This model adequately approximates the performance measures.
Approximating the performance measures of M/G/1/K systems is a difficult, challenging, and important problem for applications in science and engineering.
Only one neural network (NN) is used as critic module for approximating the performance function in the solution procedure, and thus the architecture is simpler than the typical action-critic structure, which needs more computational load from neural networks.
When constructing the Kriging model, the proposed method only approximates the performance function in some region of interest, i.e., the region where the sign of response tends to be wrongly predicted.
The most effective features were obtained from abstract-level contexts, achieving 0.982 accuracy and 0.986 F1 score; this closely approximates the performance using all the features.
The similar outcome of the manifold PPCA and PLDA, which employ different extensions for the optimization problem, indicates that the achieved result may have approximated to the performance limit of the dataset using the probabilistic framework.
Nonlinear dynamic analysis results indicate that the column-loss response of the braced frames is approximated to the performance target and thus the proposed retrofit design approach is feasible for practical applications.
As it can be observed from Figure 14, there is a small gap between the simulation results and our mathematical model, although the model still approximates the real performance.
The Kriging model approximates the structural performance both in the design domain and in the stochastic domain, which allows to decouple the uncertainty quantification process and the optimization process.
The efficiency of this new method relies upon the advantages of SS in evaluating small failure probabilities and the Kriging model with active learning and updating characteristic for approximating the true performance function.
