Software Engineering

14.2.1 
Assemble the experimental mechanisms
The trade-study definition should provide sufficient information concerning the 
investigation to determine the tools, models, or simulations that are needed to sup-
port the trade study. These experimental mechanisms must be assembled and quali-
fied. Trade-study tools and models should provide abstract representations of each 
architectural alternative. A computer-based simulation provides dynamic modeling 
FIGURE 14.2
Radar or spider chart example.

---

253
14.2 
Establish the trade-study environment
mechanisms that emulate expected operational or behavioral responses of a design 
representation to stimuli and conditional situations. Experimental mechanisms 
should be defined in a manner that enables each architectural alternative to be eval-
uated under the same conditions to ensure the consistency of the data collected.
The experimental mechanisms should express the architectural alternatives at a 
level of engineering detail that reflects the maturity of the overall software archi-
tecture. The mechanisms must exhibit the design characteristics needed to assess 
the competing alternatives against the evaluation success criteria. Experimental 
techniques must address the scenarios that will be used to exercise the alternatives 
and generate the data required for comparative analysis. Scenarios should identify 
the preconditions and sequence of events under which the experiment will be con-
ducted. Unlike a test case, it is not necessary to establish the expected outcomes for 
a scenario. The results of the experiment are to be captured and analyzed to deter-
mine how the architectural alternatives reacted to each scenario.