Dealing with maintenance disruption management in highway network and infrastructure

Disruptions in transportation networks are causing delays and driving up costs for daily commutes and goods delivery. Particularly, infrastructure degradation poses significant risks, with aging highways, bridges, and pavements prone to interruptions that intensify congestion and delays. To mitigate these vulnerabilities, this study addresses the need for effective maintenance approaches that strengthen highway reliability and minimize its traffic impacts across the network. The focus of this research is on highway maintenance, a fundamental component in ensuring the reliability of transportation networks, though it presents planning challenges. Effective maintenance must balance technical requirements with minimal disruption to traffic flow. This study introduces a framework that optimizes maintenance scheduling by integrating traffic simulation models with a Mixed-Integer Linear Programming (MILP) approach. This integration prioritizes planned maintenance interventions based on infrastructure condition and traffic impacts. It enables an efficient scheduling method that aligns maintenance needs with real-world traffic demands. In developing the proposed framework, the research employs a mixed-methods approach, beginning with traffic simulation to replicate real-world conditions and assess the impact of various planned maintenance scenarios on traffic congestion. These insights are then applied to a MILP optimization model, which prioritizes and schedules maintenance interventions to reduce traffic disruption. Also, the framework includes a specific model focused on bridge maintenance, factoring in rerouting strategies and bridge degradation considerations. To manage the computational demands of large-scale highway networks, the study introduces a solution algorithm to enhance optimization efficiency. This algorithm incorporates a data merging process to standardize segment lengths for manageable work zones. Also, it uses k-means clustering, the algorithm groups work zones to determine optimal clustering. Additional data enrichment steps include incorporating traffic impact and degradation levels, organizing clusters into a sequenced yearly maintenance plan that combines traffic volume with infrastructure needs. A case study on a 60-kilometer Italian highway network validates the effectiveness of the proposed approach. This case study, informed by two years of traffic and maintenance data, enables realistic scenario modeling for lane closures and traffic diversions., The results section evaluates the impacts of proposed framework on key performance indicators, such as queue length, delay, and traffic volume, demonstrating the ability of the model to reduce disruptions in high-traffic networks. Clustering performance is validated through Silhouette scores and interval analyses, aligning clusters with degradation levels. Gantt charts visualize maintenance schedules, indicating how each scenario improve the performance. This structured approach allows for an adaptable plan that reduces traffic disruptions while addressing infrastructure degradation.