Pemanfaatan Artificial Intelligence Dalam Mengukur Kinerja Ruas Jalan Margonda Raya Kota Depok

Suardika, Bayu; Malkhamah, Siti; Rizka Fahmi Amrozi, Mukhammad · 2025 · Crossref

DOI: 10.46447/ktj.v12i1.665

archive: archived pipeline: cataloged verified

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

Summary

This study addresses the critical issue of traffic congestion on Jalan Margonda Raya, a primary collector road in Depok City, Indonesia. Rapid population growth and increasing vehicle ownership have led to significant traffic instability, particularly during peak hours. The research aims to analyze the road’s traffic performance using the Indonesian Highway Capacity Guidelines (PKJI, 2023) while evaluating the efficacy of Artificial Intelligence (AI) in traffic surveys. The motivation stems from the need to improve data accuracy and minimize the human bias inherent in conventional manual observation methods, thereby supporting more reliable transportation planning decisions. The methodology involved a comprehensive traffic survey conducted on a weekday, December 11, 2024, from 05:30 to 19:00 WIB. Traffic volume data were collected using CCTV recordings processed by DataFromSky, an AI-based software utilizing computer vision and machine learning to automatically detect, track, and classify vehicles. This approach was chosen to handle high-volume traffic efficiently and objectively. Spot speed measurements were conducted using a Radar Gun during morning, midday, and afternoon periods to capture varying traffic conditions. The collected data were analyzed to determine road capacity, average speed, degree of saturation (DS), and Level of Service (LOS). The results indicate that Jalan Margonda Raya operates under severe congestion during peak hours. The highest traffic volume reached 4,161 passenger car units (pcu)/hour, with a total daily count of 10,465 pcu/hour. The vehicle composition was dominated by motorcycles (78%), followed by passenger cars (20%), heavy vehicles (1.47%), and buses (0.23%). The degree of saturation peaked at 0.90 during the morning rush (06:45–07:45), corresponding to a Level of Service (LOS) E, which signifies unstable traffic conditions approaching maximum capacity. Average speeds during this period were low, recorded at approximately 31 km/h. In contrast, midday and afternoon periods showed improved performance with LOS D and higher average speeds ranging from 32 to 39 km/h. The study confirms that AI implementation provided efficient, reliable, and objective data collection compared to manual methods. The significance of this research lies in demonstrating the practical utility of AI in modernizing traffic data collection. By reducing human error and bias, AI tools like DataFromSky offer a robust alternative for real-time traffic monitoring and performance assessment. The findings highlight that Jalan Margonda Raya faces structural congestion issues similar to other major urban roads in Indonesia, necessitating targeted interventions. The study concludes that integrating AI into transportation planning can enhance the precision of traffic analysis, providing a strong foundation for responsive policy-making and the development of smarter, more efficient 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.

• traffic density