Infrastructure Development

This study explores how Artificial Intelligence (AI) can support green building construction by reducing carbon emissions and utility costs. AI techniques, including Random Forest (RF) and Genetic Algorithms (GA), analyze site data and optimize decisions affecting quality, safety, and sustainability. Statistical tools like MAE, RMSE, and RMSLE evaluate model accuracy. Results showed over 96% prediction accuracy, with GA models achieving an R² of 0.96 for CO₂ prediction. Cross-validation was used to ensure reliability and prevent overfitting, demonstrating AI’s strong potential in sustainable construction.

Introduces an innovative machine vision-based safety monitoring and warning system for civil engineering construction sites. Designed for all-weather, real-time use, it enhances image clarity in both daytime (using a new de-hazing algorithm with guided filtering) and nighttime (via infrared imaging and noise-reduction filters). A key feature is its image anomaly detection framework, which identifies dangerous behaviors through posture rotation and color deviation recognition. The system significantly improves safety by automatically detecting and warning of hazards, with experiments confirming the effectiveness of its recognition algorithms in low-visibility and low-light environments.

This paper examines the complex challenges of road design and safety in the Middle East and North Africa (MENA) region, shaped by diverse geography, rapid urbanization, and extreme climates. Despite major infrastructure investments, road safety remains a concern due to poor design, inadequate safety measures, and risky driving behaviors. The study explores the impact of climate, technology, and urban growth on transportation systems and highlights the importance of governance, public awareness, and smart infrastructure. Through case studies and best practices, it offers strategies to enhance road networks and safety across both urban centers and remote areas in the MENA region.

This study developed a hybrid decision-making model to assess the feasibility of financing tourism infrastructure in the V4 countries, based on predicted tourist movement, infrastructure accessibility, and expert evaluations of service quality and development potential. It introduced a novel five-layer neuro-fuzzy model to provide both quantitative and qualitative assessments, using expert insights on tourism growth prospects. The model was tested and validated using real data from 43 regions, offering a new approach to evaluating tourism infrastructure investment potential across varying regional conditions.

Examines the overlooked issue of substandard housing in U.S.-Mexico border communities known as *colonias*, home to over 500,000 people. Since 1991, the HUD Community Development Block Grant (CDBG) program has required set-aside funds to improve living conditions in colonias across Arizona, California, New Mexico, and Texas. Analyzing data from 2014–2023 and expert interviews, the study highlights how funds are primarily used for infrastructure and water/sewer projects, though usage varies by state. It concludes that while CDBG funding is vital, program improvements—such as expanding eligibility and strengthening local capacity—are needed to better meet colonia community needs.