STATISTICAL QUALITY CONTROL(SQC)

Learning objectives

After studying this chapter you should be able to:

1. Describe categories of statistical quality control

2. Explain the use of descriptive statistics in measuring quality characteristics
3. Identify and describe causes of variations
4. Describe the use of control charts
5. Describe the differences between x- bar, R-, p- and c- charts
6. Explain the meaning of process capability and process capability index
7. Explain the tern six sigma
8. Explain the process of acceptance sampling
9. Describe the challenges in measuring quality in a service organization

Introduction

We need tools that help us identify and solve quality problems. These tools come from the area
of statistics and are used to help identify quality problems in the production process as well as in
the product itself.

Statistical quality control is the term used to describe the set of statistical tools used by quality
professionals. SQC can be divided into three categories;

1. Descriptive statistics are used to describe quality characteristics and relationships.

Included are the mean, standard deviation, range and the measure of distribution of data.
2. Statistical process control involves inspecting a random sample of the output from a
process and deciding whether the process is producing products with characteristics that
fall within a predetermined range. SPC answers the question whether the process is
functioning properly or not.
3. Acceptance sampling is the process of randomly inspecting a sample of goods and
deciding whether to accept the entire lot based on the results. It determines whether a
batch of goods should be accepted or rejected.

Sources of variations: Common and assignable causes

If you look at bottles of soda in a crate, you will notice that no two bottles are filled to exactly
the same level. Some are slightly higher and others slightly lower. This type of differences exists
in all products available it could be differences in weights or dimensions. These type of
differences are completely normal. No two products are exactly the same due to slight
differences in materials, workers, machines, tools and other factors. These causes are referred to
as common or random causes of variations. These are based on random or common causes
which are unavoidable.

The second type of variation that can be observed involves variations where the causes can be
precisely identified and eliminated. These are called assignable causes of variations. Examples
of this type of variations are poor quality in raw materials, an employee who needs more training
or a machine in need of repair. In each of these problems the cause can be identified and fixed.

Descriptive statistics

Descriptive statistics is useful in describing certain characteristics of a product and a process.

The most important descriptive statistics are measures of central tendency such as the mean,
measures of variability such as standard deviation and range and measure of distribution of data.

The mean

The arithmetic average or the mean is a statistics that measures the central tendency of a set of
data. Knowing the central point of a set of data is highly important. Mean is computed summing
all observations and dividing with the total number of observations as expressed in equation 1
n

∑ x1 (1)
x= i=1
n

Where x=the mean

x i=observationi ,i=1

n = number of observations

the range and standard deviation

There are btwo measures which can be used to determine the amount of variation in a data. The
first is the range which is the difference between the largest and smallest observations. Another
measure of variation of data is standard deviations which is mathematically expressed in
equation 2

√
n
σ= ∑ ¿¿¿¿ (2)
i=1

Where σ =standard deviation of a sample

x=the mean

ni=observation i=1… .n
 n=the number of observations∈the sample

Small vales of range and standard deviations mean that the observations are closely clustered
around the mean. Their large values means the observations are spread out around the mean.

Distribution of data

A third descriptive statistic used to measure quality characteristic is the shape of the distribution
of the observed data. When the distribution is symmetric, there are the same number of
observations below and above the mean. This is a result of normal variation.. When there is
disproportionate number of observations either above or below the mean, we say the data has a
schewed distribution. Figs 1 & 2

Figure 1 Normal distribution with varying standard deviations

 Figure 2. Differences between symmetric and asymmetric distributions

STATISTICAL PROCESS CONTROL (SPC) METHODS

SPC methods extend the use of descriptive statistics to monitor the quality of product and
process. Using SPC we want to determine the amount of variation that is common or normal. We
also monitor the production process to make sure production stays within the normal ranges. The
most common tool used here are control charts.

Developing control charts

A control chart (process charts or quality control charts) is a graph which shows whether a
sample of data falls within the normal variation. A control chart has an upper and lower control
limit that separates common from assignable causes of variations. The common range of
variation is defined by use of control chart limits. We say a process is out of control when a plot
of data reveals that one or more data fall outside the control limits.

Let us consider an example of a coca cola Fizz bottling operation in Fig. 3

The x axis is sample number taken from the process over time. Y-axis represents the quality
characteristic being monitored (ounces of liquid). The centre line (CL) of the chart is the mean,
or average of the quality characteristic 16.0 ounces. Upper control limit is the maximum
acceptable variation (UCL) and lower control limit (LCL) is the minimum acceptable variation
from the mean of a process. Variations due to normal causes fall within the control limits either
below or above the CL. If a sample of observations falls outside the control limits then we need
to look for assignable causes.

Upper and lower control units are calculated at ±3 standard deviations from the mean. If the data
exhibits a normal distribution then, the ±3 standard deviations will cover 99,74 % of data and if
set at ±2 standard deviations then 95.44 % of data will be captured.