Regression models that predict the compressive strength, tensilestrength, modulus of elasticity, and drying shrinkage of concrete made with several coarse aggregates are presented in this paper. The findings are part of a comprehensive research study being conducted in texas for determining the effect of aggregates on the performanceof concrete pavements. Using statistical analysis of laboratory test data from concrete samples made with eight commonly used aggregates of known chemical composition, regression models were developed topredict the concrete properties just mentioned. The predicted values were then compared with the laboratory test results, and the models were used to predict the concrete properties of 11 additional untested coarse aggregates also used for pavement construction in texas.The predicted concrete properties were within the pavement concreterange reported for these types of concrete mixtures. These prediction models can estimate preliminary information for coarse aggregatesor coarse aggregate blends from chemical composition data before casting and testing of concrete samples. This will allow for initial screening of proposed new sources of coarse aggregate or tentative blends of new or existing aggregate sources of known chemical composition. This paper appears in transportation research board record no. 1284, Cement, admixtures, and concrete 1990.
Abstract