Measurement and Metrics - Requirements Engineering and Management for Software Development Projects - page 144

10.3.4 Relative Effort Metrics

Relative effort metrics help us in assessing the reasonableness of the importance
given to activities. Absolute metrics do not tell us about the importance given to an
activity. These metrics inform us of the relative importance being accorded to an
activity in comparison with other activities. We compute the following metrics.

10.3.4.1 Importance to Requirements Engineering in the Project

This metric is computed using the formula:

RRE ¼ ERE TE ð Þ 100

Where
RRE Relative importance to Requirements Engineering in the project

ERE Effort in person hours spent on all activities of Requirements Engineering

TE Total effort in Person hours spent on all activities of the project

This metric is expressed as a percentage. We compare this percentage to the
organizational benchmark and take appropriate corrective and preventive action as
well as improve the process to control the variance in it. The data for this can be
easily obtained from the organizational timesheets for the project.
While there is no established benchmark, 20 % is used in full life cycle (from
requirements to acceptance testing and delivery) development projects to gauge
the reasonableness of the amount of effort spent on requirements engineering.
When a project's RRE exceeds the organizational benchmark, we carry out an
analysis to ascertain what pitfalls caused this slide and draw lessons for future
projects. If this metric falls below the organizational benchmark, we analyze the
project to ascertain what best practices allowed the savings to improve our process.
But if we spent less time than the organizational benchmark, it could result in:

1. Higher DIR whose origin is in requirements engineering
2. Receiving more CRs

We can see a correlation between the effort spent on requirements engineering and
the quality or stability of requirements. If the above two outcomes are absent, it
establishes that we implemented some the best practices. So, this metric helps us in
analyzing the requirements engineering activity and its impact on quality or CRs.

10.3.4.2 Quality Control of the URS and SRS

The main activities for ensuring that quality is in-built in the URS and SRS are
verification consisting of peer review and managerial review and validation. The
formula for computing this metric is::