<p>Physics-Informed Neural Networks (PINNs) have emerged as a transformative modeling approach that merges data-driven learning with physical laws, offering a compelling solution to longstanding challenges in civil and transportation engineering. This review provides a comprehensive examination of PINN applications across diverse domains, including pavement analysis, structural health monitoring, geotechnical modeling, material characterization, and transportation system dynamics. A systematic literature review was conducted following Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines, covering 100 peer-reviewed articles published up to mid-2025 and sourced from major scientific databases. The studies were categorized into thematic clusters, highlighting advancements in modeling surface distresses, structural performance, and dynamic vehicle behavior, as well as applications in bridge engineering, tunnel design, and soil-structure interaction. The review analyzes core architectural developments, optimization strategies, and hybrid PINN frameworks tailored for engineering contexts. Despite these advances, the field faces significant limitations, including inconsistent evaluation practices, limited model generalization across domains, heavy dependence on synthetic or simplified datasets, and the absence of rigorous uncertainty quantification frameworks. These gaps hinder the benchmarking and scalability of PINN models in practical engineering environments. Addressing these issues is essential for enabling reproducible, trustworthy, and field-applicable PINN-based solutions. Emerging trends such as integration with Digital Twins and IoT-enabled infrastructures signal a promising direction for real-time, physics-consistent decision-making in civil systems. This work offers valuable insights for researchers and practitioners aiming to harness the power of PINNs in advancing smart, resilient, and sustainable infrastructure.</p>