Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network

  Advanced Search  

Integrating heterogeneous data sources for better freight flow analysis and planning

Summary: Integrating heterogeneous data sources
for better freight flow analysis and planning
Josť Luis Ambite, Genevieve Giuliano, Peter Gordon, Qisheng Pan, Sandipan Bhattacharjee
University of Southern California, Los Angeles, CA 90089
ambite@isi.edu, {giuliano,gordon,qpan,sandipab}@usc.edu
We present ongoing work on developing an automated integration system for freight flow analysis
and planning. To overcome the limitations of current estimation methods for commodity flows, we use
reliable secondary sources, such as employment data, and derive estimates in a principled way by means
of a computational workflow. When available, we extract the data automatically from online sources, so
that the system maintains the estimations continuously updated. This project will allow planners and
policymakers to make more informed decisions by accessing the most recent data and enhance the ability
to explore different scenarios.
1 Introduction
Economic restructuring and globalization have vastly increased the volume of commodity flows by
all transport modes. In the US, intercity ton-miles have increased approximately with GNP, but truck and
air transport have increased faster than other modes. In 1994, for example,, of the total bill of $420 billion
the US spent on transportation, trucks carried 79%, i.e. $331 billion (USDOT, 1998). Total US ton-miles
of freight increased from 2989 billion in 1980 to 3710 billion in 1998. Over the same period intercity
truck ton-miles almost doubled (from 555 to 1027 billion), and domestic air ton-miles increased from 4.5


Source: Ambite, Josť Luis - Information Sciences Institute & Department of Computer Science, University of Southern California


Collections: Computer Technologies and Information Sciences