Evaluation of Cost Improvement Models When Programs Experience Unplanned Production Decreases

Abstract : As military and other governmental budgets decline and impacted project deadline changes require instantaneous responses, cost analysts' tasks become more and more formidable. Inaccurate estimates can lead to misappropriation of resources and can thus create delays in goods reaching warfighters. This thesis aims to avail cost estimators of more reliable projection tools and to challenge the status quo of cost estimating, the production rate cost improvement model, when programs face reductions in lot quantities. The findings reveal that the status quo proves efficient under many cost profiles, but clearly does not estimate as well when a program suffers lot quantity reduction coupled with loss of cost efficiency. Prior research recognized the importance of changes in lot quantity to cost estimating, but definitive guidance never surfaced with regards to choosing a model. Monte Carlo simulation allows us to vary cost-affecting variables and isolate conditions where the use of a fixed cost, cost improvement model provides more accurate estimates than does the status quo. While no model for estimation should be discounted without exploration of its usefulness, we argue that the fixed cost model should be considered for use based on its ability to predict increases in average unit cost.