Seismic risk mitigation in buildings using a new method to encode a joint weighting function in multi-attribute utility theory

Abstract

This paper provides a new simple, nonparametric method that directly elicits a joint weighting function to be immediately associated with multi-attribute utility under uncertainty. The method allows to include the subjective perception of probabilities and consequences in the decision-making process while dealing with multi-attributes. The concept was only possible when dealing with one outcome; this study demonstrates that it can be achieved for multiple outcomes by mean of a parametric method. First, the newly proposed procedure is described. Then, the experimental protocol is presented while applied to the performance-based earthquake engineering (PBEE) methodology. The decision-making process of PBEE is improved by including the preferences of the decision maker, which is usually the owner. This improvement is important, especially in seismic risk mitigation, because the problem involves low probabilities, which may be inaccurately estimated or even unknown, and these are coupled with catastrophic events. Such combined factors, when objectively weighed, lead to decisions counter to the distorted perception of probabilities and consequences that reflect the implicit desires of the decision maker. Using the newly proposed method, the owner, whose own preferences are elicited, is more likely to embrace the solution since he is involved in the decision process. The results and implications of the analysis are discussed showing the difference in results while including the joint weighting function in the decision-making procedure. The method, which is a step forward in multi-attribute utility theory, is not restricted to the decision-making process applied to PBEE but can be implemented to any multi-attribute decision analysis problem.