Abstract
Creep of concrete is an important design consideration. National design codes typically provide empirically based models for the estimation of creep deformation. Such models estimate a creep coefficient (f), which is the ratio of the creep strain (ec) to the elastic strain (ee), which is used to predict the creep strain (ec) at any age. This paper assesses the accuracy of the creep coefficients (f) predicted by the latest American Association of State Highway and Transportation Officials (AASHTO), Load and Resistance Factor Design (LRFD) Bridge Design Specification. The accuracy of the AASHTO Model was evaluated by comparing the predicted with the actual (measured) creep coefficients (f), on a range of concretes under laboratory-controlled conditions, for six mixes. The six mixes comprised three aggregate types (quartzite, granite and andesite) and two strength grades (water-cement ratios of 0.55 and 0.4). In the case of all six mixes, at all ages considered, the AASHTO LRFD Model underpredicted the creep coefficient (f). Furthermore, the predictions were most and least accurate in the case of the quartzite and andesite concretes, respectively. In the case of all six mixes, a highly significant linear relationship (maximum P = 1.4E-11 %) was found between the predicted f and actual f. When considering all six mixes, an overall coefficient of variation (ωall) of 68.5 % was obtained. The results of this investigation were compared to those of 19 other models and the AASHTO LRFD Model was the 16 th most accurate model of the 20 models assessed to date. Introduction Creep magnitude is an important design consideration for the durability, long-term serviceability and the load carrying capacity of structures. Creep strain can result in loss of serviceability due to unacceptable deflections or loss of prestress. Creep strain can be determined by relatively costly and time-consuming laboratory testing or estimated by means of empirically based models of various complexities. In general, the more deformation sensitive the structure, the more laboratory testing is justified. In cases where an approximation of creep is required, design code-type models suffice. Almost all code-type creep models express creep strain in terms of the creep coefficient, (f), as shown in Eq. 1.