PSI - Issue 84

Galileo Tamasi et al. / Procedia Structural Integrity 84 (2026) 725–732

728

2.2. Change Management and Data-Driven Operational Windows

The identification of optimal time windows for executing works is logically embedded in the Change Management process, which requires an exhaustive evaluation of how the operational change interacts with the environment. This methodology emphasizes the importance of assembling both qualitative and quantitative data to support risk-based asset management decisions (Dehghani et al., 2025). By identifying optimal time windows through historical analyses of traffic volumes and weather conditions, operators can minimize risk exposure for both workers and users (Ouyang, 2021). Such proactive planning is vital, considering that work zone crashes, particularly those involving large trucks and detours, are empirically associated with higher levels of injury severity and total harm (Khattak and Targa, 2004). 2.3. High-Fidelity simulation for institutional and safety validation The final stage of this integrated approach involves the use of high-fidelity microscopic traffic simulations to validate the proposed mobility plans. These simulations provide the quantitative evidence required by safety management protocols to prove that the proposed "alternative scenario" is viable and safe. Furthermore, data-driven asset management allows for prioritizing repair and replacement decisions based on probability of failure and user impacts (Chae et al., 2024). This predictive capability transforms simulation data into a cornerstone of institutional transparency. From an organizational safety perspective, this structured methodology prevents independent decisions from combining to create unforeseen negative impacts on the system (Marais et al., 2006). This approach reinforces the role of the highway operator as a proactive actor that manages network evolution through a robust, risk-based methodology. 3. Dynamic safety mitigation: V2I communication and advanced separation analysis While the reallocation of light-vehicle traffic to alternative networks provides a necessary capacity buffer, it introduces a significant shift in the operational risk profile. Much like the separation of general aviation from commercial traffic in the aerospace sector, rerouting non-professional drivers onto national and provincial roads creates a "skill gap" challenge. Unlike professional truck drivers, car commuters often possess lower driving skills and are less accustomed to the heterogeneous geometries and unforeseen obstacles of secondary roads. Consequently, the risk of accidents linked to the transition between road types is non-negligible (Li and Bai, 2009). To address this, a second mitigation strategy is proposed, founded on two pillars: the continuous analysis of vehicle separation and the deployment of Vehicle-to-Infrastructure (V2I) communication systems. 3.1. V2I systems and precursor event detection This strategy represents a paradigm shift from traditional "crash mitigation factors" - which rely on historical collision data - toward a proactive approach based on precursor events. By leveraging Advanced Driver Assistance Systems (ADAS) and V2I connectivity, the infrastructure can detect and warn against near-miss scenarios before a collision occurs. The focus shifts to identifying critical reductions in separation distances from obstacles and other vehicles, as well as deviations from the ideal trajectory. This real-time monitoring is essential for operationalizing safety in work zones where lane configurations are frequently altered (Porretto et al., 2025). Within a robust Safety Management System, these precursor events serve as the primary indicators for organizational safety, allowing for the correction of systemic drifts before they escalate into accidents (Marais et al., 2006). By analyzing these precursors, highway agencies can move beyond reactive management, fostering a culture of continuous learning and proactive risk reduction (National Academy of Engineering, 2004). 3.2. Quantitative risk modeling and the three-tier Assessment framework The core of this strategy lies in the continuous collection of high-fidelity data, which allows for the constant updating of occurrence models. By employing logistic regression techniques, "missed separation" events can be correlated with a comprehensive set of parameters, including environmental factors such as precipitation (rain, snow,