First-order approximation and model management is a methodology for a systematic use of variable-fidelity models or approximations in optimization. The intent of model management is to attain convergence to high-fidelity solutions with minimal expense in high-fidelity computations. The savings in terms of computationally intensive evaluations depends on the ability of the available lower-fideli...

ing logical rules should be justified and based on the data that we can observe from a real system. In other words, if a logical rule is abstracted, the input data for the abstracted logical rule should be observable in the real system, and its output data also should be comparable to the observable results of the real system. An abstracted model does not provide more I/O data than the original...

In this paper, a novel multi-fidelity modelling-based optimisation framework is developed for the robust design of composite structures. The proposed provides significant savings on computation time compared to both conventional and high-fidelity modelling methods while maintaining an acceptable level accuracy. Artificial neural networks (ANNs) multi-level approach are incorporated into formula...

Multifidelity Monte Carlo (MFMC) estimation combines lowand high-fidelity models to speedup the estimation of statistics of the high-fidelity model outputs. MFMC optimally samples the lowand high-fidelity models such that the MFMC estimator has minimal mean-squared error for a given computational budget. In the setup of MFMC, the low-fidelity models are static, i.e., they are given and fixed an...