A compressive laboratory testing program, field testing program, numerical analysis, and life-cycle cost analysis were conducted to evaluate the beneficial effects of incorporating shrinkage-reducing admixture (SRA), polymeric microfibers (PMF), and optimized aggregate gradation (OAG) into an internally cured concrete (ICC) mix for rigid pavement application. Results from the laboratory program indicate that all ICC mixes outperformed the standard concrete (SC) mix. All ICC mixes showed a decrease in drying shrinkage compared to the SC mix. Based on the laboratory program, three ICC mixes and one of the SC mixes were selected for the full-scale test subjected to a heavy vehicle simulator for accelerated fatigue testing. Extensive testing and analysis have shown that ICC mixes incorporating SRA, PMF, and OAG can be beneficially used in pavement applications to achieve increased pavement life.

To promote the application of pervious concrete (PC) in heavy-duty pavement engineering, a thick plate (approximately 50 to 100 cm) paving structure can be used, and its failure form mainly by fatigue compression. Therefore, compressive fatigue tests were carried out under fatigue loads in four stress levels (S): 0.6, 0.7, 0.8, and 0.9, at three loading frequencies of 10, 15, and 20 Hz. The results showed that the fatigue life (N) and fatigue residual strength are controlled by S, while loading frequency showed no statistically significant effect on them. The fatigue failure of PC will not occur under a stress level of 0.6. The survival rate of PC and the fatigue life of uniaxial compression obey a Weibull distribution with two parameters. The material constants of uniaxial compression fatigue of PC are 0.0464 to 0.052, which are similar to ordinary concrete. There are two forms of fatigue failure: one is the shearing along the vertical central axis and the other is shear failure at an angle of 15 to 30 degrees with the vertical central axis.

The fracture of longitudinal reinforcing steel causes the loss of load-carrying capacity in reinforced concrete (RC) members. Results of large-scale reverse cyclic column tests have indicated that the fracture of longitudinal reinforcement is influenced by the amount of buckling experienced by the reinforcing steel. Similar behavior was observed in a material test as reinforcing bars fractured in a brittle manner when pulled in tension after buckling. Brittle fracture occurred after the bending strain from buckling exceeded the critical bending strain. A material test was developed to quantify the critical bending strain, called the buckled bar tension test. The rib radius of the reinforcing bar was found to influence the magnitude of the critical bending strain. Additionally, the results of column tests indicated that the critical bending strain of the longitudinal reinforcement affected the column displacement capacity. Finally, a relationship between axial displacement and strain from bending was developed.

Due to the difficulties of direct tensile tests on concrete, available data are limited and conflicting. Cyclic tensile tests on concrete have been carried out by the authors of this paper to investigate the effect of stress level on fatigue behaviors of concrete. Four different stress levels—0.95, 0.90, 0.85, and 0.80—have been employed. The relationship between stress level and fatigue life can be expressed as a logarithmic linear function S = a + blogNf at different probability of failure p, which indicates logarithmic normal distribution can describe the fatigue lives of concrete. The maximum strain accumulation can be distinguished as three typical stages: the rapid accumulation stage, the stable accumulation stage, and the accelerated accumulation stage. The dissipated energy decreases during the first few cycles and then steadily increases as the acceleration to failure is reached. The dissipated energy increases with the increasing of stress levels. As the dissipated energy can reflect both the strength and damping effect of concrete, a power function is selected in this paper to describe the relationship between the fatigue lives and the average dissipated energy. The F-test and t-test of the power model confirm the significance of the averaged dissipated energy in the model. The high value of R2 for the predicted model shows good predictability.

Despite rigorous efforts in the derivation of various fatigue damage models for concrete, damage predictions of sufficient accuracy are still limited to loading conditions similar to those of the experiments used for developing the models. Most models are void of salient factors affecting the fatigue behavior of concrete such as frequency, stress ratio, and loading waveform, and the approaches used in developing such models tend to be rudimentary. Therefore, further investigation is required. In this study, damage models are expressed for residual concrete strength and fatigue secant modulus using experimental data from tested cylindrical specimens, a damage function, and a stress-life model in the literature. The number of cycles leading to failure, required for normalizing the fatigue cycles for each specimen, is obtained using a proposed secondary strain rate model. The aforementioned influencing factors incorporated into the damage function result in robust models that account for variations in loading parameters.