Pedestrian routing is important in a multitude of public spaces, especially those characterized by a large number of newcomers. Their needs may be diverse, with priority for the shortest path, the less crowded or the less polluted one, the accessibility for reduced mobility, or the sheltering from unfavorable weather conditions. Hence, typical graph-based routing must be enriched to support multiple policies, at the choice of each person. The paper proposes a systemic approach and a set of services for orientation and accessibility, which are both community-driven and data-driven, for correctly perceiving the routing necessities and the surrounding situation. The response time to a pathfinding query depends on the types of policies applied and not only on their number, because each of them contributes to the customization of the weighted graph, although it refers to the same physical space traversed by pedestrians. The paper also presents results of loading tests for up to 5000 Virtual Users, inspired from real-life requirements and executed on a graph that models a real building in our university; different policies are applied to assess performance metrics, with simulated community feedback and sensor data.

This paper investigates the integration of intrusion detection systems (IDSs) within Digital Twin (DT) architectures to enhance cybersecurity in industrial environments. Using the CICIDS2017, CIC Modbus, and 4SICS 2015 datasets, we evaluate the performance of Random Forest (RF) and Support Vector Machine (SVM) in detecting network intrusions. Results indicate that RF achieves an accuracy of 99.9% for CICIDS2017, with high precision, recall, and low false positives. In contrast, SVM exhibits an accuracy of 94.2% for the same dataset, struggling with high rates of false positives and moderate recall. Similarly, for 4SICS 2015, RF demonstrates an accuracy of 93%, being balanced and reliable for industrial applications, while SVM shows only 88% accuracy, with a low precision of 65% and a high false alarm rate. For the CIC Modbus dataset, RF displays an accuracy of 95% in validation and 93% in testing, highlighting strong detection in ICS networks. However, SVM maintains an accuracy of 88%, with weak separation between benign and malicious traffic, and a higher misclassification rate. Our findings highlight the importance of DT-IDS integration in real-time threat detection and system resilience, paving the way for future research in deep learning-based IDS solutions.

As global temperatures rise, cities struggle with heat stress and the limitations of traditional cooling strategies. This study introduces “seasonal urban exodus”—temporarily relocating urban residents to cooler areas during summer—as a behavioral climate adaptation strategy driven by the need for thermal comfort. To assess social feasibility, a survey was conducted among 163 urban residents in Romania. The dataset was analyzed using linear regression and machine learning algorithms (Random Forest and K-Means clustering). The results show that 77.9% of respondents would relocate for 1–2 months if they had adequate destination infrastructure, while a 2 °C temperature increase would cause 46% to migrate temporarily. Predictive modeling identified barriers related to heat (p = 0.009) and transportation (p = 0.016) as the most significant predictors of relocation intention. These results suggest that seasonal mobility is a viable social response to urban heat islands. However, while this adaptation strategy improves individual thermal comfort, further interdisciplinary research—including life-cycle assessments, travel emission calculations, and the evaluation of rural energy systems—is absolutely necessary to determine the net carbon balance and environmental viability of these relocation patterns.