Complexity based cost estimating relationships for space systems

As cost estimating methods evolve and grow increasingly important in the lifecycle of complex engineering feats, effective cost analysis methods are finding their way into the early system level design phases. What follows outlines and how to develop a parametrically determined cost estimating relationship (CER) that can be used to support the system design tradeoffs that take place throughout the study and development phases of space systems project. There are three steps in building a CCER. First, the key technical elements that determine the complexity of the subsystem are hypothesized; secondly, the elements are ranked with respect to the characteristics of the same class in our historical database; finally, all individual ranks are combined to quantify the overall subsystem complexity. At this point the systems engineer must make use of his or her experience and expertise to verify the results given by the CCER by answering the following questions, among others: was subsystem A more complex than subsystem B? If so did it affect costs accordingly? What about the schedule or time lapse to complete the product? How did these three parameters interact? Once the CCER answers the above questions, it can be used to support the design process of a new subsystem, by evaluating the impact on complexity of specific design choices (i.e. battery type, solar cell type and therefore needed area for a given power requirement, etc.). As an example a satellite subsystem complexity based CER (CCER) is built by applying the complexity index theory introduced by David A. Bearden in his paper, "A complexity-based risk assessment of low-cost planetary missions: when is a mission too fast and too cheap?" presented at the Aerospace Corporation 4th IAA international conference on low-cost planetary missions, in May 2000. By using technical characteristics to build our CCER, we are able to bridge the gap between cost, schedule, risk and the design of a system. System engineers can finally adopt algorithms that determine complexity and, through complexity, determine the cost and schedule impact of a design choice. Therefore, just as we are used to determining battery and solar array sizes based on our power needs, and defining radiating surfaces based on our thermal dissipation needs, we can aspire to define a design choice based on its cost and schedule impacts. In the future perhaps we can start to see complexity budgets based on cost and time limits, side by side with mass and power budgets. As we mention budgets we need to consider ranges of values and hence, distributions. Therefore, as we collect the data to build our CER, which is the basis of all of our work, we must keep track of the uncertainty that lies in the values we use. Keeping track of these values allow us to determine complexity distributions associated with our design elements and hence establish ranges of cost schedule distributions and quantify risk.