A method for evaluation of subsystem alternate designs

The current trend in aerospace business is to provide cost-effective programs and designs that are optimized to provide the most for the dollar but not overdesigned beyond the necessary requirements. Too many dollars can be spent on achieving results far above those required to meet the function for which the design is intended. A systematic quantitative method is required for deciding how much should be spent in a design effort to obtain better performance and reliability. This paper is an explanation of one method of combining these factors and an example of how it was applied to a project in evaluating alternate potential designs and subsystems. First, a set of criteria was established for use by the design engineers in evaluating their alternate designs. These criteria were ranked as to their relative importance for each subsystem. Weighting factors for the criteria for each subsystem were derived from the project managers through a polling procedure. A figure of merit (FOM) model was derived for combining the weighting factors and the criteria values. Sensitivity analyses were conducted by varying the criteria values and the weighting factors, which resulted in a range of FOM values. Alternate subsystem designs were evaluated and the FOM ranges of values were used as a basis for comparison. A clear maximum FOM range classified an alternate as the optimum choice. For three separate systems, a total of some 190 alternate designs were evaluated, generally 4–5 alternates per subsystem. Some of the selections were clearly optimal, but others were not clear choices; in some cases, the evaluation even produced results contrary to intuition. This evaluation technique was not used as a replacement for the decision maker, but as a tool to help him account for all important factors in his decision-making process.