The Demand for Water Transportation: Application of Discriminant Analysis to Commodities Shipped by Barge and Competing Modes in Ohio River and Arkansas River Areas.

Abstract

This dissertation develops a methodology for simulating transport demand functions for barge transportation. Discriminant analysis is utilized to calibrate a modal choice model from disaggregate observations of individual shipments. The modal choice model stems earlier development by Moses and Lave (1971), Allen (1969) and Beuthe (1968). Transport demand functions for barge shipments are developed for coal used for energy and metallurgical market. Barge demand functions are developed for commodities other than coal including chemicals and refined petroleum products. Data utilized in the analysis represents a total of 815 shipments totaling 145.9 million tons shipped annually for 14 major commodity groups utilizing 9 transport modes. When the discriminant models were calibrated from logarithmic transformed data, 83-100 percent of the individual shipments were classified correctly. Further testing by a random holdout procedure showed no significant upward bias in classification. Conventional rail and unit train modes comprised about 54 percent of the shipments and 33 percent of annual tonnage in the sample.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Aug 01, 1980
Accession Number
ADA096551

Entities

People

  • Lloyd G. Antle

Tags

Communities of Interest

  • Biomedical
  • Energy and Power Technologies
  • Ground and Sea Platforms

DTIC Thesaurus Topics

  • Commerce
  • Computer Programming
  • Computers
  • Data Science
  • Discriminant Analysis
  • Economic Analysis
  • Economics
  • Elastic Properties
  • Environmental Protection
  • Freight Transportation
  • Geography
  • Information Science
  • Inland Waterways
  • Investments
  • Regression Analysis
  • Surveys
  • Transportation

Readers

  • Industrial Economics
  • Regression Analysis.
  • Riverine Ecology

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Machine Learning Algorithms