Data-Driven Decision Support for Urban Transport and Logistics Networks in Large Agglomerations

Abstract

Growing urbanization and increasing traffic intensity create a need for transparent tools that support urban transport management and city logistics planning. The aim of this study is to propose an interpretable rule-based decision-support framework that transforms short-term traffic-count data into relative traffic-load categories for preliminary traffic assessment and planning. A pilot case study was carried out using minute-level traffic count data from four selected locations in the Prague metropolitan area: Stodůlky, Barrandov, Radlice, and Velká Chuchle. The data were aggregated into hourly intervals and subsequently evaluated according to predefined daily traffic periods. To assess relative traffic intensity within each monitored location, the Relative Traffic Intensity Index was calculated using a percentile-based reference value. Based on predefined analytical thresholds, each daily period was classified into one of four relative traffic-load categories: low, medium, high, or critical. Within the Barrandov dataset, both the morning and afternoon peak periods showed high relative traffic intensity and were classified as critical under the primary reference setting. Sensitivity analysis using alternative percentile references showed that these peak-period classifications remained stable at Barrandov, whereas some classifications, particularly at Velká Chuchle, were more sensitive to the selected reference value. The proposed framework provides a transparent and interpretable approach for converting traffic count data into operationally understandable information for preliminary traffic assessment and planning. Due to the limited temporal and spatial scope of the available data and the absence of independent congestion indicators, the results should be interpreted as a pilot demonstration of the proposed analytical procedure rather than as an externally validated or fully generalized predictive traffic management system.