ABSTRACT This paper presents a comprehensive literature review of recent advances in autonomous vehicle (AV) technologies, focusing on artificial intelligence, multi-sensor fusion, automated control, and safety assurance frameworks. Autonomous vehicles promise to transform mobility by integrating AI-driven perception and decision-making systems, with the potential to significantly reduce human-driver errors, which are responsible for over 90% of road crashes, and to improve transportation safety and accessibility. We review recent advances in AV perception, planning, and safety, with particular emphasis on industry safety standards such as ISO 26262 for functional safety. The review highlights that purely empirical validation is impractical, as prior studies estimate that billions of test miles would be required to achieve statistical confidence, motivating hybrid safety approaches that combine formal analysis (FTA/FMEA) with large-scale simulation and targeted on-road testing. Key enabling technologies, including sensor platforms, multi-modal fusion, AI-based planners, and validation workflows such as simulators, hardware-in-the-loop testing, and vehicle trials, are surveyed. Finally, emerging regulatory and deployment trends, including UNECE initiatives and public trust considerations, are discussed, along with open challenges such as robust handling of rare edge-case scenarios and the need for continuous, data-driven safety feedback loops. By integrating technical advances with safety assurance frameworks, ethical considerations, and regulatory developments, this review provides actionable insights for engineers and policymakers supporting safe and scalable AV deployment.

ABSTRACT Timber-Cardboard Sandwich (TCS) composites have emerged as sustainable, lightweight structural systems for rapid deployment in temporary housing, particularly in post-disaster contexts. While prior studies have addressed structural configuration and assembly methods, little is known about their long-term serviceability, particularly time-dependent deformation due to creep under natural environmental exposure. This study presents the first investigation into the short- and long-term creep performance of TCS beams under uncontrolled ambient weather conditions. Twelve full-scale TCS beams incorporating either recycled or scrap cardboard cores were subjected to sustained flexural loading over periods up to 9 months. Key variables included core material type, loading duration and exposure to hygrothermal conditions. Post-creep flexural testing was conducted to evaluate residual strength, stiffness and failure modes. Results show that beams with recycled cores exhibited more stable and consistent creep performance than those with scrap cores. Both variants met acceptable deflection criteria under short-term loading, supporting their viability for service conditions. While environmental exposure significantly influenced creep deformation, post-creep bending strength and failure modes remained consistent with control specimens, indicating no substantial degradation of structural performance. The findings establish a foundational dataset for evaluating the long-term behaviour of TCS composites and provide critical insights for their application in sustainable temporary construction.

ABSTRACT The paper presents the development of fragility curves for two types of frames considering reduced beam section (RBS) connections, based on the results obtained from extensive nonlinear time history analysis (NLTHA). Two different classes of MRF are considered with bolted and welded connections. Due to some limitations inherent in both experimental and empirical based fragility curves, analytical fragility curves are considered to be the most reliable ones for the assessment of seismic vulnerability. In the present work, first, the force-displacement behaviour of bidirectional bolted connections (ordinary and RBS) was obtained by experimental test and numerical simulation, while the force-displacement behaviour of welded connections (ordinary and RBS) was obtained by numerical simulation using finite element analysis in ABAQUS. Subsequently, moment–rotation (M-θ) curves for all the connections were extracted from the backbone curves to obtain the behaviour of the plastic hinges of each connection. These plastic hinges were then used to derive the accurate force-displacement behaviour of the medium-rise frame modelled in SAP 2000. Finally, NLTHA was performed to obtain the maximum roof drift, considered the engineering demand parameter (EDP) in this study. Finally, standard goodness-of-fit test was performed using standard statistical methods and probability distribution function was fit to maximum roof drift data. A comparison of the fragility functions for each type of frame with different RBS-column connections was carried out. The results indicate the significant low seismic vulnerability of the considered medium-rise frames with RBS for both the bolted and welded connections, in comparison to the ordinary frame.

ABSTRACT The primary objective of this study is to develop a machine learning model capable of predicting tunnel-induced settlement resulting from excavation processes. To achieve this, a set of finite element – based datasets was generated from rigorously calibrated and validated numerical models. These datasets were then employed to construct an ensemble learning model for settlement prediction. The input variables were selected to include key geotechnical and geometric factors, specifically the surrounding soil properties and tunnel configuration parameters, which collectively govern the deformation behaviour during excavation. The selected soil parameters comprise cohesion, internal friction angle, Poisson’s ratio, and the elastic modulus. The tunnel-related parameters include tunnel diameter, depth of cover, and excavation method. The maximum surface settlement above the tunnel was designated as the output variable. Two machine learning models – Gradient Boosting and XGBoost – were developed for this purpose. The advantage of employing these two algorithms lies in their complementary strengths, where their integration helps to mitigate the individual limitations of each model especially in the case of relatively small data-set. The findings confirm the effectiveness of the proposed approach in predicting excavation-induced tunnel settlement, demonstrating its practical utility for risk assessment and mitigation planning.

ABSTRACT This paper describes a new computational framework for characterising the effects of reclaimed asphalt pavement (RAP) parameters, sources and contents, on the structural performance of hot mix asphalt, in terms of dynamic modulus (E*), of using machine learning (ML) techniques. The ML models were developed for five RAP content percentages (0%, 10%, 20%, 30%, and 40%) for three different sources. The ML models showed promising results with Taylor diagrams and R2 values of 87% to 99% for all RAP sources and content levels. The results of pavement design using ML data showed the thicknesses of the asphalt mixture are slightly different depending on the RAP source. The same trend was found in the mixing and construction temperatures. Furthermore, analysis of sustainability showed that higher and stiffer RAP (higher E*) results in a more sustainable asphalt mixture because of lower thicknesses, and consequently lower aggregate requirements and CO2 emissions, for all ML methods and fuel types. Lastly, the incorporation of higher RAP content from any sources, the use of ML methods, and the adoption of cleaner fuel types results in massive material savings, higher energy efficiency, and improved sustainability in asphalt mix production for all RAP sources.

ABSTRACT To investigate the damage and failure mechanisms of external thermal insulation systems (ETIS) under horizontal seismic loads, this study conducted quasi-static tests to examine the hysteretic behaviour of ETIS on frame-structures. The research focused on analysing the influence of infill walls, insulation material types, and the bonding ratio of insulation boards on the seismic performance of ETIS on frame-structure. The experimental results demonstrated that in thin-coat ETIS, the ETIS on frame beams and columns were the first to fail under horizontal loading. Systems relying solely on adhesive bonding exhibited more severe seismic damage at lower bonding ratios. Ceramic tile claddings on these systems also detached during simulated seismic action. In contrast, at a 70% bonding ratio, delamination was confined to the corners of infill walls, while other areas remained intact. Rock wool board systems maintained complete integrity under all conditions, even with ceramic tile claddings. Furthermore, lower strength of infill wall blocks and insulation materials directly exacerbated seismic damage. For ceramic tile-clad systems, a 50% bonding ratio prevented tile detachment in phenolic/EPS-based systems, whereas rock wool systems exhibited no tile detachment regardless of bonding ratio.

ABSTRACT This study explores the combined influence of coarse steel slag (SS) and glass fibre (GF) on the durability properties of concrete. Among different SS, the current study utilises Linz-Donawitz slag (LDS) as a replacement for natural coarse aggregates (NCA) at varying percentages of 0%, 25%, 50%, 75%, and 100%. Additionally, GF is introduced in concrete mixes at concentrations of 0.15%, 0.3%, 0.45%, and 0.6%. Concrete performace was evaluated through tests on compressive strength, sulphate and acid resistance, chloride penetration, carbonation, freeze–thaw resistance, water permeability, electrical resistivity, and ultrasonic pulse velocity. The findings indicate that the integration of 50% of LDS into concrete resulted in an improvement in compressive strength, sulphate resistance, sulphuric acid resistance, hydrochloric acid resistance, chloride penetration resistance, carbonation resistance, water permeability, and ultrasonic pulse velocity by approximately 10.09%, 13.64%, 10.55%, 18.05%, 9.20%, 8.42%, 9.89%, and 4.54%, respectively. Incorporating 0.45% of GF into 50% LDS concrete further enhanced these properties by up to 27.15%, 20.45%, 19.42%, 33.25%, 14.94%, 25.05%, 25.17%, and 13.40%, respectively, owing to the combined pozzolanic and reinforcing effects of LDS and GF. The experimental findings demonstrated that the combined use of LDS with GF leads to superior performance in both strength and durability aspects.

ABSTRACT This study presents a mechanical model based on the component method (CM) as a reliable analytical alternative to experimental testing for assessing the strength and stiffness of boltless beam-to-column (BTC) connections in steel pallet racks (SPR). Developed within the Eurocode (EN 1993-1-8) framework, the model systematically analyses the M-θ behaviour of cold-formed steel (CFS) connections with slender geometry. By identifying critical components and using spring analogies, it quantifies the contribution of individual elements, offering insights into failure mechanisms and design optimisation. The model’s predictions were validated against experimental results for a 5-tab beam-end connector (BEC). The findings highlight the influence of column profile and contact area on strength and stiffness while indicating the need for refinement to capture column stiffening effects more accurately. Although experimental methods remain indispensable for specific configurations, the CM-based model provides a cost-effective and efficient tool for the connection analysis. By reducing reliance on costly experimental testing, this model contributes to the advancement of semi-rigid BTC connection design in steel pallet racks, offering a practical and scalable solution for engineers and designers.

ABSTRACT Extreme weather events intensify landslide risks, which can significantly impact the safety and functionality of road networks. Effective risk management in this context requires identifying vulnerable areas supported by appropriate assessment methodologies. This study evaluated landslide risks on state highways using historical data and future climate projections, considering the SSP1–2.6 and SSP5–8.5 scenarios. The most recent Brazilian climate risk assessment methodology was applied, with a focus on the state of Santa Catarina in southern Brazil. The analysis combined multiple local criteria within a geoprocessing environment, resulting in maps that classify road segments into five risk levels, ranging from very low to very high. Projections indicate that, without adaptation strategies, the extent of the state road network exposed to medium and high risk could nearly double over the coming decades. Additionally, the results were compared with the original outputs of the methodology, highlighting the importance of considering regional approaches, especially in highly vulnerable areas such as Santa Catarina. This study represents one of the first applications of this methodology outside its original national context, providing insights for its replication in other regions and contributing as a decision-support tool for planning road infrastructure adaptation to climate change.

ABSTRACT Accurate and reliable bridge asset information is crucial for the efficient management of road networks, however the scale of road networks makes collection and maintenance of this data a significant burden. While current studies have demonstrated the feasibility of the detection of bridge assets from aerial LiDAR, this research addresses the added complexities of network-scale application of such techniques, considering variability in structure types, environments, and data quality. A two-pronged LiDAR processing method was developed, employing geometric segmentation and point LAS classification. The methods were tested on public LiDAR datasets covering 6 km2 and showed average identification accuracies of 64% and 93%, with detection efficacy strongly influenced by on and under bridge use combinations. This paper also presents the first benchmarking of bridge measurement accuracy from aerial LiDAR, with median accuracies for bridge length and width of 90% and 66% as compared to ground truth data.