Correlation of Creep and Shrinkage Models with Field Observations

Abstract

This paper presents a comparison between laboratory creep and shrinkage data and the performance of a long span box girder structure. As a part of a long term instrumentation and monitoring program on the North Halawa Valley Viaduct, a number of laboratory creep and shrinkage tests were performed on the concrete used in the viaduct. The data are compared with current theoretical models for the prediction of creep and shrinkage. These models all underestimate the long term effect on the concrete used in this structure. This is attributed in part to the nature of the basalt aggregates available on the Hawaiian Island. It is shown that with as little as 28 days of data from creep and shrinkage tests, the prediction models can be modified to provide a far more accurate prediction of the long term performance of the structure.