A critical review of probabilistic live load models for buildings: Models, surveys, Eurocode statistics and reliability-based calibration

Abstract

The appropriate assignment of gravity live loads is of fundamental importance in establishing a design base for safe and economical structures. This paper presents a critical review of the probabilistic modelling of live loads in buildings, providing a historical overview of the theoretical models found in the literature, as well as the load surveys that provided the empirical evidence for their development. As a general rule, these models divide the live load into sustained and extraordinary components, represented as a Poisson square wave and spike processes, respectively, with the differences lying in the underlying hypotheses and process parameters. A comparison of different models is presented, and it is shown that the model parameters currently in use in background documents on the reliability of the Eurocodes do not agree well with survey data, leading to an overestimation of extreme loads. While the Eurocodes lack clear specification regarding the exceedance probability for design loads, this study demonstrates, using a model with modified parameters, that the office design load corresponds to an approximate 27% exceedance probability over 50 years for a reference area of 20 m${}^2$. The impact of employing this model on the reliability-based calibration of the Eurocode partial safety factors for loads is also examined, and it is found that the average reliability falls somewhat below the prescribed 50-year target reliability index of 3.8.