Abstract
In 2022, Ukraine is suffering an invasion which has resulted in acute impacts playing out over time and geography. This paper examines the impact of the ongoing disruption on traffic behavior using analytics as well as zonal-based network models. The methodology is a data-driven approach that utilizes obtained travel-time conditions within an evolutionary algorithm framework which infers origin-destination demand values in an automated process based on traffic assignment. Because of the automation of the implementation, numerous daily models can be approximated for multiple cities. The novelty of this paper versus the previously published core methodology includes an analysis to ensure the obtained data is appropriate since some data sources were disabled due to the ongoing disruption. Further, novelty includes a direct linkage of the analysis to the timeline of disruptions to examine the interaction in a new way. Finally, specific network metrics are identified which are particularly suited for conceptualizing the impact of conflict disruptions on traffic network conditions. The ultimate aim is to establish processes, concepts and analysis to advance the broader activity of rapidly quantifying the traffic impacts of conflict scenarios.
Introduction: Mobility in Conflict Zones
Traditional urban planning relies heavily on physical infrastructure—loop detectors, traffic cameras, and household surveys—to understand and manage city-wide movement. However, in a conflict zone, this infrastructure is often the first to fail, either through direct destruction or abandonment. Understanding traffic patterns during such times is not merely an academic exercise; it is critical for managing mass evacuations, ensuring the delivery of humanitarian aid, and assessing the real-time impact of infrastructure damage on civilian life.
This research addresses the "data poverty" inherent in conflict zones by exploring the utility of pervasive, crowd-sourced traffic data. By leveraging navigation provider APIs, the study aims to provide a reliable "pulse" of urban mobility even when ground conditions are chaotic and dangerous.
Methodology and Data Integration
The researchers employed an Automated/Rapid Transport Planning framework. Unlike traditional models that start with population data to estimate demand, this study uses an "inverse" methodology to back-calculate travel patterns from observed speeds.
Data Extraction and Processing
The study utilized two primary data sources:
- TomTom API Data: Real-time travel times and speeds were extracted for six major cities—Kyiv, Kharkiv, Lviv, Dnipro, Odesa, and Mariupol—as well as the national highway network.
- OpenStreetMap (OSM): This provided the underlying road network geometry for traffic assignment.
The data extraction process focused on identifying unique travel time values to ensure the robustness of the pervasive data stream, especially as some "live" features were occasionally disabled for security reasons during the invasion.
Metric Calculation and OD Estimation
Two key indicators were used to quantify network conditions:
- Coefficient of Variance (CoV): Used to measure travel time uncertainty, reflecting how much travel times fluctuated day-to-day.
- Congestion Index: The ratio of actual travel time to free-flow travel time.
To estimate Origin-Destination (OD) matrices, a customized genetic algorithm-based metaheuristic was applied. This evolutionary algorithm iteratively adjusted demand values until the modeled traffic speeds matched the observed TomTom data, allowing for the automated generation of daily mobility models.
Timeline of Disruptions
A crucial component of the methodology was the synthesis of a timeline of 51 key war events, including airstrikes, curfews, and the establishment of "green corridors." This timeline allowed researchers to directly correlate military actions with shifts in traffic behavior.
Key Insights and Results
The analysis revealed several distinct patterns in how the conflict reshaped mobility:
- The "Inner vs. Outer" Paradox: In most cities, inner-city regions showed lower travel time variability than the outskirts. While city centers were frequently targeted, traffic there remained "consistently" congested. Conversely, outer regions experienced wild fluctuations, likely driven by irregular migration and the sudden appearance of checkpoints.
- Lviv as a Humanitarian Sink: Lviv’s traffic patterns uniquely reflected its status as a destination for millions of internally displaced persons, acting as a "sink" for those fleeing conflict zones further east.
- Quantifiable Impact of Siege and Evacuation: In Mariupol, the opening of evacuation corridors triggered a massive spike in travel time variability. The model estimated a 300% increase in trip productions during these intense periods.
- Congestion and Uncertainty: Invasion events led to congestion increases of up to 50% in Kyiv and 55% in Kharkiv. Travel time uncertainty (CoV) jumped by 135% in Dnipro and 100% in Kharkiv during the study period.
- Recovery Trends: Following the de-occupation of the Kyiv region on April 2, 2022, the data captured a clear "return to normalcy," with morning congestion levels steadily decreasing as the immediate military threat subsided.
Limitations and Future Directions
A primary limitation of the current framework is its assumption of a static network model. The model assumes road capacity remains constant, whereas in reality, bridges are destroyed and roads are rendered impassable. Currently, if a major artery is lost, the model may still attempt to assign traffic to it unless manually updated.
Future research will focus on a "destructive variant" of the network design problem. This involves using evolutionary algorithms to simultaneously estimate both travel demand and infrastructure integrity, inferring road closures based on the sudden, unexplained disappearance of traffic from specific links. This advancement would further enhance the ability to rapidly quantify the impacts of conflict on urban infrastructure.
Summary updated