Implmenting FCD system in Belgrad

Nikolić Marko K.; Jović Jadranka J.; Vukanović Smiljan M. · 2015 · DOAJ

DOI: 10.5937/tehnika1505845N

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 investigates the feasibility of implementing a Float Car Data (FCD) system for real-time travel time estimation in Belgrade, Serbia. The research is motivated by the need for cost-effective, dynamic traffic monitoring solutions to address urban congestion, as traditional fixed detectors are expensive and limited in coverage. FCD systems utilize GPS-equipped vehicles to collect location and speed data, offering a scalable alternative. The authors specifically assess whether the existing fleet of GPS-equipped postal vehicles in Belgrade provides sufficient data coverage and reliability to support such a system. The methodology involves a pilot study using data from 220 postal vehicles equipped with GPS devices and wireless communication sensors. Data collection occurred over a representative week in April 2014, with a focus on a specific "representative day" (April 7, 2014) between 06:30 and 18:30. The vehicles recorded position, time, speed, and operational status every 15 seconds on average. The researchers processed this data using map-matching algorithms to reconstruct vehicle trajectories on the city’s road network. They analyzed spatial coverage, temporal distribution of data points, and the number of vehicle passes per road link. The study also calculated the minimum number of vehicles required to achieve recommended coverage standards (60–80% of the network with at least two passes per link within 10-minute intervals), referencing established literature and statistical models. The results indicate that the current postal fleet is insufficient for a reliable FCD system in Belgrade. While the 220 vehicles covered approximately 5,350 km of the road network during the observation period, this represented only about 40% of the total urban road length. Coverage of the primary road network peaked at 90% during midday hours but dropped significantly during morning peak hours (06:30–08:30), when only 40 vehicles were active. Analysis of 10-minute intervals revealed that even during the most active period, only 10% of the road network was covered. The study calculated that a minimum of 390 to 1,100 vehicles would be necessary to meet the standard requirement of covering 60% of the network with sufficient data density. Since the postal fleet provided only 95 active vehicles at its peak, it falls well below the threshold. The authors conclude that the existing postal vehicle fleet cannot independently support a robust real-time travel time estimation system in Belgrade due to inadequate spatial and temporal coverage. To establish a reliable FCD system, the city must integrate data from additional vehicle sources, such as taxis or commercial fleets, to increase the sample size. The study highlights that while FCD technology offers significant advantages in cost and coverage, its effectiveness is strictly dependent on having a sufficiently large and diverse pool of probe vehicles to ensure statistical reliability across the entire road network.

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.

• exposure measurement