Page 215 - Intro to Space Sciences Spacecraft Applications

202
Intmduction to Space Sciences and Spacecraft Applications
considerations are not to be isolated from one another, and each may con-
tain a number of factors within itself. Each factor must be identified and
the relative importance of each specified in the evaluation function by use
of weighting factors. An example may be given for the construction of a
road for which we would like to minimize the time between destinations
and maximize the efficiency (traffic control), but, of course, minimize the
cost. The parameter may look like:

ROAD(tirne x w1, efficiency x w2, cost x w3) (9- 1 )
where time, efficiency, and cost represent the important variables of the
parameter and wlr w2, and w3 represent the weighting factors assigned to
each variable. These variables must all be considered, but some may be
more important than others and will be given higher weighting factors in
the actual parameter statement indicating their priority. Some of these vari-
ables may also conflict with the desires of some of the others. For instance,
minimizing the travel time on our road by cutting a hole through a moun-
tain may not fit into the cost restraints with which we also must ded.
A considerable amount of effort must be put toward this stage of the
design process since it directly affects the outcome of evaluation of pos-
sible solutions to the problem.

Alternative Solutions. Having decided on the desired performance dur-
ing the last stage, the designer may now use his knowledge and analytical
skills to synthesize solutions to the problem. Formulation of solutions
prior to the last stage should be avoided to prevent biased ideas. This is
the fun part of systems engineering where the designer may dream up
exotic solutions to the stated objectives and have them compared to the
more conventional solutions that will also be brought forward. (Who
knows if the exotic solution won’t out-perform the conventional solutions
when the evaluation stage is reached!)

Test or Analysis. Before a good evaluation can be performed, as much
information as possible on each solution must be obtained. In the test or
analysis stage, each solution is tested (using models or simulations), if
possible. Alternately, its performance in each of the value model factors
may be estimated. This will result in a performance that can be compared
to the desired performance model formulated earlier.

Evaluation. The results of the test or analysis stage for each solution are
compared to the desired performance function as well as to the results for