Direct loss-based seismic design of low-rise base-isolated reinforced concrete buildings

Abstract

This paper proposes a procedure to design low-rise base-isolated structures achieving a specific target level of earthquake-induced loss (e.g., dollars, downtime) while complying with a predefined minimum level of structural reliability. The procedure is “direct” since the target loss is specified at the first step of the process, and virtually no design iterations are required. Direct loss-based design (DLBD) is enabled by a simplified loss assessment module involving: (1) surrogate probabilistic seismic demand models representing the probability distribution of peak horizontal displacements and accelerations on top of the isolation layer conditional on different ground-motion intensity levels; (2) approximations of the superstructure response; (3) simplified consequence models for isolation system and superstructure based on damage-to-loss ratios (economic loss or repair time); (4) simplified consequence models for acceleration- and drift-sensitive non-structural components, based on storey loss functions of a potential inventory of components. Given some basic geometrical parameters of the superstructure, DLBD provides the isolation system's force-displacement curve and the required superstructure's strength complying with a selected loss target. The members' structural detailing follows the principles of direct displacement-based design, sectional analysis, and the general theory of base isolation. The procedure is illustrated by designing six reinforced concrete wall buildings (two-, three-, and four-storeys) base isolated with lead rubber bearings, to achieve predefined targets of expected repair time. Repair time is benchmarked against a more refined method adopting a cloud-based non-linear time history analysis, finding a maximum underestimation of 17%, thus confirming the dependability of DLBD. Such error is almost entirely attributable to the simplified estimation of peak floor accelerations, and it could be potentially eliminated by refining such estimation.