Agent-Based Approach in Freight Systems: Towards A Door-To-Door (D2D) Freight Optimization

Details

• Creators:
  Aredah, Ahmed ; Rakha, Hesham
• Corporate Creators:
  Virginia Tech ; Virginia Polytechnic Institute and State University ; Virginia Tech Transportation Institute
• Corporate Contributors:
  United States. Department of Transportation. University Transportation Centers (UTC) Program ; Morgan State University.Safety and Mobility Advancements Regional Transportation and Economics Research Center. University Transportation Center (UTC) ; United States. Department of Transportation. Office of the Assistant Secretary for Research and Technology
• Subject/TRT Terms:
  Automation Cost Effectiveness Energy Consumption Freight Traffic Greenhouse Gases Optimization Ship Large-Scale Simulation Ship Longitudinal Motion ShipNetSim Simulation Supply Chain Management Taxation

  More +
• Publication/ Report Number:
  SM33
• Resource Type:
  Tech Report
• Geographical Coverage:
  United States
• Edition:
  Final Report: September 2024 - August 2025
• Corporate Publisher:
  Morgan State University.Safety and Mobility Advancement Regional Transportation and Economics Research Center (SMARTER) University Transportation Center (UTC)
• Abstract:
  This report presents CargoNetSim, an open-source, modular, high-fidelity optimization and simulation framework developed to model, analyze, and optimize multi-modal freight transportation movement. Integrating agent-based modeling (ABM) and system dynamics (SD), CargoNetSim bridges micro-level operational decisions and macro-scale network behaviors, enabling granular quantification of fuel/energy consumption and total transportation costs. The framework comprises four validated modules: NeTrainSim (rail), ShipNetSim (maritime), INTEGRATION (road/truck), and TerminalSim (terminal operations), all synchronized via a central integration hub leveraging RabbitMQ for real-time data exchange. A cost optimization module precedes simulation by filtering infeasible or sub-optimal routes based on time, monetary value, emissions, and energy metrics, reducing computational demand without compromising accuracy. The system is calibrated using authoritative U.S. freight datasets, cost factors, energy coefficients, and carbon taxation structures aligned with 2030 policy forecasts. A comprehensive case study is conducted for transcontinental container transport from Madrid, Spain, to multiple inland U.S. destinations (Kansas City, Chicago, Dallas, and Los Angeles). Simulated costs reveal significant deviations from pre-estimated costs—e.g., $8,671.42 vs. $4,455.04 for Kansas City—demonstrating the critical role of dynamic simulation in capturing real-world inefficiencies such as rail stoppages, terminal congestion, and customs delays. Sensitivity analysis further shows that rail becomes cost-effective when transporting over eight containers at a $24.08/hour MVOT, shifting to 14 containers at a $96.32/hour MVOT, indicating economies of scale as a decisive factor. The results underscore CargoNetSim’s utility as a decision-support tool for logistics managers and policymakers, enabling evaluation of trade-offs between operational efficiency and environmental effects. The system's ability to simulate emissions, dwell times, fuel use, and modal transfers across international corridors positions it as a pivotal platform for strategic freight planning and supply chain policy development.
• Format:
  PDF
• Funding:
  69A3552348303
• Collection(s):
  University Transportation Centers
• Main Document Checksum:
  urn:sha-512:15078f6928032acfbc7cc58b7a5f0065ee382f2f3c8a63d755d8f84f6bec0283095b597c35f4421d8b73eddef3a8d7b6edfd43d6c7f10c2fec23abd1fc2b7903
• Download URL:
  https://rosap.ntl.bts.gov/view/dot/87917/dot_87917_DS1.pdf
• File Type:
  [PDF - 691.05 KB ]