Compressive strength is one of the most critical mechanical properties of various types of concrete and is the main input variable for structural concrete design. Recently, with the advances in concrete technology, applications of fiber-reinforced concrete (FRC), including ultra-high performance concrete (UHPC), have grown rapidly. These new types of concrete are well known to exhibit superior mechanical characteristics such as compressive strength, fracture toughness, and durability compared to conventional concrete and thus are popularly used in urgent repair jobs where compressive strength is an important parameter to determine the required curing time until open to the public. Considering the importance of compressive strength in practice, this study aims to evaluate the effect of age and maturity on the compressive strength characteristics of three different types of concrete, namely UHPC with micro and macro steel fibers, FRC, and plain concrete, and to propose a new design strength criterion for UHPC. To this end, 180 concrete cube specimens were tested at 12 different ages between 3 and 126 days. The results indicated that irrespective of the type and presence of fibers, UHPC gained more than 90% of their ultimate compressive strength after only 21 days, while FRC and plain concrete specimens required a longer time (i.e., 28 days) to achieve 90% of their ultimate strength. Therefore, UHPC may adopt a 21-day compressive strength as a design input instead of a 28-day compressive strength commonly required for structural concrete specified by many codes of practice. Moreover, the obtained experimental results were compared with existing compressive strength predictive models in the codes of practice.

This study aims to develop an electrically heated sidewalk system to efficiently clear ice and snow from pedestrian walkways, ensuring safety and minimizing environmental impact during winter. While extensive research has been conducted on heated pavement and slab systems for vehicles and aircraft, there is a notable gap in studies focusing on heated sidewalk systems for pedestrians, which are unique due to their disjointed configurations. Concrete mixtures containing 2.2% carbon fibers (CFs) and 0.5% carbon nanotubes (CNTs) by cement weight were used to cast the electrically heated concrete side blocks. No. 3 structural rebars served as electrodes to physically connect the side blocks and distribute electrical power to them. A laboratory-scale prototype, consisting of a 5 × 5 block array (750 mm × 750 mm), was constructed to evaluate heating performance. The surface and internal temperatures were measured using an infrared camera and thermocouples, respectively, while consistently powering the prototype with a power supply. The blocks were connected in a parallel electrical configuration to operate the system at low voltage levels. The results indicated a surface temperature increase of 16–20 °C over two hours with a power density ranging from 620 to 830 W/m², which was sufficient for deicing pedestrian walkways.

In thick continuously reinforced concrete pavement (CRCP), thicknesses greater than 356 mm, the Texas Department of Transportation (TxDOT) places longitudinal steel in two mats, with equal amount of steel in both top and bottom mats. These longitudinal steel layouts may not be optimal for restraining concrete volume changes from temperature and moisture variations (environmental loading), which are most severe near the surface. This study mechanistically evaluated and optimized two-mat reinforcement layouts by placing more steel in the top layer while reducing the amount in the bottom layer. A total of 2160 three-dimensional finite-element models were developed in ANSYS to examine how variations in steel depth and spacing influence steel stress, concrete stress, and crack width under different environmental loading conditions. The results indicated that top-focused configurations with reduced bottom reinforcement satisfied all performance thresholds for steel and concrete stresses as well as crack width, reducing concrete stress by about 4% and 11% for 356 mm slabs and about 13% for 381 mm slabs, and reducing crack width by about 2% for 356 mm slabs and about 5% for 381 mm slabs. These optimized designs achieved performance better than the existing TxDOT two-mat design standards while requiring less steel amount. The findings in this study will be validated in the field experiment in the second phase of the study.

This study presents the field performance of two emerging novel concrete technologies: (1) carbon nanotube (CNT)-based cementitious strain sensors for in situ sensing and monitoring, and (2) a bio-based phase change material (PCM) cover designed to reduce early-age temperature gradients and strain amplitudes associated with cracking potential in concrete pavements. Both technologies were implemented simultaneously in a newly constructed concrete slab section at the Kyle Fire Station in Kyle, Texas. The CNT-based cementitious strain sensors demonstrated reliable performance, with resistance measurements strongly correlating with temperature and strain ( R ² > 0.8), confirming their applicability under field conditions. The PCM study examined the potential of a novel PCM cover method to reduce early-age thermal cracking potential. Although winter ambient temperatures did not reach the 29 °C melting point required for full phase change activation of PCM, the external cover still reduced surface heat gain, absorbing solar radiation and lowering peak temperatures by approximately 1.5 °C during the covered period. Correspondingly, the PCM section experienced a maximum 30% less peak thermal strain in one cycle and reduced moisture-induced expansion compared to the control section during the PCM-covered period. The reduced temperature gradients and strain amplitudes suggest that the PCM cover system may reduce early-age cracking potential. Because the PCM was applied externally and not incorporated into the concrete mixture, it is not expected to directly affect mixture-level mechanical properties. Overall, the results provide a comprehensive assessment of CNT-based strain sensors and PCM covers under field conditions, highlighting their potential for practical implementation in future pavement construction projects.

Alcohol-impaired driving remains one of the leading contributors to traffic fatalities and severe injuries in the United States, with crash risk escalating nonlinearly with increasing impairment and compounded by the growing prevalence of polydrug use. Despite this, driving while intoxicated (DWI) drivers are predominantly treated as a homogeneous population in crash analyses, obscuring critical differences in behavioral subgroups and their interaction with crash pathways. To address this limitation, this study analyzes 49,346 police-reported impaired driving crash records from the Texas Crash Record Information System (2017–2025) using a three-step hybrid analytical framework that explicitly combines Cluster Correspondence Analysis (CCA) to identify latent behavioral profiles, semi-data-driven Bayesian Network (BN) structure learning to infer directed associational pathways linking pre-crash conditions to injury outcomes, and cluster-wise pathway heterogeneity analysis to quantify variation in injury risk across profiles. Four distinct latent behavioral profiles were identified among impaired road users, and five directed associational pathways linking pre-crash conditions to injury outcomes were identified, with harmful event type serving as the central mediating node across most pathways. Cluster-wise heterogeneity analysis revealed that injury risk was substantially associated with behavioral profile membership and Blood Alcohol Concentration (BAC) classification, with the highest-severity profile consistently associated with the greatest relative risk across all pathways (highest RR = 4.14). Notably, a directional reversal under geometric conflict conditions indicated that crash environment rather than BAC level alone was more strongly associated with injury risk within the highest-severity profile (Non-Aggravated DWI RR = 3.46 (95% credible interval: 3.01–3.81) vs. Aggravated DWI RR = 2.88 (95% credible interval: 2.47–3.22)), suggesting that the legal BAC threshold alone may be insufficient to capture the full spectrum of high-risk impaired driving behavior. These findings provide a behavioral evidence base for designing targeted, profile-specific intervention strategies consistent with Safe System and Vision Zero principles.