Introducing variance-based global sensitivity analysis for uncertainty enabled operational and economic aircraft technology assessment

Assessing the efficacy of aircraft and technologies is a crucial step in the aeronautic product development. Due to their prospective nature, such analyses are subject to various uncertainties, which decrease their transparency and can act as a barrier for innovation. Therefore, a comprehensive and systematic inclusion of uncertainties is needed. For the use case of this paper being Hybrid Laminar Flow Control (HLFC), these comprise a number of inputs from different domains such as the design, operation, environment, or economics, leading to the underlying research question: Which parameters contribute to what extent to the output uncertainty of HLFC efficacy? To answer this, this paper introduces variance based global sensitivity analyses (GSA) in combination with a discrete event simulation framework. Considered outputs of the assessment comprise the Net Present Value (NPV) as an economic metric as well as the fuel efficiency as an operational performance indicator. A first monte-carlo simulation revealed an equivocal assessment outcome with Delta-NPV values ranging from -$ 11M up to +$ 22M. Results of a subsequent GSA reveal the cruise mach speed being the dominant element (39%) in this overall uncertainty, followed by the lack of knowledge in specific fuel consumption penalty (23%) and overall drag reduction potential (9%). Furthermore, this study includes an applied comparison of different GSA methods, i.e. Fourier Amplitude Sensitivity Test and the Sobol' method. While computed sensitivity indices are fairly equal among all alternatives, the lack of quantification of higher order terms of certain methods may pose a disadvantage if interaction effects are high. The contribution of this study helps to direct further research and development efforts regarding HLFC, e.g. to focus on a robust and aerodynamically less sensitive airfoil design. More generally, the use of GSA methods and further uncertainty quantification efforts in prospective assessments can significantly improve the trust in the analysis and with that, help in terms of a commercially successful HLFC development