i .e., a “will-cost.” One cannot, of course, expect a cost model to compensate for deficiencies in design definition. [2] The detailed estimate reflects a cost forecast for an orderly, well-executed program with no serious technical or management problems. It is ordinarily produced in a competitive environment in which it is recognized by those involved that only the lowest rationalizable estimate can win. Critics of parametric costing sometimes argue that parametric models predict things that are more massive to be more expensive, whereas “everyone knows it costs more to create something lighter.” This is a specious argument. The general trending of cost estimating relationships is clearly correct, e.g., large structures are more expensive than small ones of the same general design. Mass reduction for a given design is handled in a parametric model through “difficulty factors,” sometimes called complexity factors. Difficulty factors range from about 0.3 to about 2.0 with 1.0 being typical aerospace design. If the mass reduction is accomplished by more sophistication, e.g., fuel cells rather than batteries, the model identifies the higher cost through higher cost of the subsystem elements. There are no historical analogies for space manufacturing. No such history exists. Therefore, one cannot directly use parametric methods directly. (There is an indirect, “back-door” method which I describe below.) Consequently, the space manufacturing studies have tended to use enumerative estimating, i.e., estimated numbers of people employed multiplied by salaries and schedule times with materials and capital cost. Cost estimating experience indicates these estimates should not be compared to the SPS reference system costs. Enumerative estimates are only accurate when detailed program plans have been developed; even then they tend to assume a “success” schedule and are likely to be biased low by a factor of roughly two. Enumerative estimates based on conceptual TABLE 3 COST DELTA FACTORS
RkJQdWJsaXNoZXIy MTU5NjU0Mg==