Page 117 - Software and Systems Requirements Engineering in Practice

86 S o f t w a r e & S y s t e m s R e q u i r e m e n t s E n g i n e e r i n g : I n P r a c t i c e

and make modifications; e.g., changing, adding, or deleting artifacts
and their relationships will automatically be reflected everywhere the
related objects are shown.
Navigation of Complex System Requirement Sets
Navigating through volumes of text can be very difficult. Even when
items are traced to each other in databases, as projects get larger, trace
matrices increase in size as the square of the number of requirements
and it can become a daunting task to find information for an impact
or coverage analysis. Moreover, for someone not familiar with the
domain, navigation can be a time-consuming, trial-and-error process.
Navigation with a well-crafted model is no more difficult than finding
a route with a map. Touch, zoom, touch, zoom, etc., and you are there.
Finding related objects is as easy as doing an ad hoc query (remember,
everything is in a database). As models scale well, ease of navigation
remains the same regardless of model size, although it might take a
few more mouse clicks to find the material of interest.

Rapid Review of Business Processes
and Requirements Relationships
Reviewing diagrams is significantly faster and more thorough than
reviewing similar material in text. We have found that model reviews
tend to be more culture and language neutral than reviewing text
documents. Furthermore, if the models are extended to support the
Unified Requirements Modeling Language (URML) concepts (see
later Section 4.5), then the relationships between hazards, threats,
processes, product features, business goals, and functional and
nonfunctional requirements can all be viewed at the same time by
distributed teams [Berenbach and Gall 2006]. Pictures tend to be less
ambiguous than text, and relationships (or the lack thereof) are
immediately apparent.

Metrics for Quality and Progress
Unlike text, models are mathematical structures (directed graphs). It
is therefore possible to define metrics for quality and progress, and
then semiautomatically extract the metrics at periodic intervals
[Berenbach and Borotto 2006]. The rapid extraction and analysis of
metrics improves transparency and product quality.
Semiautomatic Generation of Project Plans
and Requirements Database Content
With a properly crafted model, where one of the goals of the modeling
effort is the automatic generation of downstream artifacts such as
project plans and the population of requirements databases, the
manual transcription of information can be avoided, along with the
potential for transcription errors [Berenbach 2003].