Lea Harris

(Advisor: Prof. Dimitri Mavris] 

will defend a doctoral thesis entitled, 

A Probabilistic Approach for Margin Allocation Tradeoffs Pertaining to Early Two-Stage-To-Orbit Launch Vehicle Design

On 

Friday June 23rd, 1:00pm, EDT
Collaborative Design Environment (CoDE) 

Weber Space Science and Technology Building (SST II) 

And 

Click here to join the meeting

Meeting ID: 230 015 258 685
Passcode: WGb63Z

The evolution of complex aerospace vehicle designs traced by the progression of novel missions and state-of-the-art technology has created a continuous need to adapt to new uncertainties in the design process. This sequence introduces more uncertainties into the decision-making with increased sources and unknowns. Navigating early decision-making for requirements is often a trade between the predicted uncertainties and willingness to take on risk. Less distinguishable uncertainties are difficult to address with substantive data or examples, which drives traditional design margin allocation to account for uncertainties through a deterministic proportionate value related to the risks of design variation. There are significant complications in adapting margin standards for new missions and novel technologies. Without applicable or accessible data to inform the impacts of uncertainties more effectively, the industry is driven to apply conservative margins or iterate as the design is more defined, which can result in costly redesigns and schedule slips. The needed quantitative and probabilistic design methods are often implemented on the disciplinary design level during conceptual and preliminary design phases. There remains a need for quantitative methods in the systems engineering discipline to reduce uncertainties in decisions for estimated performance constraints and allocate margins. 

This research focuses on enabling a probabilistic design approach to tradeoffs during performance requirements and margin allocation to address underlying design uncertainties. The Probabilistic Uncertainty in Margin Allocation (PUMA) Framework developed by this research provides a foundation for decomposing the uncertain measures in a requirements decomposition and translating it into a quantitative modeling environment. The multidisciplinary and hybrid fidelity simulation environment harnesses conceptual design tools to identify drivers of uncertainty, employ model reductions for more accessible design simulations, and estimates response variability as a function of uncertainty parameters. The framework demonstration uses a Two-Stage-To-Orbit, single stack, launch vehicle concept with a mission to deliver a crewed payload to the ISS. The demonstrated scenarios for design and margin trades inform the likelihood of meeting aerodynamic, structural, and propulsion-based estimated performance measures. The tradeoff scenarios explored in this research increase the understanding of total variability due to uncertainties embedded in the estimated performance measures. The study evaluates the impacts of augmenting the design or adjusting constraints and allocated margins to meet the desired confidence level. The novel capability this framework provides is a quantitative approach to understanding uncertainty, substantiating decisions, and improving communication of uncertainty during the formulation phase of design. 

• Prof. Dimitri Mavris – School of Aerospace Engineering (advisor) 
• Prof. Daniel Schrage – School of Aerospace Engineering 
• Prof. Graeme J. Kennedy – School of Aerospace Engineering 

• Dr. Adam Cox – Research Engineer II, School of Aerospace Engineering 
• Mr. Bob Jurenko – Solutions Architect for Government Launch Services, Rocket Lab USA, Inc