[30]

The Effect of Uncertainty in Analyses of Advanced Life Support Systems  L.F. Rodriguez¹, A.J. Both¹, A.B.O. Soboyejo² and K.C. Ting^{2  1}Rutgers University, New Brunswick – NJ-NSCORT, The Ohio State University, Columbus

     During the development of complex, novel systems, such as Advanced Life Support (ALS) Systems, research and technology development is a necessity to provide knowledge bases for achieving system and mission requirements.  When considering technologies still in the developmental phase, it is essential to consider the uncertainty of the observed and reported data.  When the data is utilized in subsequent analysis, results will be expected to carry an inherited amount of uncertainty.  One example is an ALS metric such as ESM, which is considered a valuable piece of decision support information, that will likely carry some degree of uncertainty, and decision makers should be aware of the extent of uncertainty.

     By considering the variability of the data utilized to evaluate a model, it is possible to determine the amount of variability in the result by propagating the error in the data throughout the calculation.  An ongoing ALS System top-level modeling effort, supported by the NJ-NSCORT at Rutgers University,  provides an ideal platform for demonstration of this error propagation technique.  Limiting this effort is information describing the uncertainty of data reported by ALS System researchers.  Another source of uncertainty may originate from the need of making estimates for incomplete information during analyses.  Therefore, reasonable assumptions regarding the uncertainty in data generated by the ALS community have been made and non-trivial uncertainty in ESM calculations has been found based on the assumptions underlying this study.  To mitigate the effect of uncertainty on the decision process, it is necessary to initiate data collection of information describing the uncertainty of a process at the fundamental research level.  This will enable researchers and decision makers to have a better understanding of the uncertainty in their analyses used in the ALS decision making process.