Evaluation of Extreme Events and Reliability in the Context of the Mars Sample Return Mission

• Activity Number: 493 - Uncertainty Quantification for the Mars Sample Return Mission
• Type: Invited
• Date/Time: Thursday, August 11, 2022 : 8:30 AM to 10:20 AM
• Sponsor: Uncertainty Quantification in Complex Systems Interest Group
• Author(s): Richard Smith* and Dawn Sanderson and Maggie Johnson and Amy Braverman
• Companies: University of North Carolina Chapel Hill and University of North Carolina Chapel Hill and Jet Propulsion Laboratory, California Institute of Technology and Jet Propulsion Laboratory, California Institute of Technology
• Keywords: Extreme Value Theory; Reliability; Uncertainty Quantification; Computer Model Simulations

Abstract

The Mars Sample Return (MSR) mission involves a complex interplay of statistical and engineering calculations. The key statistical requirement is a target probability of less than one in a million that the biosphere is contaminated by Martian material. Achieving this requires a statistical model involving both reliability analysis and extreme value distributions to assess the probability of a rare event. A Bayesian analysis is adopted so that the results may be expressed in terms of the posterior mean of the failure probability. A key component of the analysis is the use of computer model simulations to assess probabilities of failure, but it’s not feasible to run millions of simulations of a computer model so some form of extreme value approximation will be required. The talk will compare different methods of achieving the target probability and assess their overall effectiveness.