Page 108 -

CHAPTER

4 SOFTWARE PROCESS AND
PROJECT METRICS

KEY easurement is fundamental to any engineering discipline, and soft-
CONCEPTS
ware engineering is no exception. Measurement enables us to gain
backfiring . . . . . . . 94
Minsight by providing a mechanism for objective evaluation. Lord
defect removal Kelvin once said:
efficiency . . . . . . . 98
function points . . . 89 When you can measure what you are speaking about and express it in numbers, you
metrics collection 100 know something about it; but when you cannot measure, when you cannot express
it in numbers, your knowledge is of a meager and unsatisfactory kind: it may be the
project metrics . . . 86
beginning of knowledge, but you have scarcely, in your thoughts, advanced to the
process metrics . . 82
stage of a science.
quality metrics. . . 95
The software engineering community has finally begun to take Lord Kelvin's
size-oriented
metrics . . . . . . . . . 88 words to heart. But not without frustration and more than a little controversy!
Software metrics refers to a broad range of measurements for computer soft-
statistical process
control . . . . . . . . 100 ware. Measurement can be applied to the software process with the intent of
SSPI. . . . . . . . . . . . 84 improving it on a continuous basis. Measurement can be used throughout a
software project to assist in estimation, quality control, productivity assess-
ment, and project control. Finally, measurement can be used by software engi-
neers to help assess the quality of technical work products and to assist in
tactical decision making as a project proceeds.

QUICK What is it? Software process and Why is it important? If you don’t measure, judge-
LOOK product metrics are quantitative ment can be based only on subjective evaluation.
measures that enable software With measurement, trends (either good or bad)
people to gain insight into the efficacy of the soft- can be spotted, better estimates can be made,
ware process and the projects that are conducted and true improvement can be accomplished over
using the process as a framework. Basic quality time.
and productivity data are collected. These data What are the steps? Begin by defining a limited set
are then analyzed, compared against past aver- of process, project, and product measures that are
ages, and assessed to determine whether quality easy to collect. These measures are often nor-
and productivity improvements have occurred. malized using either size- or function-oriented met-
Metrics are also used to pinpoint problem areas rics. The result is analyzed and compared to past
so that remedies can be developed and the soft- averages for similar projects performed within the
ware process can be improved. organization. Trends are assessed and conclusions
Who does it? Software metrics are analyzed and are generated.
assessed by software managers. Measures are
often collected by software engineers.

79