Research for Early Warning

This paper models cumulative cost and schedule for research and development contracts using the non-linear Rayleigh distribution. Parametric estimation, using the actual data to date of an executing contract, yields independent forecasts of final contract cost and schedule that give early warning of potential execution difficulties to decision-makers. Overall contract cost and schedule risk can also be calculated. The model was rigorously tested and validated against 107 completed development contracts drawn from the DOD database over a 35 year period. A software application was also developed to graphically portray trends and includes automated business insights. In addition, a plan assessment module was developed to evaluate plan realism. This module can be used to assess the realism of a contractor's offer during source selection and to assess plan realism early in contract execution, even before actual cost data are available. The module may also be used to assess the realism of research and development funding profiles.