Measurement and Metrics - Requirements Engineering and Management for Software Development Projects - page 137

2. Size of artifacts using work register. The size would be different for information
artifacts and code artifacts. Size of information artifacts is often measured in
number of pages, number of requirements, number of classes and so on. Code
artifacts are generally measured in LOC (Lines of Code), Function Points,
Object Points or SSU (Software Size Units). Measuring software size forms
part of the software estimation subject and is beyond the scope of this topic.
Interested readers may refer ''Software Estimation: Best Practices, Tools and
Techniques for Software Project Estimators'' (2009) by Murali Chemuturi and
published by J.Ross Publishing, Inc, USA for more information.
3. Defects uncovered in software artifacts, both the information artifacts and code
artifacts. Defects are measured as integer numbers. They may also be classified
as critical, major and minor or some such categorization. We often associate the
origin of the defect with the measurement of defects. We obtain defect data
from defect reports or defect resolution tools and the organizational defect
definition which would define the class (critical, major or minor) and the
origins.
4. We also make note of the dates on which activities are performed. We obtained
the scheduled dates from the project schedule. We obtain the actual dates from
the work register and the timesheets.
5. We also measure the change requests placed on the project to gauge their
impact on the project and the stability requirements.
So, the primary measures are effort, size, defects, dates and changes. From these
measures we derive a host of metrics for the project out of which a subset is
relevant for the requirements engineering and management. In the following
sections, we discuss those metrics that are relevant to requirements engineering
and management.
We derive metrics in the following five classes:
1. Productivity Metrics
2. Change Request Metrics
3. Quality Metrics
4. Relative Effort Metrics
5. Schedule Metrics
Productivity metrics assist us in assessing the efficiency of handling the
function.
Change request metrics allows us to assess the stability of requirements in the
project.
Quality metrics assist us in understanding how well we performed the function
and the level of quality in our deliverables.
Relative Effort metrics assist us in understanding the importance we are giving
to the function in the overall scheme of software development.
Schedule metrics assist us in understanding how well we met the delivery
commitments to our client.
Now let us look at each of them in a little greater detail.