INTERNATIONAL JOURNAL OF ARTIFICIAL INTELLIGENCE
ISSN: 2692-5206, Impact Factor: 12,23
American Academic publishers, volume 05, issue 05,2025
Journal:
https://www.academicpublishers.org/journals/index.php/ijai
page 1996
FREIGHT FLOW SIMULATION AND OPTIMIZATION ON THE UZBEKISTAN–
KYRGYZSTAN–CHINA RAILWAY USING AGENT-BASED MODELING
Bo‘riyev Shuxrat Xamroqul ugli
Introduction
In the era of globalized trade and interconnected supply chains, efficient freight transportation
systems are essential for economic growth. Central Asia’s geographic position between Europe
and East Asia offers unique opportunities to develop strategic land-based corridors. The
proposed Uzbekistan–Kyrgyzstan–China (UKC) railway is expected to become a vital link in
the Belt and Road Initiative (BRI), facilitating cargo transport between China and Central Asia,
and further extending to Europe and the Middle East.Despite its potential, the UKC railway’s
operational efficiency remains unexplored in terms of route performance, throughput, and
customs-related bottlenecks. This paper aims to evaluate and simulate freight flow along the
proposed UKC corridor using agent-based modeling (ABM), identifying key constraints and
optimization strategies. By leveraging modern simulation tools, we assess how logistics
performance can be enhanced through digitalization, infrastructure development, and
institutional coordination.
Methods
This research utilizes agent-based modeling (ABM) as a method to simulate the dynamic
behavior of freight transportation systems along the UKC corridor. The AnyLogic simulation
platform was selected for its ability to model complex, multi-agent logistics systems, including
variable cargo volumes, infrastructure constraints, and customs clearance processes.
Model components include:- **Agents:** Containers, freight trains, terminals, border
checkpoints, customs officials.- **Environment:** A geospatial representation of the UKC
railway line including node distances, border crossings, and elevation data.
- **Processes:** Loading/unloading, customs clearance, intermodal transfers.
- **Performance indicators:** Average delivery time, cargo throughput per day, terminal
utilization rate.
Input data was collected from national statistics agencies and international reports (ADB,
UNESCAP), while delay probabilities were estimated based on expert assessments. Three
scenarios were modeled: baseline (current infrastructure), optimized infrastructure, and digital
customs integration.
Results
Simulation results show that under baseline conditions, the average delivery time from Kashgar
(China) to Tashkent (Uzbekistan) is 9.8 days. In the optimized infrastructure scenario, which
includes double-track segments and expanded terminals, the delivery time decreased to 7.2 days.
INTERNATIONAL JOURNAL OF ARTIFICIAL INTELLIGENCE
ISSN: 2692-5206, Impact Factor: 12,23
American Academic publishers, volume 05, issue 05,2025
Journal:
https://www.academicpublishers.org/journals/index.php/ijai
page 1997
Further improvements with digital customs systems reduced delays at border checkpoints by
35%.
The throughput capacity of the corridor improved from 18 to 29 trains per day in the optimized
scenario, and container loss rates dropped by 15%. Table 1 summarizes the key performance
metrics across the three scenarios:
**Table 1. Performance Comparison Across UKC Railway Simulation Scenarios**
| Scenario
| Avg. Delivery Time | Trains/Day | Delay Time | Terminal Utilization |
|--------------------------|--------------------|------------|------------|----------------------|
| Baseline
| 9.8 days
| 18
| 22 hours | 76%
|
| Optimized Infrastructure| 7.2 days
| 24
| 16 hours | 83%
|
| Digital Integration | 6.5 days
| 29
| 10 hours | 91%
|
Discussion
The simulation findings suggest that targeted infrastructure upgrades and technological
integration can substantially improve the operational efficiency of the UKC railway. Border
delays, identified as the most critical bottleneck, can be mitigated through unified customs
procedures and shared digital platforms. The gains from optimizing terminal layout and
increasing capacity at major intermodal hubs also significantly reduced congestion-related
delays.The agent-based approach provides a flexible and scalable tool to test different policy
and investment scenarios. For example, the implementation of electronic documentation and
pre-arrival processing at border points can reduce clearance times by up to 50%.
These findings align with global best practices observed on other corridors such as the China–
Kazakhstan–Russia route. However, the UKC route's advantage lies in its shorter geography
and integration with southern Eurasian markets, offering a strong complement to existing
networks.
Conclusion
Agent-based modeling of the UKC railway demonstrates that freight flow optimization can
unlock substantial gains in efficiency and capacity. Policy reforms targeting customs
automation and infrastructure modernization are key enablers of success. Future work should
focus on integrating real-time data feeds from railway operations to enhance model accuracy.
Additionally, stakeholder coordination across borders remains essential for sustainable corridor
development.
References:
AnyLogic Simulation Software. (2023). _User Guide and Applications in Transport Logistics_.
2. Asian Development Bank. (2022). _Railway Connectivity in Central Asia_.
3. UNESCAP. (2023). _Digital and Sustainable Transport Corridors_.
4. World Bank. (2022). _Trade Facilitation and Logistics Performance Index_.
5. Kyrgyz Ministry of Transport. (2023). _Feasibility Studies on Transit Corridors_.
6. Zhang, X., & Liu, H. (2020). _Freight Simulation for Strategic Corridors_. _Transport
Modelling Review_, 8(2), 155–174.
INTERNATIONAL JOURNAL OF ARTIFICIAL INTELLIGENCE
ISSN: 2692-5206, Impact Factor: 12,23
American Academic publishers, volume 05, issue 05,2025
Journal:
https://www.academicpublishers.org/journals/index.php/ijai
page 1998
7. Tashkent Freight Logistics Hub. (2023). _Operational Report_.
8. Zhao, W. (2021). _Digital Border Management in Belt and Road Corridors_. _Global
Logistics Review_, 6(1), 45–67.
