Collaborative Research: Maximizing the Expected Value of Engineered Systems Through Coordinated Verification

Virginia Polytechnic Institute And State University, Blacksburg VA

Investigators

Abstract

Verification is a crucial process in the development of complex systems during which engineers build confidence that their system functions as intended. Because resources are limited, engineers must be strategic in the planning and execution of verification activities. Many large engineered systems are composed of parts and subsystems developed by different firms, which complicates system verification. Much of the verification process is reactive rather than proactive and impacted by schedule and budget pressure. This research project will create and evaluate a mathematical framework of incentives as a coordination mechanism among engineering teams during system verification. Individual teams will define their own verification strategies while maximizing the value of the overall system. If successful, this research will benefit the public through increased safety and efficacy of commercial products and public services. It will enable U.S. industry to be more competitive by making system verification more efficient and effective. The project also includes active recruiting and mentoring activities that will advance the interests and skills of women and other underrepresented groups in engineering. The objective of this research is to mathematically derive value-maximizing system verification strategies for multi-firm systems engineering projects. This method enables computing verification strategies that maximize a firm's objective function, while accounting for the interdependencies between the firms, the components they develop, and the value of the overall system. Moreover, the framework will account for and enable the design of contractual agreements between firms in order to ensure that locally optimal firm verification decisions contribute to the global goal of maximizing system value. To achieve this research objective, mathematical methods from operations research, decision theory, and economics will be combined and advanced into one framework. The mathematical foundations of the research include partially observable Markov decision processes for deriving optimal verification strategies for each firm, multiscale decision theory to mathematically capture the interdependencies between firms, and contract theory to design effective contracts that can implement globally optimal verifications strategies. Studies of engineering projects will provide empirical evidence about these methods. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.