Abstract
Designing a system that is verifiable with respect to mission requirements is difficult, given that there is no a priori system to verify. Rather than ask if the hypothetical system is verified, we appeal to probabilistic bounds on system performance to determine how likely will a new system will meet the mission requirements. For these bounds, we use available capability performance probabilities that characterize the hypothetical system. Assuming cost estimates that directly vary with respect to system performance estimates, we give an algorithm that finds the set of multi-objective optimal solutions. We use the algorithm and cost estimates to construct regions that estimate the actual set of optimal solutions; we demonstrate this approach by using generated data that shows these regions are good estimators of the set of optimal solutions. The generated data adheres to proven probabilistic relationships between the system performance data and capability performance data.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
