Conference Agenda

Inland waterway transport on the Rhine corridor forms a critical component of European freight logistics, moving high volumes of bulk cargo under increasingly variable operating conditions. Traffic along the corridor is frequently disrupted by hydrological extremes, infrastructure failures, meteorological events, and operational incidents. However, the relationship between disruption frequency and actual traffic impact remains insufficiently researched. This gap is significant, as infrastructure investment and resilience planning are often based on event-frequency statistics, implicitly assuming that the most frequent disruptions also cause the greatest impacts.Recent studies have examined this issue using Notices to Skippers data from the EURIS portal and negative binomial regression to estimate the effects of short- and long-term events on daily vessel passages at lock systems along the Rhine, Main, and Danube. Related AIS data-based research has addressed vessel emissions, typhoon-induced network resilience, and Rhine water-level forecasting. However, these studies do not integrate disruption taxonomies with corridor-wide AIS-derived traffic measurements.Existing analyses have two main limitations: they assess impacts only at locks rather than across the full navigable corridor, overlook threshold effects and interactions between concurrent disruptions. Consequently, the relative contribution of each disruption category to traffic impacts on the Rhine remains unclear. Hence, a disruption attribution framework that links classified disruption events with corridor-wide traffic performance indicators is proposed. Disruptions are grouped into hydrological, meteorological, infrastructural, and operational categories and represented through time- and location-specific features. Traffic conditions are measured continuously along river segments using AIS and Notice to Skippers data, enabling the estimation of speed deviations and accumulated delays. A data-driven modelling approach is used subsequently to estimate segment-level traffic impacts and attribute these impacts to individual disruption categories. This enables a transparent assessment of how different disruption types contribute to corridor-wide traffic delays. The final results quantify the contribution of different disruption categories to traffic delays and reveal that disruption frequency does not necessarily correspond to traffic impact, suggesting that frequency-based monitoring may misrepresent risk priorities. The resulting category-level attribution weights may provide waterway authorities, port operators, and policymakers with a transparent, data-driven basis for prioritizing infrastructure investment, maintenance scheduling, and targeted resilience measures.