The field of transportation and environmental systems is rapidly advancing with the integration of artificial intelligence (AI) and machine learning techniques. Recent studies have focused on developing innovative models to predict traffic flow, traffic volume, and carbon emissions, as well as optimizing transportation systems and managing uncertainties in open-pit mines. A common theme among these studies is the use of hybrid models that combine traditional approaches with AI-driven techniques to improve prediction accuracy and robustness. For instance, the use of graph neural networks, transformers, and attention mechanisms has been explored to capture complex spatial and temporal dependencies in traffic and environmental data. Notably, the development of probabilistic models, such as the Pretrained Probabilistic Transformer, has enabled the estimation of uncertainty in traffic predictions, which is crucial for decision-making in intelligent transportation systems. Furthermore, the integration of multiple data sources, including satellite images, point-of-interest data, and real-time traffic data, has been shown to improve the accuracy of predictions and enable more informed decision-making. Noteworthy papers include: TrafficPPT, which introduces a Pretrained Probabilistic Transformer for city-scale traffic volume prediction, demonstrating significant improvements in accuracy and uncertainty estimation. OpenCarbon, which proposes a contrastive learning-based approach for high-resolution carbon emission prediction using open data, achieving a 26.6% performance gain on R2. ADFormer, which presents an Aggregation Differential Transformer for passenger demand forecasting, capturing holistic spatio-temporal relations and demonstrating effectiveness and efficiency in experiments.