EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Applications of Stochastic Processes in Modeling Uncertainty and Random Variation: Evidence from Hourly Traffic Flow Dynamics

Chalang Suleiman Hussein

Qubahan Journal of Social Science · 07.03.2026 · 3 (1) · 1-16

DOI: 10.48161/qjss.v3n1a13 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Qubahan Journal of Social Science, online erschienen am 07.03.2026. Verfasst von Chalang Suleiman Hussein. Die Arbeit nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Stochastic processes provide a rigorous probability-based foundation for analyzing systems that evolve over time under uncertainty and random variation. This study develops and evaluates an empirical stochastic-process framework for modeling hourly traffic-flow dynamics using a real multi-junction traffic dataset. The framework integrates Markov-chain state transitions, high-traffic event modeling, random-variation analysis, and Monte Carlo simulation to examine how traffic conditions transition among low, moderate, high, and congested states. The dataset contains 48,120 hourly observations from four traffic junctions, with no missing values or duplicate records. We defined traffic states using junction-specific calibration quartiles, and we reserved the final month of observations for out-of-sample evaluation. The results show strong temporal dependence in traffic dynamics. Lag-one autocorrelation ranged from 0.630 to 0.963, while weekly autocorrelation remained high for Junctions 1 and 2, indicating persistent operational cycles. Markov transition matrices revealed substantial state persistence, particularly in low and congested states, while upward and downward transitions captured short-run changes in traffic intensity. One-step stochastic forecasts achieved an average state-classification accuracy of 68.72%, with an average 95% prediction-interval coverage of 89.10%. High-traffic event counts exhibited overdispersion across all junctions, supporting the use of negative-binomial robustness alongside Poisson event-rate estimation. Monte Carlo simulation generated interpretable weekly traffic scenarios and uncertainty intervals for mean, peak, and total vehicle counts. The study contributes a transparent empirical framework that connects classical stochastic-process theory with operational traffic monitoring and uncertainty-aware decision support.

Zitieren

APA

Hussein, C. S. (2026). Applications of Stochastic Processes in Modeling Uncertainty and Random Variation: Evidence from Hourly Traffic Flow Dynamics. Qubahan Journal of Social Science, 3(1), 1-16. https://doi.org/10.48161/qjss.v3n1a13

Vancouver

Hussein CS. Applications of Stochastic Processes in Modeling Uncertainty and Random Variation: Evidence from Hourly Traffic Flow Dynamics. Qubahan Journal of Social Science. 2026;3(1):1-16. doi:10.48161/qjss.v3n1a13

BibTeX
@article{Hussein2026,
  author = {Hussein, Chalang Suleiman},
  title = {Applications of Stochastic Processes in Modeling Uncertainty and Random Variation: Evidence from Hourly Traffic Flow Dynamics},
  journal = {Qubahan Journal of Social Science},
  year = {2026},
  volume = {3},
  number = {1},
  pages = {1--16},
  doi = {10.48161/qjss.v3n1a13}
}
Kostenlos starten →