Information System Development Life Cycle Models - High Level Models and Methodologies for Information Systems - page 36

2.11 Hybrid System Development Life Cycles

An environment of increasing competitiveness demands systems that are safe and
trustworthy. They also have to be adaptable and flexible to the changes that can

happen at any moment, in a fast paced, ever-evolving world. This brings compli-
cations for the development process. A common response to these issues is the
combination of different system development life cycles.
SDLC models each have their own peculiar characteristics, which can be both
advantageous and detrimental, depending on the type of project requirements and
features. Once a project is hypothesized, a model is chosen to

fit its purposes. But if

the particular characteristics of the project do not necessarily c model,

it is possible to combine guidelines from more than one. This combination is
primarily done to harness the qualities of a model and reduce its weaknesses by
incorporating the strengths of another model (Rahmany 2012 ).
An example of combined life cycle models is the case of a development process
that is being guided by the spiral model, but that later in the process demands a
change in the requirements. To accommodate this need, the agile model could be
incorporated (Rahmany 2012 ).
Madachy et al. ( 2006 ) have outlined a hybrid SDLC model which they named
the scalable spiral model. The main purpose was to combine a plan-driven approach
with an agile one. Development is organized into thoroughly planned increments,
which take into consideration relatively stable, initial requirements. However, upon
the release of each increment, it is crucial to have an agile team focusing on market,
competition, and technological analysis, as well as user feedback and renegotiation
of the characteristics of the next increments. The model ultimately aims at simul-
taneously catering for the challenges of rapid change and the need for risk man-
agement and dependability (Madachy et al. 2006 ).
Munassar and Govardhan ( 2010 ) have also attempted a similar approach, which
they simply named the hybrid model. They picked up essential traits from such
different models as waterfall, iteration, V-shaped, spiral, and extreme programming
(XP). It consists of a series of seven steps that are interconnected with each other:
planning, requirements (at which point risk analysis is undertaken), design,
implementation (including testing), integration development, deployment, and
maintenance. Although the process appears to be outlined in the same style of a
waterfall approach, the relationship between the different stages is

fit one speci

fluid and mul-
tidirectional, accounting for possible changes in requirements and the need to revise
design features after testing. The authors argue that this approach would permit to
combine the best characteristics of each model: It promotes good habits of de

ne-
before-design and design-before-code (like the waterfall model), while, at the same
time, it avoids the dangers of rigid development by introducing early testing. It is
also iterative, but still incorporates risk analysis, and the test-based approach allows
for high usability (Munassar and Govardhan 2010 ).