Impact analysis of road infrastructure parameters on bus speed and transit regularity

Abstract

This paper performs a quantitative assessment of how road infrastructure conditions and parameters impact the speed and regularity of urban bus transit. The relevance of this topic stems from the fact that decisions regarding repairs and traffic management are often made based on general engineering criteria, whereas the actual effect on public transport (changes in travel time, headway variability, and punctuality) is rarely evaluated in a formalized manner. The information base consists of combined GPS/AVL data (trajectories, speeds, segment travel times, and stops) and the results of field infrastructure surveys (pavement quality, IRI/PCI metrics, stop and node parameters, the presence of dedicated lanes, and conflict points). An approach to route segmentation and spatio-temporal alignment of data sources is proposed. This approach enables the creation of a unified observation table using the "segment–time" key and the calculation of target KPIs: operational speed, travel time, regularity (headway/time variability), and delay metrics. Based on this framework, a statistical model is designed to interpretably link infrastructure factors with quality indicators, allowing for a scenario-based assessment of the effects of repairs or management interventions (e.g., IRI reduction, bus bay construction, or adjustments to traffic signal timings). The practical outcome of this study lies in formulating recommendations for prioritizing interventions on sections that contribute the most to time losses and transit irregularity.