1.1 Model Structure and Information Requirements

Transportation data is unique because, as Morgenstern (1, p. 44) notes, "When objects to be accounted for are moving about considerably, it may be impossible to obtain a desired precision." A survey of users of freight transportation data showed that academics most frequently used intercity tons by mode, rail car and TOFC loadings, and rail ton-miles. Consultants used rail car and TOFC loadings, followed by rail ton-miles, and then intercity tons by mode. Consultants viewed the carloadings data as more accurate than the intercity ton-miles reported for various modes (2, pp. 71-74).

1.2 Data Sources

Foremost among transportation data sources has been the work of Smith, whose Transportation Facts and Trends has been continued by Wilson as Transportation in America (TIA). This set of data tables is an annual publication with quarterly updates, and includes both macroeconomic data and more detailed information. For example, transportation’s share of the GNP was 16.3 percent in 1996, down from 17.2 percent ten years prior (3, p. 4). TIA provides some basic analysis of the data collected. Bar and pie charts display financial information, such as revenues of intercity carriers, petroleum consumption by mode, while line graphs show trends in trailers and containers purchased and ships built in the U.S.

National economic measures reported by TIA include GNP, population, intercity ton-miles and passenger-miles. Modal data report amounts spent for carrying freight by highways, in categories of intercity and local truck, and intercity bus. Rail revenues are grouped into one statistic, while water is segmented into international, coastal/intercoastal, inland waterways, Great Lakes, locks and channels. Oil pipeline revenues are reported as regulated and non-regulated, and airfreight is either domestic or international. Freight forwarder costs are considered in "other carriers" and additional costs of loading, unloading and other traffic operations and included, as well (3, p. 40). Two tables of much-used data are domestic intercity tons and ton-miles by mode, with historical data (in 5 year increments) extending back to 1950 (3, pp. 44-46). Transportation of petroleum is reported by ton-miles, by mode, the only major industry to receive this detailed treatment (3, p. 59). Employment, numbers of vehicles, miles of intercity airways, highways, waterways, railroads and pipelines are reported, as are distances covered, by mode, in for-hire passenger trips and interstate freight.


2.1 Traffic Zone Definition

The existing data sources are described as agglomerated over product lines and geographical boundaries. While the original data elements may have been individual shipments, described by origin and destination and by product type and amount, the reported data have been broadly grouped to obscure much potentially useful information. National data may be allocated according to some logical bases (e.g., state populations, employment by SIC) to arrive at the more useful estimates. Further allocation may be made on the basis of county size (or groups of counties), recognizing that collecting originally at the county (or smaller zone) would be more accurate, but require more resources than is generally practical. To be useful for planning purposes, i.e., to evaluate proposed transportation investments, data for smaller zones will be more valuable than national or state data.

2.2 Network Characteristics, Nodes and Links

The links and nodes for intermodal networks vary little from a single-mode network. The main difference is that the nodes need to be better defined, in terms of capacity (e.g., units or tons per hour, passengers per day). For example, bottlenecks may occur at ocean ports if ships needing unloading must wait for previous outbound traffic to be cleared

2.3 GIS Database Structure and Data Management Capabilities

An organizational scheme for transportation data was developed by Jack Faucett Associates (JFA) along four transportation system attributes: "supply, demand, performance, and impacts" (4, p. 2). Supply data describes the physical facilities (i.e., characteristics and financial condition of the service providers). Demand data describes the quantity of travel, temporal and spatial distribution, and users’ behavioral characteristics. Performance data measures how the transportation systems meet the needs of the users; cost data and safety data are found in this category. Impacts data describes the effects of transportation on environmental and other societal goals (4, p. 5).

Within the supply attributes, JFA sub-divides the data into systems, service, facilities, condition, and project data, and maintains separate data files for highway, rail, transit systems, ports and inland waterways, and air. Demand attributes are organized under the headings of economic data, demographic data, land use data, commodity flow data, travel data, and travel behavior data (4, pp. 7-16).


  1. Oskar Morgenstern, On the Accuracy of Economic Observations, 2nd ed. Princeton: Princeton University Press, 1963.
  2. Clyde Kenneth Walter, "Transportation Data: Theory and Perceptions of Accuracy," Transportation Journal, 26 (Summer 1987), pp. 67-81.
  1. Rosalyn A. Wilson, Transportation in America, 15th ed. Landsdowne, VA: Eno Transportation Foundation, Inc., 1997.
  2. Jack Faucett Associates, Inc., NCHRP Project Update, March 1996, p. 2.

Multimodal Investment Analysis: Phase 1 Contents

CTRE is an Iowa State University center.

Address: 2711 S. Loop Drive, Suite 4700, Ames, IA 50010-8664

Phone: 515-294-8103
FAX: 515-294-0467

CTRE Communications: Marcia Brink
CTRE Webmaster: Michele Regenold


Iowa State University--Becoming the Best