[STATISTICAL QUALITY CONTROL]

QUALITY CONTROL

Quality control in construction typically involves insuring compliance with minimum

standards of material and workmanship in order to insure the performance of the facility
according to the design. These minimum standards are contained in the specifications
described in the previous section. For the purpose of insuring compliance, random samples
and statistical methods are commonly used as the basis for accepting or rejecting work
completed and batches of materials. Rejection of a batch is based on non-conformance or
violation of the relevant design specifications.

An implicit assumption in these traditional quality control practices is the notion of an

acceptable quality level which is a allowable fraction of defective items. Materials obtained
from suppliers or work performed by an organization is inspected and passed as acceptable if
the estimated defective percentage is within the acceptable quality level. Problems with
materials or goods are corrected after delivery of the product.

TOTAL QUALITY CONTROL

In contrast to this traditional approach of quality control is the goal of total quality control. In
this system, no defective items are allowed anywhere in the construction process. While the
zero defects goal can never be permanently obtained, it provides a goal so that an
organization is never satisfied with its quality control program even if defects are reduced by
substantial amounts year after year. This concept and approach to quality control was first
developed in manufacturing firms in Japan and Europe, but has since spread to many
construction companies. The best known formal certification for quality improvement is the
International Organization for Standardization's ISO 9000 standard. ISO 9000 emphasizes
good documentation, quality goals and a series of cycles of planning, implementation and
review.

Total quality control is a commitment to quality expressed in all parts of an organization and
typically involves many elements. Design reviews to insure safe and effective construction
procedures are a major element. Other elements include extensive training for personnel,
shifting the responsibility for detecting defects from quality control inspectors to workers,
and continually maintaining equipment. Worker involvement in improved quality control is
often formalized in quality circles in which groups of workers meet regularly to make
suggestions for quality improvement. Material suppliers are also required to insure zero

Quality control | Anil, Rameez –BEM ,SPA, New Delhi 1

 [STATISTICAL QUALITY CONTROL]

defects in delivered goods. Initially, all materials from a supplier are inspected and batches of
goods with any defective items are returned. Suppliers with good records can be certified and
not subject to complete inspection subsequently.

The traditional microeconomic view of quality control is that there is an "optimum"

proportion of defective items. Trying to achieve greater quality than this optimum would
substantially increase costs of inspection and reduce worker productivity. However, many
companies have found that commitment to total quality control has substantial economic
benefits that had been unappreciated in traditional approaches. Expenses associated with
inventory, rework, scrap and warranties were reduced. Worker enthusiasm and commitment
improved. Customers often appreciated higher quality work and would pay a premium for
good quality. As a result, improved quality control became a competitive advantage.

Quality control is difficult to apply, particular in construction. The unique nature of each
facility, the variability in the workforce, the multitude of subcontractors and the cost of
making necessary investments in education and procedures make programs of total quality
control in construction difficult. Nevertheless, a commitment to improved quality even
without endorsing the goal of zero defects can pay real dividends to organizations.

Quality Control by Statistical Methods

An ideal quality control program might test all materials and work on a particular facility. For
example, non-destructive techniques such as x-ray inspection of welds can be used
throughout a facility. An on-site inspector can witness the appropriateness and adequacy of
construction methods at all times. Even better, individual craftsmen can perform continuing
inspection of materials and their own work. Exhaustive or 100% testing of all materials and
work by inspectors can be exceedingly expensive, however. In many instances, testing
requires the destruction of a material sample, so exhaustive testing is not even possible. As a
result, small samples are used to establish the basis of accepting or rejecting a particular work
item or shipment of materials. Statistical methods are used to interpret the results of test on a
small sample to reach a conclusion concerning the acceptability of an entire lot or batch of
materials or work products.

The use of statistics is essential in interpreting the results of testing on a small sample.
Without adequate interpretation, small sample testing results can be quite misleading. As an
example, suppose that there are ten defective pieces of material in a lot of one hundred. In

Quality control | Anil, Rameez –BEM ,SPA, New Delhi 2

 [STATISTICAL QUALITY CONTROL]

taking a sample of five pieces, the inspector might not find any defective pieces or might
have all sample pieces defective. Drawing a direct inference that none or all pieces in the
population are defective on the basis of these samples would be incorrect. Due to this random
nature of the sample selection process, testing results can vary substantially. It is only with
statistical methods that issues such as the chance of different levels of defective items in the
full lot can be fully analyzed from a small sample test.

There are two types of statistical sampling which are commonly used for the purpose of
quality control in batches of work or materials:

The acceptance or rejection of a lot is based on the number of defective (bad) or

nondefective (good) items in the sample. This is referred to as sampling by attributes.
A quantitative quality measure or the value of a measured variable is used as a quality
indicator. This testing procedure is referred to as sampling by variable

Whatever sampling plan is used in testing, it is always assumed that the samples are
representative of the entire population under consideration. Samples are expected to be
chosen randomly so that each member of the population is equally likely to be chosen.
Convenient sampling plans such as sampling every twentieth piece, choosing a sample every
two hours, or picking the top piece on a delivery truck may be adequate to insure a random
sample if pieces are randomly mixed in a stack or in use. However, some convenient
sampling plans can be inappropriate. For example, checking only easily accessible joints in a
building component is inappropriate since joints that are hard to reach may be more likely to
have erection or fabrication problems.

Another assumption implicit in statistical quality control procedures are that the quality of
materials or work is expected to vary from one piece to another. This is certainly true in the
field of construction. While a designer may assume that all concrete is exactly the same in a
building, the variations in material properties, manufacturing, handling, pouring, and
temperature during setting insure that concrete is actually heterogeneous in quality. Reducing
such variations to a minimum is one aspect of quality construction. Insuring that the materials
actually placed achieve some minimum quality level with respect to average properties or
fraction of defectives is the task of quality control.

Quality control | Anil, Rameez –BEM ,SPA, New Delhi 3

 [STATISTICAL QUALITY CONTROL]

Statistica1 quality control (SQC) is the term used to describe the set of statistical tools used
by quality professionals. Statistical quality control can be divided into three broad categories:

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

Included are statistics such as the mean, standard deviation, the range, and a measure of the
distribution of data.

2. Statistical process control (SPC) 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 of 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. Acceptance sampling
determines whether a batch of goods should be accepted or rejected.

All three of these statistical quality control categories are helpful in measuring and evaluating
the quality of products or services. However, statistical process control (SPC) tools are used
most frequently because they identify quality problems during the production process. The
quality control tools do not only measure the value of a quality characteristic but they also
help us identify a change or variation in some quality characteristic of the product or process.
This will help to identify specific tools used for measuring this variation.

SOURCES OF VARIATION: COMMON AND ASSIGNABLE CAUSES

COMMON CAUSES: No two products are exactly alike because of slight differences in
materials, workers, machines, tools, and other factors. These are called common, or random,
causes of variation. Common causes of variation are based on random causes that we cannot
identify. These types of variation are unavoidable and are due to slight differences in
processing. For example: if you look at blueberry muffins in a bakery, you will notice that
some are slightly larger than others and some have more blueberries than others, bottles of a
soft drink in a grocery store, you will notice that no two bottles are filled to exactly the same
level. These types of differences are completely normal.

ASSIGNABLE CAUSES: 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 variation. Examples of this type of variation are poor quality in raw

Quality control | Anil, Rameez –BEM ,SPA, New Delhi 4

 [STATISTICAL QUALITY CONTROL]

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. Also, if the problem is allowed
to persist, it will continue to create a problem in the quality of the product. For example, if
the average bottle of a soft drink called Cocoa Fizz contains 16 ounces of liquid, we may
determine that the amount of natural variation is between 15.8 and 16.2 ounces. But if it
shows an average of 15.6 ounces of liquid would signal a problem. The machine may need to
be readjusted. This would be an assignable cause of variation. We can assign the variation to
a particular cause (machine needs to be readjusted) and we can correct the problem (readjust
the machine).

DESCRIPTIVE STATISTICS

It can be helpful 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 the standard deviation and range, and measures of the distribution of
data.

The Mean: The arithmetic average, or the mean, is a statistic that measures the central
tendency of a set of data. Knowing the central point of a set of data is highly important. To
compute the mean we simply sum all the observations and divide by the total number of
observations. The equation for computing the mean is

where _ the mean

xi _ observation i, i _ 1, . . . , n
n _ number of observations

The Range and Standard Deviation: It tells us how spread out the data is around the mean.
There are two measures that can be used to determine the amount of variation in the data. The
first measure is the range, which is the difference between the largest and smallest
observations. Another measure of variation is the standard deviation. The equation for
computing the standard deviation is

where σ standard deviation of a sample

x the mean
xi _ observation i, i _ 1, . . . , n
n _ the number of observations in the sample

Quality control | Anil, Rameez –BEM ,SPA, New Delhi 5

 [STATISTICAL QUALITY CONTROL]

Normal distributions with varying standard deviations Differences between symmetric and skewed distributions

STATISTICAL PROCESS CONTROL METHODS

Statistical process control methods extend the use of descriptive statistics to monitor the
quality of the product and process. Using statistical process control we want to determine the
amount of variation that is common or normal. Then we monitor the production process to
make sure production stays within this normal range. That is, we want to make sure the
process is in a state of control. The most commonly used tool for monitoring the production
process is a control chart. Different types of control charts are used to monitor different
aspects of the production process.

Developing Control Charts

A control chart (also called process chart or quality control chart) is a graph that shows
whether a sample of data falls within the common or normal range of variation. A control
chart has upper and lower control limits that separate common from assignable causes of
variation. The common range of variation is defined by the use of control chart limits. A
process is said to be out of control when a plot of data reveals that one or more samples fall
outside the control limits.

Types of Control Charts

Control charts are one of the most commonly used tools in statistical process control. They
can be used to measure any characteristic of a product, such as the weight of a cereal box, the
number of chocolates in a box, or the volume of bottled water. The different characteristics
that can be measured by control charts can be divided into two groups: variables and
attributes. A control chart for variables is used to monitor characteristics that can be measured
and have a continuum of values, such as height, weight, or volume.

Quality control | Anil, Rameez –BEM ,SPA, New Delhi 6

 [STATISTICAL QUALITY CONTROL]

1) % of values captured by different ranges of SD 2) Chance of Type I error for ±3σ

Mean (x-Bar) Charts

A control chart used to monitor changes in the mean value of a process. A mean control chart
is often referred to as an x-bar chart. It is used to monitor changes in the mean of a process.
To construct a mean chart we first need to construct the centre line of the chart. To do this we
take multiple samples and compute their means. Usually these samples are small, with about
four or five observations. Each sample has its own mean, the center line of the chart is then
computed as the mean of all ‘k’ sample means, where ‘k’ is the number of samples:

To construct the upper and lower control limits of the chart, we use the following formulas:

Quality control | Anil, Rameez –BEM ,SPA, New Delhi 7

[STATISTICAL QUALITY CONTROL]

Quality control | Anil, Rameez –BEM ,SPA, New Delhi 8

[STATISTICAL QUALITY CONTROL]

Quality control | Anil, Rameez –BEM ,SPA, New Delhi 9

 [STATISTICAL QUALITY CONTROL]

Constructing a Mean (x-Bar) Chart from the Sample Range

Quality control | Anil, Rameez –BEM ,SPA, New Delhi 10

 [STATISTICAL QUALITY CONTROL]

Factors for three-sigma control limits of and R-charts-Table-1

Source: Factors adapted from the ASTM Manual on Quality Control of Materials.

Range (R) Charts

Range (R) charts are another type of control chart for variables. Whereas x-bar charts
measure shift in the central tendency of the process, range charts monitor the dispersion or
variability of the process. The method for developing and using R-charts is the same as that
for x-bar charts. The center line of the control chart is the average range, and the upper and
lower control limits are computed as follows:

where values for D4 and D3 are obtained from Table-1

Quality control | Anil, Rameez –BEM ,SPA, New Delhi 11

 [STATISTICAL QUALITY CONTROL]

CONTROL CHARTS FOR ATTRIBUTES

Two of the most common types of control charts for attributes are p-charts and c-charts.

P-charts are used to measure the proportion of items in a sample that are defective.
Examples are the proportion of broken cookies in a batch and the proportion of cars produced
with a misaligned fender. P charts are appropriate when both the number of defectives
measured and the size of the total sample can be counted. A proportion can then be computed
and used as the statistic of measurement.

Quality control | Anil, Rameez –BEM ,SPA, New Delhi 12

 [STATISTICAL QUALITY CONTROL]

C-charts count the actual number of defects. For example, we can count the number of
complaints from customers in a month, the number of bacteria on a petri dish, or the number
of barnacles on the bottom of a boat. However, we cannot compute the proportion of
complaints from customers, the proportion of bacteria on a petri dish, or the proportion of
barnacles on the bottom of a boat.

1) To construct the upper and lower control limits for a p-chart, we use the following
formulas:

2) To construct the upper and lower control limits for a c-chart, we use the following
formulas:

An example to illustrate the p-chart

Quality control | Anil, Rameez –BEM ,SPA, New Delhi 13

 [STATISTICAL QUALITY CONTROL]

ACCEPTANCE SAMPLING

Acceptance sampling, the third branch of statistical quality control, refers to the process of
randomly inspecting a certain number of items from a lot or batch in order to decide whether
to accept or reject the entire batch.

A sampling plan is a plan for acceptance sampling that precisely specifies the parameters of
the sampling process and the acceptance/rejection criteria. The variables to be specified
include the size of the lot (N), the size of the sample inspected from the lot (n), the number of
defects above which a lot is rejected (c), and the number of samples that will be taken.

single sampling: in which a random sample is drawn from every lot

double sampling: This provides an opportunity to sample the lot a second time if the
results of the first sample are inconclusive.
multiple sampling: similar to double sampling plans except that criteria are set for
more than two samples. The decision as to which sampling plan to select has a great
deal to do with the cost involved in sampling, the time consumed by sampling, and
the cost of passing on a defective item.

Operating characteristic (OC) curve

A graph that shows the probability or chance of accepting a lot given various proportions
of defects in the lot.

Quality control | Anil, Rameez –BEM ,SPA, New Delhi 14