Validating Human Performance Models of the Future Orion Crew Exploration Vehicle

Abstract

NASA's Orion Crew Exploration Vehicle (CEV) will provide transportation for crew and cargo to and from destinations in support of the Constellation Architecture Design Reference Missions. Discrete Event Simulation (DES) is one of the design methods NASA employs for crew performance of the CEV. During the early development of the CEV, NASA and its prime Orion contractor Lockheed Martin (LM) strived to seek an effective low-cost method for developing and validating human performance DES models. This paper focuses on the method developed while creating a DES model for the CEV Rendezvous, Proximity Operations, and Docking (RPOD) task to the International Space Station. Our approach to validation was to attack the problem from several fronts. First, we began the development of the model early in the CEV design stage. Second, we adhered strictly to M&S development standards. Third, we involved the stakeholders, NASA astronauts, subject matter experts, and NASA's modeling and simulation development connnunity throughout. Fourth, we applied standard and easy-to-conduct methods to ensure the model's accuracy. Lastly, we reviewed the data from an earlier human-in-the-loop RPOD simulation that had different objectives, which provided us an additional means to estimate the model's confidence level. The results revealed that a majority of the DES model was a reasonable representation of the current CEV design.

Predictive mental workload modeling for semiautonomous system design: Implications for systems of systems

“…Full formal verification, validation, and accreditation of preliminary models of future or conceptual systems is extremely difficult, due to the inaccessibility or impracticality of data collection, and/or due to the cost and time demands associated with these activities. The model verification and validation step applied recommended practices [Modeling and Simulation Coordination Office, 2006] which have been successful in other similar efforts to model future systems [Wong, 2010]. The practices for this research effort included Desk Checking, Face Validation, Reviews, and Walkthroughs to insure consistency of the model logic with the architecture and task structure.…”