Engineering vs. Scientific Problems
Engineering vs. Scientific Problems
m Scientific Problems:
Ù Usually characterized by a small number of well-defined, relatively independent, first-order effects.
Ù Input and required output usually well defined.
Ù Measure of success usually well defined.
Ù Usually well-suited to algorithmic solution.
e.g. channel routing for minimum no. of tracks
m Engineering Problems:
Ù Characterized by a large number of interacting second-order effects.
Ù Often difficult to define inputs and outputs precisely.
Ù Measure of success often time-dependent and ill-defined.
Ù Require the use of good experimental technique, modular design, iteration.
e.g. Area routing on n layers with signal skew constraints, stub-length constraints, via constraints, and cross-talk constraints.