Fusing Shipper Behavior Models between Markets and Approaches

Understanding shipper behavior is critical for informed freight transportation planning and policy development. Despite the availability of various modeling approaches—including traditional analytical methods and emerging artificial intelligence (AI) techniques—significant variability persists across commodity types, shipment distances, and market scales. This project addresses the need for a unified and systematic framework to compare, integrate, and enhance shipper behavior models. Building on the complementary expertise of the Principal Investigator (PI) and Co-Principal Investigator (Co-PI), the study will conduct comparative analyses of existing models, focusing on the integration of AI-based and analytical approaches such as multinomial logit (MNL) models. The research will examine model performance across diverse market conditions and geographies, using the Commodity Flow Survey (CFS) data as a foundational resource. Emphasis will be placed on developing fusion techniques to bridge methodological gaps and improve predictive accuracy, particularly in the face of imbalanced datasets common in freight data. By unifying modeling strategies and addressing data limitations, this work aims to deliver a robust framework with enhanced generalizability and practical utility. The expected outcomes include improved forecasting tools, better policy support, and more effective use of publicly available data for national and regional freight planning efforts.

Language

• English

Project

• Status: Active
• Funding: $397,500.00
• Contract Numbers:

  69A3552348338

• Sponsor Organizations:

  Office of the Assistant Secretary for Research and Technology

  University Transportation Centers Program
  Department of Transportation
  Washington, DC  United States  20590
• Managing Organizations:

  Center for Freight Transportation for Efficient and Resilient Supply Chain

  University of Tennessee Knoxville
  Knoxville, TN  United States  37996
• Project Managers:

  Bruner, Britain

  Kaplan, Marcella

• Performing Organizations:

  University of Tennessee, Knoxville

  Knoxville, TN  United States 

  Texas A&M University, College Station

  Zachry Department of Civil Engineering
  3136 TAMU
  College Station, TX  United States  77843-3136
• Principal Investigators:

  Han, Lee

  Wang, Bruce

  Zhang, Yunlong

• Start Date: 20250901
• Expected Completion Date: 20260831
• Actual Completion Date: 0
• USDOT Program: University Transportation Centers Program

Subject/Index Terms

• TRT Terms: Data analysis; Freight transportation; Multinomial logits; Predictive models; Shipping
• Subject Areas: Data and Information Technology; Freight Transportation; Planning and Forecasting;

Filing Info

• Accession Number: 01984314
• Record Type: Research project
• Source Agency: Center for Freight Transportation for Efficient and Resilient Supply Chain
• Contract Numbers: 69A3552348338
• Files: UTC, RIP
• Created Date: Mar 25 2026 4:39PM