AN INTELLIGENT LOCATION AND STATE REORGANIZATION OF TRAFFIC SIGNAL

Behzadi, Saeed · 2020 · Crossref

DOI: 10.3846/gac.2020.10806

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 inefficiency of storing traffic signal data as separate entities in geospatial databases, proposing instead an intelligent method to derive signal location and state directly from road network attributes. The motivation stems from the reality that traffic signals exist based on road conditions and parameters; therefore, their spatial and non-spatial information should be extractable from road data rather than maintained as independent records. The study aims to reduce database redundancy and improve the integration of traffic information within Geospatial Information Systems (GIS). The methodology involves two main components: extracting spatial information and determining non-spatial states. For spatial identification, the algorithm analyzes road coordinates to detect intersections where three or more road endpoints converge. It assigns a unique integer ID to roads sharing an intersection and uses decimal encoding to specify the signal’s face relative to the road direction. For non-spatial information, the method utilizes Annual Average Daily Traffic (AADT) data from connected roads. It calculates the mean and standard deviation of AADT values for opposing road directions at each intersection. Based on these statistics and predefined thresholds ($T'$, $T_1$, $T_2$), the algorithm applies specific rules to determine if a signal is needed and to classify its state (e.g., "Amber-Red" blinker or "Green-Amber-Red" steady signal) and timing. The proposed method was implemented on a study area comprising approximately 180 roads and compared against existing database records of real traffic signals. The results demonstrated high accuracy: the overall accuracy for recognizing traffic signal locations was 94%, and the accuracy for recognizing signal states was 89%. Specifically, the model correctly identified 42 signal locations and 42 signal states. Discrepancies were attributed to the manual, experience-based nature of real-world signal placement and the reliance on user-defined thresholds for the algorithm. The significance of this work lies in its potential to streamline GIS databases by eliminating the need for separate traffic signal data layers. The findings suggest that road network attributes are sufficient for inferring signal presence and configuration. The authors conclude that while the current method requires manual threshold definition, future improvements could involve using optimization algorithms, such as Genetic Algorithms, to automatically determine optimal threshold values. This approach offers a more rational, data-driven alternative to manual signal placement and enhances the efficiency of traffic data management in urban transportation systems.

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.

• signage environment