mean.
These limits define the boundaries for process variability that is
considered normal (within control).
• Upper Control Limit (UCL) = Mean + 3*(Standard Deviation)
• Lower Control Limit (LCL) = Mean - 3*(Standard Deviation)
• For a range control chart, the limits would be based on the
average range of the samples.
Plot the Control Chart
Draw the Chart: On graph paper or using statistical software:
1. X-axis (Horizontal): Time or sequence of samples.
2. Y-axis (Vertical): Measurement values.
3. Central Line (CL): Represents the process mean.
4. Upper and Lower Control Limits (UCL and LCL): Lines that represent the
thresholds beyond which the process is considered out of control.
Plot Data Points: Plot the individual measurements or subgroup
averages over time.
Analyze the Control Chart
Look for Patterns: Check for any signs of special-cause variation:
1. Points outside of Control Limits: Indicates out-of-control process.
2. Non-random Patterns: Such as trends, cycles, or unusual patterns.
Determine Process Stability: The process is considered stable if most
points are within control limits and no non-random patterns are
detected.
Investigate and Act
Identify Causes of Variation: If the chart shows out-of-control signals or
patterns, investigate potential causes.
Implement Corrective Actions: Modify the process to eliminate special-cause
variation and to improve process stability.
Continue Monitoring: After changes are made, continue to use the control
chart to ensure the process remains stable.
• Adjust and Recalculate as Necessary
Review and Recalculate Limits: As process improvements are made and data
collected over time, revisit the calculation of the mean and control limits to
ensure they reflect the current state of the process.
Attribute Causes
• Definition: Attribute data refers to data that can be counted for recording and analysis
of defects (qualitative data). It is data that fits into discrete categories and is binary or
nominal in nature.
• Examples: Examples of attribute data include the number of defective items in a batch,
presence or absence of a feature, and pass/fail outcomes. Each measurement simply
represents whether the item or outcome meets the criteria or does not meet them.
• Control Charts: The control charts used for attribute data include:
• p-chart (proportion chart) used for monitoring the proportion of defective items
in a sample.
• np-chart used for monitoring the number of defective items when sample sizes
are constant.
• c-chart used for monitoring the count of defects per unit when the opportunity for
defects is constant.
• u-chart used for monitoring the number of defects per unit when the number of
units or sample size can vary.
•
Variable causes
• Definition: Variable data refers to data that can be measured on a continuous scale
(quantitative data). It reflects varying degrees of characteristic or dimension in a
measurement.
• Examples: Examples include the length of a part, the temperature of a batch
process, the pressure exerted by a machine, or the time taken to complete a
process. These characteristics are measured on a continuous scale and give more
precise information about the process.
• Control Charts: The control charts used for variable data include:
• X̄ and R chart (mean and range chart) used to monitor the process mean and
variability when sample sizes are small.
• X̄ and S chart (mean and standard deviation chart) used for larger sample sizes.
• I-MR chart (individual and moving range chart) used for monitoring individual
observations and their immediate variability.
• Analysis Focus: The analysis focuses on the stability of the process mean and
