Measuring the Use of Features in a Requirements Engineering Tool - An Industrial Case Study

Keywords Measuring the Use of Features in a Requirements Engineering Tool - An Industrial Case Study

The context of the case study is a company within a large
group of companies, or more precisely the production part
of the company in scope. This organisation employs
around 600 persons and competence within several
subject fields (mechanics, electronics, and software
development) is needed. In many cases the organisation
works closely with partners to customise the products for
different applications.

The organisation has used Focal Point for more than
three years. The activities performed within the system are
mainly of two different types. Firstly, strategic product
planning that takes place during a short period of time
once a year, which is a task that involves all product
managers and aims at establishing an overall budget and a
roadmap for the enterprise. Project proposals in Focal
Point are prioritised and visualised, and budget negotiations
are included in the roadmap. It is important to note,
though, that this activity did not take place during the case
study’s time of measurement. Second, Focal Point is used
on regular basis to implement the roadmap, i.e. investigate
and decide which requirements that can be accepted
within the budget. Product managers use the system to
make inquiries to development managers.

Date published
Document type
technical white paper
Defines standard
Replaced/Superseded by document(s)
Cancelled by
Amended by
File MIME type Size (KB) Language Download
Paper Feature Use in RE Tool borg_serps04.pdf application/pdf   150.61 KB English DOWNLOAD!
File attachments

Measuring how features are actually used in a system has
several potential benefits, for instance an improved
requirements selection process. We have investigated how
the use of features in Focal Point—a highly interactive,
web-based RE tool—can be measured using log-file
analysis, and we have interviewed key persons (user,
developer, and consultant) in order to validate and
discuss the deployed method as well as the results.
Our measurements show that different view and edit
features are dominating. There is a 90% agreement
between measured use and other sources such as interviews
with stakeholders and an automated solution based
on direct mapping from accesses of certain .jsp-files
(JavaServer Pages) to feature categories. All stakeholders
believe that the information generated is useful, but the
development manager would like more detailed information
based on logging individual mouse-clicks.

Andreas Borg, Joachim Karlsson, Stefan Olsson, Kristian Sandahl
Visit also