Balancing Design Functional Coupling and Sensitivity to Noise to Achieve the Design Target 2007-01-1207
The primary objective in design is to achieve the target value of the design's response function. If a design fails to achieve the target value, it most likely fails in two ways: inconsistent functional output and in design involving multiple response functions, unable to converge to the multiple target values in spite of iterative adjustment of the design parameters. The former is symptom of a design not able to perform in the presence of variability, i.e., noise. The latter is symptom of a design that fails to perform in the presence of functional coupling. Both problems are best addressed at the conceptual stage of the design at which only design solution that is inherently robust to noise and functionally uncoupled is entertained. If this is not possible, the alternative is to exploit the interaction between control variables and variables that are sources of noise and functional coupling to render the design insensitive to them. Thus situations may arise wherein the same control variables are involved in reducing the effect of functional coupling and of noise. In these situations, balancing the conflicting needs to reduce design functional coupling and sensitivity to noise is the key to achieving the design targets. In this paper, we formulate the metrics for design functional coupling and design sensitivity to noise. These metrics are then used to balance the conflicting needs of reducing the two in an illustrative example.