TECHNOLOGY FOR CREATING A DIGITAL MODEL OF A SECTION OF A ROAD NETWORK TO IDENTIFY THE LOCATIONS OF A PROBABLE ACCIDENT USING THE QGIS GEOGRAPHIC INFORMATION SYSTEM

Lozovoy, Nikolai; Glagolev, Sergey; Shchetinin, Nikolai; Zagorodnii, Nikolai; Borovskoy, Alexey; Sokorev, Sergey · 2022 · Crossref

DOI: 10.34220/2311-8873-2022-91-101

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the increasing frequency of road traffic accidents (RTAs) in urban areas, driven by the expansion of road networks and rising vehicle density. The authors argue that traditional statistical analyses of the "driver-vehicle-road-environment" system are insufficient for identifying all contributing factors. To mitigate social and economic risks associated with RTAs, the study proposes a methodology for creating a digital model of road network sections using the open-source Geographic Information System (QGIS). This approach aims to identify locations with a high probability of accidents, enabling proactive infrastructure reorganization before incidents occur. The methodology involves integrating various data sources within QGIS to construct a comprehensive digital model. Road network data was obtained from specialized providers, specifically using road graphs from NEXTGIS, which include graphical data for roads, sidewalks, and pedestrian crossings, along with attribute information. A three-dimensional model was created by projecting the road graph onto a digital terrain model. Accident data, specifically locations of fatal RTAs in the Belgorod Region from 2015 to 2020, was overlaid onto the map. The authors performed data correction to ensure geodetic accuracy, removing points located outside the region or country. The experimental design involved a multi-stage filtering process to isolate high-risk road segments. Initially, 82,804 road objects were filtered to retain 28,968 public roads. The road graph was then split at nodes to identify straight segments, and further filtered for lengths between 600 and 1,700 meters, resulting in 90,677 segments. Subsequent filtering targeted two-lane asphalt roads, yielding 7,122 segments. To assess pedestrian risk, a buffer analysis was applied to identify segments near pedestrian crossings, highlighting 208 critical areas. Similarly, for four-lane roads, 78 straight segments of the specified length were identified, with buffer analysis revealing 24 unregulated pedestrian crossings in proximity. The study concludes that QGIS provides an effective tool for analyzing large datasets to pinpoint specific road sections prone to accidents. By identifying these high-risk zones, authorities can implement timely corrections and reconstructions to enhance road safety. The authors posit that this methodology serves as a foundation for creating a "digital twin" of the road network, allowing for the monitoring of current safety conditions and the forecasting of probable accident locations. This approach supports the reduction of social risks by enabling preventive measures rather than reactive responses to traffic incidents.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• incidence prevalence