Although an accurate reliability assessment is essential to build a resilient infrastructure, it usually requires time-consuming computation. To reduce the computational burden, machine learning-based surrogate models have been used extensively to predict the probability of failure for structural designs. Nevertheless, the surrogate model still needs to compute and assess a certain number of training samples to achieve sufficient prediction accuracy. This paper proposes a new surrogate method for reliability analysis called Adaptive Hyperball Kriging Reliability Analysis (AHKRA). The AHKRA method revolves around using a hyperball-based sampling region. The radius of the hyperball represents the precision of reliability analysis. It is iteratively adjusted based on the number of samples required to evaluate the probability of failure with a target coefficient of variation. AHKRA adopts samples in a hyperball instead of an n-sigma rule-based sampling region to avoid the curse of dimensionality. The application of AHKRA in ten mathematical and two practical cases verifies its accuracy, efficiency, and robustness as it outperforms previous Kriging-based methods.

Benefit-cost analysis (BCA) aims to help people make better decisions. But BCA does not always serve this role as well as intended. In particular, BCA’s aim of aggregating all attributes of concern to decision makers into a single, best-estimate metric can conflict with the differing world views and values that may be an inherent characteristic of many climate-related decisions. This paper argues that new approaches exist that can help reduce the tension between the benefits of providing useful, scientifically based information to decision makers and the costs of aggregating uncertainty and differing values into single best estimates. Enabled by new information technology, these approaches can summarize decision-relevant information in new ways. Viewed in this light, many limitations of BCA lie not in the approach itself, but with the way it is used. In particular, this paper will argue that the problem lies in a process that begins by first assigning agreed-upon values to all the relevant inputs and then using BCA to rank the desirability of alternative decision options. In contrast, BCA can be used as part of a process that begins by acknowledging a wide range of ethical and epistemological views, examines which combinations of views are most important in affecting the ranking among proposed decision options, and uses this information to identify and seek consensus on actions that are robust over a wide range of such views.