Statistical Quality Control

SQC
TQM vs. SQC

• TQM: focuses on customer-driven quality standards, managerial

leadership, continuous improvement, quality built into product
and process design, quality identified problems at the source,
and quality made everyone’s responsibility.
• Talking about solving quality problems is not enough!
• SQC: Tools come from statistics used to help identify quality
problems in production process as well as in product itself
• Descriptive statistics
• Statistical process control (SPC)
• Acceptance sampling
SQC tools
• Descriptive statistics: used to describe quality characteristics and relationships.
• Included are statistics such as mean, standard deviation, range, and a measure of
distribution of data
• Statistical process control (SPC): involves inspecting a random sample of output
from a process and deciding whether process is producing products with
characteristics that fall within a predetermined range.
• SPC answers question of whether process is functioning properly or not.
• Acceptance sampling: process of randomly inspecting a sample of goods and
deciding whether to accept the entire lot based on the results.
• Acceptance sampling determines whether a batch of goods should be accepted
or rejected.
Common causes of variation

• Common, or random, causes of variation:

• No two products are exactly alike because of slight differences in materials, workers,
machines, tools, and other factors (soft drink bottles not filled to same level)
• QC task: find out range of natural random variation in a process.
• Example: if average bottle of a soft drink contains 330 ml of liquid, we may determine
that amount of natural variation is between 325 and 335 ml. If this were case, we would
monitor the production process to make sure that the amount stays within this range.
• If production goes out of this range — bottles are found to contain on average 310 ml
— this would lead us to believe that there is a problem with the process because the
variation is greater than the natural random variation
Assignable causes of variation

• Causes can be precisely identified and eliminated.

• Poor quality in raw materials, an employee who needs more
training, or a machine in need of repair.
• In each of these examples the problem can be identified and
corrected.
• If problem is allowed to persist, it will continue to create a problem
in the quality of the product.
• In example of the soft drink bottling operation: bottles filled with
310 ml of liquid would signal a problem. The machine may need
to be readjusted.
Descriptive Statistics

• Most important descriptive statistics are measures of central tendency

such as:
• Mean
• Measures of variability: standard deviation and range
• Measures of the distribution of data
The Mean
• The arithmetic average, or the mean: a
statistic that measures the central
tendency of a set of data.
• Knowing the central point of a set of data
is highly important
• The equation for computing the mean is
Range and Standard Deviation
• There are two measures that can be used
to determine the amount of variation in
the data:
• Range: difference between largest and
smallest observations
• Standard deviation: measures amount of
data dispersion around the mean
 The standard deviation
• Small values of range and standard deviation mean that
observations are closely clustered around the mean.
• Large values of the range and standard deviation mean that the
observations are spread out around the mean

Same mean!
Distribution of Data

• Shape of distribution of the observed

data.
• When a distribution is symmetric,
same number of observations below
and above the mean. (normal
variation) is present in data.
• When a disproportionate number of
observations are either above or
below the mean, data has a skewed
distribution
Statistical Process Control

• SPC: extend the use of descriptive statistics to monitor the quality of

the product and process
• Helps determine amount of variation that is common or normal.
• Then we monitor the production process to make sure production
stays within this normal range.
• Ascertain the process is in a state of control.
• The most commonly used tool for monitoring the production
process is a control chart
Developing Control Charts

• Control chart: graph that shows whether a sample of data falls within
common or normal range of variation.
• Has upper and lower control limits that separate common from
assignable causes of variation.
• Common range of variation is defined by use of control chart limits.
• A process is out of control when a plot of data reveals that one or
more samples fall outside the control limits
• also called process chart or quality control chart
 • x axis represents samples (#1, #2, #3, etc.) taken from process
over time.
Example: volume in • y axis represents quality characteristic being monitored (ounces
coke can of liquid).
• Center line (CL) of control chart is the mean, or average, of
quality characteristic measured
Control Limits

• Upper control limit (UCL): maximum acceptable variation from the

mean for a process that is in a state of control.
• Lower control limit (LCL): minimum acceptable variation from the
mean for a process that is in a state of control
• UCL and LCL on a control chart are usually set at ±3 standard
deviations from the mean (±3σ)
 Setting Control Limits

• Assume data exhibit a normal

distribution, At ±3 standard
deviations from the mean these
control limits will capture 99.74
percent of the normal variation.
• Control limits set at ±2 standard
deviations from mean. In that case,
control limits would capture 95.44 %
of values
 Control limits and variation

• Observations that fall outside set range represent assignable causes of variation.
• There is still a small probability that a value that falls outside the limits is still due
to normal variation (Type I error, with the error being the chance of concluding that
there are assignable causes of variation when only normal variation exists). Another
name for this is alpha risk (α), where alpha refers to the sum of the probabilities in
both tails of the distribution that falls outside the confidence limits (see next figure)
• The chance of this happening is given by the percentage or probability
represented by the shaded areas of Figure. For limits of ±3 standard deviations
from the mean, the probability of a Type I error is .26% (100% - 99.74%), whereas
for limits of ±2 standard deviations it is 4.56% (100% - 95.44%)
Types of Control Charts

• Different characteristics that can be measured by control charts can be

divided into two groups: variables and attributes.
• Control chart for variables: used to monitor characteristics that can be
measured and have a continuum of values, such as height, weight, or
volume
• Control chart for attributes: used to monitor characteristics that have
discrete values and can be counted. Often they can be evaluated with a
simple yes or no decision (taste, color...etc.)
Control Charts for Variables

• Used to monitor characteristics that can be measured and have a

continuous scale, such as height, weight, volume, or width.
• When an item is inspected, the variable being monitored is measured and
recorded
• Most commonly used control charts for variables monitor both central
tendency of data (mean) and variability of data (either standard deviation
or the range)
• X-bar chart
• Range (R) chart
 Mean (x-bar) Charts
• A mean control chart is often referred to as an x-bar chart.
• Used to monitor changes in the mean of a process.
• To construct a mean chart: construct center line of the chart.
• Take multiple samples and compute their means. Usually these samples
are small, with about four or five observations. Each sample has its own
mean,
Chart construction
Example

• A quality control inspector at the Cocoa Fizz soft drink company has taken
twenty-five samples with four observations each of the volume of bottles
filled. The data and the computed means are shown in the table. If the
standard deviation of the bottling operation is 0.14 ounces, use this
information to develop control limits of three standard deviations for the
bottling operation
Solution
Resulting
chart
 Constructing chart: method 2
• Use sample range as an estimate of variability of process.
• Range: difference between largest and smallest values in sample.
• Spread of range captures variability of the data.
• In this case control limits would be constructed as follows:

The Table
Factors for three-sigma control
limits of x and R-charts
Example II

• A quality control inspector at Cocoa Fizz is using the data from previous
example to develop control limits. If the average range (R) for the twenty-
five samples is .29 ounces (computed as 7.17) and the average mean (x) of
the observations is 15.95 ounces, develop three-sigma control limits for the
bottling operation.
Solution
 Range (R) Charts
• x-bar charts measure shift in central tendency of process,
• Range charts monitor dispersion or variability of process.
• Method for developing and using R-charts is same as that
for x-bar charts.
• Center line of control chart is the average range, and the upper
and lower control limits are computed as follows: