The practical application of fiber-reinforced concrete (FRC) in structural components has gained growing interest due to structural advantages such as improved tensile strength, distributed load transfer, crack width control, as well as superior durability. To this end, reliable structural assessment techniques and analytical models have been developed, placing emphasis on tension-softening behavior owing to the bond and pull-out mechanisms of fibers at a local crack. However, these models could not be directly applicable to evaluate the multiple cracking mechanisms of high-performance fiber-reinforced concrete (HPFRC), which exhibits strain-hardening behavior. To overcome this challenge, this paper presents a probabilistic analytical technique. This approach has employed the simplified diverse embedment model (SDEM). Then, an HPFRC member was modeled with multiple segments considering the most probable number of cracks. It was assumed that material properties had a normal probability distribution and were randomly assigned to each segment. To have reliable results, 10,000 analyses were performed for each analysis case and validated using experimental test data. Based on the analysis results, the actual strain-hardening tensile behavior of an HPFRC member could be reasonably predicted with the number of segments chosen on the basis of the fiber length.
Keywords: crack; fiber transfer length; high-performance fiber-reinforced concrete (HPFRC); probability; strain hardening.