Active learning finds better ship designs faster and cheaper
Robust Active Learning for More Efficient and Reliable Paths to Optimal Design
Computational Engineering, Finance, and ScienceMachine Learning
Summary
Designing things like ships often means running many costly computer simulations to find the best shapes and features. The authors present a way to pick which simulations to run, balancing between cheap but less accurate ones and expensive but precise ones. Their method helps decide when to test a design cheaply before using a costly test or to go straight to the expensive test. They showed on a complex ship design problem that their approach found better designs using the same or less computing power compared to previous methods.
What this means in practice
- •For aerospace design teams: Optimize aircraft components by efficiently choosing when to run costly simulations versus cheaper approximations to reduce development time.
- •For automotive engineering groups: Speed up vehicle design by balancing between fast but rough simulations and detailed ones to find better designs under tight budgets.
Authors
Hayriye Pehlivan Solak, Francesco Di Fiore, Jeroen Wackers, Riccardo Pellegrini, Andrea Serani, Matteo Diez, Laura Mainini
Abstract
Simulation-driven design optimization requires active learning strategies that efficiently combine models with different computational costs and levels of fidelity. We introduce Multifidelity Cost-Aware Screening (MF-CAS), which enables the joint selection of sampling location and fidelity using a surrogate that provides decomposed uncertainty intervals rather than a complete predictive probability distribution. MF-CAS exploits the contribution of each fidelity to the high-fidelity uncertainty together with its computational cost to determine whether a candidate should first be evaluated at low fidelity or directly progressed towards high fidelity. The method is assessed on the constrained 14-dimensional hydrodynamic optimization of the DTMB 5415 hull under a deliberately limited computational budget. MF-CAS achieves an approximately 9.0% resistance reduction, compared with 6.7% for the best baseline optimisation method, demonstrating that incorporating fidelity-resolved uncertainty directly into the joint location-fidelity decision can substantially improve multifidelity active learning efficiency.