Challenges of Sustaining the International Space Station through 2020 and Beyond: Including Epistemic Uncertainty in Reassessing Confidence Targets

Abstract

This paper introduces an analytical approach, Probability and Confidence Trade-space (PACT), which can be used to assess uncertainty in International Space Station (ISS) hardware sparing necessary to extend the life of the vehicle. There are several key areas under consideration in this research. We investigate what sparing confidence targets may be reasonable to ensure vehicle survivability and for completion of science on the ISS. The results of the analysis will provide a methodological basis for reassessing vehicle subsystem confidence targets. An ongoing annual analysis currently compares the probability of existing spares exceeding the total expected unit demand of the Orbital Replacement Unit (ORU) in functional hierarchies approximating the vehicle subsystems. In cases where the functional hierarchies' availability does not meet subsystem confidence targets, the current sparing analysis further identifies which ORUs may require additional spares to extend the life of the ISS. The resulting probability is dependent upon hardware reliability estimates. However, the ISS hardware fleet carries considerable epistemic uncertainty (uncertainty in the knowledge of the true hardware failure rate), which does not currently factor into the annual sparing analysis. The existing confidence targets may be conservative. This paper will also discuss how confidence targets may be relaxed based on the inclusion of epistemic uncertainty for each ORU. The paper will conclude with strengths and limitations for implementing the analytical approach in sustaining the ISS through end of life, 2020 and beyond.

Supportability Challenges, Metrics, and Key Decisions for Future Human Spaceflight

“…40 As an example, consider the roll of a die. The outcome of the roll, given known probabilities associated with each potential outcome, is associated with aleatory uncertainty.…”

“…The distribution of the number of failures that may be experienced can be used to determine the number of spares that must be provided in order to achieve a desired Probability of Sufficiency (POS), a metric defined as the probability that the number of spares provided is sufficient to cover all maintenance demands during a given period of time. 40,41 POS is a function of a variety of factors, depending on the complexity of the model being used. These typically include mission endurance, item failure rates (or Mean Times Between Failures (MTBFs)), and the number of spares provided.…”

“…An in-depth discussion of risk, reliability, and POS modeling is beyond the scope of this paper; however, a selection of references is provided here that include more detailed descriptions of various modeling approaches. 1,2,8,15,16,27,32,37,38,[40][41][42][43][44][45][46][47][48][49][50][51][52][53][54][55][56][57][58][59] Epistemic uncertainty is also a critical consideration for spares and maintenance logistics. 8,36 When failure rates are (or, at least, are assumed to be) deterministically known, POS for a given item is a deterministic value.…”

“…Instead, it must be estimated based on similarity to other components and/or statistical analysis of observed behavior. For example, the failure rate estimates for ISS ORUs are updated based on observed failures using a Bayesian approach, as described by Vitali and Lutomski 60 and Anderson et al 40 An analysis by Stromgren et al 8 found that, of the items in Maintenance and Analysis Data Set (MADS) -the data set used by ISS Safety and Mission Assurance (S&MA) to track reliability characteristics for all ISS ORUs -that have had their reliability estimates change as a result of Bayesian updates, approximately 85% of MTBF values were adjusted upwards (indicating higher than predicted reliability), while 15% were adjusted downwards. While Bayesian updating of failure rate estimates can reduce uncertainty in these estimates over time, some amount of epistemic uncertainty remains.…”

“…If test/operational experience is relatively limited or the number of test articles is small -as is often the case for human spaceflight when compared to, for example, the aviation or automotive industries -a significant amount of epistemic uncertainty may be present in failure rate estimates. 40 This means that assessment of POS values (which are a function of failure rates) result in a distribution of possible values rather than a point estimate. To account for this epistemic uncertainty, analysts must consider not only POS but also confidence, a measure of fidelity of the estimate typically expressed in terms of the probability that the real-world POS value is greater than or equal to the estimated value.…”