Smart Resilient Infrastructure & Intelligent Systems 

Earthquakes present a significant and unpredictable threat to infrastructure systems, with the potential to damage bridges, buildings, and essential lifeline networks. Even the failure of a single structural component can propagate into larger system collapses. This challenge can be addressed by strengthening the fundamental structural performance of infrastructure through improved design strategies, advanced materials, and intelligent monitoring technologies. The approach also emphasizes optimizing structural geometry, load paths, and redundancy to ensure that buildings and infrastructure can effectively redistribute forces during seismic events.  The focus of the SRIS Lab within this theme is the development of high-performance and smart structural materials that enhance strength, ductility, and durability under cyclic loading conditions. The lab also investigates innovative structural systems and design methodologies that improve overall seismic performance and structural reliability. The economic and social benefits of resilient structural design are substantial. Investments in earthquake-resistant infrastructure can significantly reduce long-term recovery costs while protecting lives and property. However, challenges remain, including the difficulty of predicting earthquakes, the high cost of retrofitting aging infrastructure, and the need to better integrate advanced research findings into practical engineering design codes and construction standards. Therefore, the SRIS Lab focuses on advancing resilient structural technologies, smart materials, and data-informed structural design approaches to improve seismic safety and infrastructure sustainability.

Cities and infrastructure systems are increasingly exposed to a wide spectrum of extreme hazards, including flooding, structural fires, hurricanes, heatwaves, and earthquakes, which may occur simultaneously or in sequence. Climate change is intensifying the frequency and severity of many of these events, increasing risks to critical infrastructure and urban communities. Research in this theme focuses on understanding and mitigating combined hazard impacts through integrated modeling, advanced structural design, and risk-informed engineering strategies. The research emphasizes the development of multi-hazard risk assessment frameworks and resilient infrastructure solutions. The economic and social benefits of resilient infrastructure investment are significant. Resilient infrastructure also enhances economy, public safety and supports long-term community sustainability. Despite these advantages, several challenges remain. Advanced multi-hazard modeling tools are often developed in research environments but are not widely implemented in engineering practice. Additionally, predicting complex and cascading disaster events remains uncertain, and existing building codes, policies, and funding mechanisms often lag behind scientific and technological advances. The research vision of the lab is therefore to advance practical, data-driven, and scalable resilience solutions for multi-hazard environments.

Modern computer vision and artificial intelligence techniques are transforming how infrastructure is assessed by automating and accelerating damage detection. High resolution images and drone video can be analyzed using deep learning models to identify cracks, corrosion, spalling, and other structural defects much faster and more objectively than manual inspections. Incorporating physical laws into these networks, known as physics informed learning, improves prediction accuracy and ensures physically consistent results. These intelligent systems can rapidly scan bridges, buildings, and dams and automatically flag regions of concern. The detected information can be integrated into digital twins and three dimensional models to support continuous structural monitoring, prioritize inspections, and enable predictive maintenance. The main advantages include faster assessments, more consistent evaluations, and earlier detection of potential failures. Challenges remain as this technology is still developing. Deep learning based monitoring requires large and well curated training datasets, careful model calibration, and robust validation to ensure reliable performance across different structure types, environmental conditions, and sensor platforms. Improving model generalization, uncertainty quantification, and long term field implementation continues to be an important research focus.

High-fidelity computational methods let engineers virtually test and optimize infrastructure with unprecedented detail. Techniques such as nonlinear finite element modeling, topology optimization, and performance-based simulation enable designers to predict how a bridge, building, or material will behave under extreme loads. For example, iterative performance-based earthquake and fire simulations help ensure structures meet explicit safety targets. When paired with modern scientific computing tools like parallel processing, GPU acceleration, and uncertainty quantification, these models can capture complex interactions among materials, geometry, and loading. The result is safer and more efficient designs: engineers can remove unnecessary material, explore novel structural forms, and identify hidden failure modes before construction begins. These tools directly support resilience by locating vulnerabilities early and optimizing structures to avoid them. However, high-fidelity models require substantial computational resources and high-quality input data. Validating simulations against real-world experiments is essential but often difficult, and the simplifying assumptions needed to make problems tractable can limit accuracy. In short, advanced computational mechanics opens powerful possibilities for safer infrastructure, but improving model reliability, validation, and scalability remains a central area of research.

Robotic systems are providing new capabilities for inspection and maintenance of infrastructure, particularly in hazardous or hard to access environments. Drones, climbing robots, and mobile ground vehicles can carry cameras, LiDAR, ultrasound, and other sensing technologies to survey bridges, tunnels, storage tanks, and tall structures. These autonomous platforms can reach areas that are difficult or unsafe for human inspectors. Multi scale robotic inspection approaches can combine computer vision and autonomous sensing to identify regions of interest such as cracks or surface defects, followed by detailed scanning to construct three dimensional damage models. These digital representations, often integrated into digital twin platforms, enable precise damage quantification and long term structural condition monitoring. Such technologies significantly reduce inspection time while improving measurement accuracy. The primary benefits include improved safety through earlier defect detection and increased operational efficiency by allowing maintenance efforts to focus on critical repairs. However, challenges remain in deploying robotic systems for infrastructure inspection. These include maintaining reliable autonomy in complex and changing environments, ensuring sufficient power supply for long duration missions, and effectively managing large volumes of sensor data for infrastructure management. Although current robotic inspection technologies show strong potential, further work is needed to make robotic infrastructure assessment a routine engineering practice.