Skip to main content

Currently Skimming:

4 Improvements for the Input Data to TARAM
Pages 23-28

The Chapter Skim interface presents what we've algorithmically identified as the most significant single chunk of text within every page in the chapter.
Select key terms on the right to highlight them within pages of the chapter.


From page 23...
... IMPROVING COMPLETENESS, ACCESSIBILITY, AND QUALITY OF INPUT DATA TO TARAM The quality of the estimated risk depends on the trustworthiness and quality of the input data. While the current TARAM Handbook requires specific input data to assess the risk of failure, the data sources to be used are not always explicitly specified.
From page 24...
... Based on the FAA briefing regarding the Seattle ACO Transport Airplane Safety Manual, when assessments require pilot human reliability and error analysis, flight test pilots' subjective judgment has been used and, if needed, simulator data have been created to support the analysis.3 Currently, there is no documented framework for the human reliability data generation or sharing processes. The incorporation of human reliability analysis into the TARAM process is discussed in Chapter 5 leading to Recommendation 5.
From page 25...
... CHARACTERIZING UNCERTAINTY IN INPUT DATA TO TARAM Most data inputs to TARAM lack uncertainty characterization. Uncertainties in the TARAM input data can be characterized by leveraging approaches used in PRA of other complex technological systems, such as those for the nuclear power plants8 and space exploration.9 Uncertainty considered in PRA arises from two distinct sources: the intrinsic random behavior of a system (referred to as "aleatory uncertainty")
From page 26...
... Based on the FAA briefing regarding the Seattle ACO Transport Airplane Safety Manual, in the current practice of COS decisionmaking for transport airplanes, risk sensitivity is sometimes studied by examining the impact of varying each input or modeling assumption on the risk outputs. The TARAM sensitivity analysis is conducted in a one-at-a-time manner and using only a predefined discrete input value or variation in a modeling assumption; thus, the current practice is limited and does not capture the entire range of epistemic uncertainty.
From page 27...
... The characterized uncertainty in TARAM input data could be an input to the uncertainty propagation and sensitivity analysis methods that are discussed in Chapter 5 to support Recommendation 7, which states the need for incorporating uncertainty analysis into the TARAM process. Through uncertainty characterization and propagation, the uncertainty for the TARAM risk outputs could be quantified to quantitatively represent the degree of confidence in the estimated risk values.
From page 28...
... . The physics-of-failure simulation would be equipped with uncertainty quantification to make it probabilistic, creating a probabilistic physics-of-failure model.16 Bayesian analysis could then be used to facilitate the updating of the data-driven DA estimate with the probabilistic physics-of-failure simulation data.


This material may be derived from roughly machine-read images, and so is provided only to facilitate research.
More information on Chapter Skim is available.