Page 162 - DMGT501_OPERATIONS_MANAGEMENT

Operations Management

Notes the degree of conformance to requirements diameter, weight, temperature, chemical composition,
hardness etc. The pros and cons of the two types of data are summarised in the Table 6.1:

Table 6.1: Comparison of Attribute and Variable Data

Characteristic Attribute Variable
Cost of measuring instrument Nil or low High
Grade of operator Unskilled/Semiskilled Skilled
Speed Quick Slow
Recording of data Simple Complex
Overall cost per observation Low High
Information value per observation Low High
Number of observations needed for valid inference Large Small

Variable data will naturally be preferred for control purpose where the characteristic concerned
is important.

6.3 Major Statistical Measures for “Central Tendency”

When data are examined, it will normally be found that a few values will be extremely high or
extremely low and most of the values tend to be concentrated within a region which is somewhere
between the two extremes. This phenomenon is known as central tendency. The measure of
central tendency is thus a parameter in a series of statistical data which reflects a central value of
the same series.
The following are the measures most commonly used to describe the central location about
which a number of observations are grouped:
1. Mean (the ordinary “average”) is usually used for symmetrical or near symmetrical
distributions, or for distributions which lack a clearly dominant single peak.
2. Mode (value which occurs most often in data) is usually for severely skewed distributions,
describing an irregular situation where two peaks are found, or for eliminating the effects
of extreme values.
3. Median (the middle value when the figures are arranged according to magnitude) is
usually used for distributions where the mode is not well defined, for reducing the efforts
of extreme values, or for data which can be ranked but are not economically measurable
shades of colour, visual appearance, odours.

Mean is the most generally used measure of central tendency in quality work. It is employed so
often to report average size, average yield, average per cent of defective etc. Control charts have
been devised to analyse and keep track of it. Such control charts can give the earliest obtainable
warning of significant changes in central values of the group.
The mode is the value which corresponds to the greatest frequency, the peak value. It is the
number that appears most often or most commonly and is in this sense most typical of the data.
Understandably, then, the mode is the measure instinctively picked out when bar charts are
used. For example, to compare sizes of inspected parts with blue print limits. It is the size of the
parts described by the tallest bar.

In contrast, the median is generally reserved for a few special situations such as destructive
testing, where it can sometimes be used, through a statistical trick, to reduce the number of parts
tested. If, for example, the average of five parts tested is used to decide whether a life test has

156 LOVELY PROFESSIONAL UNIVERSITY