A Practical Method for Uncertainty Analysis in the Aircraft Conceptual Design Phase

A methodology has been developed and tested to propagate the performance uncertainty of emerging technologies from the system level to the aircraft level. This allows one to get an insight in how performance uncertainties of an emerging technology translate to the uncertainties of aircraft key performance indicators (KPIs) such as fuel burn and maximum take-off mass. First, the Morris-One-At-a-Time method is applied which ranks the input parameters in terms of their effect on a set of predefined KPIs. This is followed by the quantification of the input uncertainties using the four-step expert elicitation process. Finally, an interval analysis is carried out which propagates the interval bounds of the input parameters to the interval bounds of the KPIs of interest. The efficient global optimization is used to find the resulting interval bounds of the aircraft KPIs. The results are visualized such that decision-makers are given a quick and easy assessment of the most-likely effect of an emerging technology together with its uncertainty interval. A case study is carried out in which the method is applied to a design study of a laminated electrical wing ice protection system (heated GLARE). The effect of this technology on the maximum take-off mass and total fuel mass is examined. The results show a most-likely reduction in total fuel mass of 1.04% compared to the conventional bleed air system, for an aircraft comparable to an Airbus A320. The uncertainty analysis shows an interval bounded by -0.11% and -1.59%. Based on this test case, the proposed method is compared to alternative approaches of uncertainty modeling and uncertainty propagation.