Screening Mississippi River Container-on-Barge Demand: A TAM–SAM–SOM Hierarchy with Pooled Machine Learning Forecasts
Year: 2026
Type: article
Abstract: The Mississippi River offers cost and environmental advantages for freight; however, containerized service remains underused. This paper presents a prefeasibility market screening framework for a hypothetical container-on-barge service at 15 Mississippi mainstem candidate nodes, using 9,797,087 import records across 12 consignee states. The Total Addressable Market, Serviceable Addressable Market, and Serviceable Obtainable Market (TAM–SAM–SOM) hierarchy applies geographic accessibility, commodity time-sensitivity, and container compatibility screens before an assumed capture rate. Pooled commodity group candidate models forecast January 2024–December 2025 and are evaluated by expanding-window, 12-month recursive holdouts; the median pair-selected weighted absolute percentage error is 21.9%. Under the Moderate scenario (100 mi radius, accelerated service, 25% assumed capture), the TAM is 2,305,427 annual physical containers, SAM is 79,227 (3.4% of TAM), and the assumed capture SOM is 19,807. A stratified internal author review confirms 585 of 600 commodity assignments (97.5%). Sensitivity analyses examine the effects of gateway-local flows, missing ZIP codes, unknown equipment, and scenario assumptions, alongside forecast performance diagnostics. These estimates do not predict realized modal diversion or establish project-level investment feasibility; they provide a transparent basis for prioritizing locations and commodity segments for shipper validation, route-level cost analysis, and operational feasibility assessment.
Authors: Alex Carroll, Elenna R. Dugundji
Institution: Massachusetts Institute of Technology
Language: English
Topic: Maritime Ports and Logistics
Subfield: Industrial and Manufacturing Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $3,650
OpenAlex ID: W7234018483
