Document Type



Master of Science (MS)


Civil Engineering

First Advisor's Name

Young-Kyun Lee

First Advisor's Committee Title

Committee Chair

Second Advisor's Name

Nii 0. Attoh-Okine

Third Advisor's Name

L. David Shen

Date of Defense



Demand forecasting is an essential element in the analysis of transportation systems. It is concerned with the behavior of consumers of transportation services and facilities. We choose geographic, demographic, and socioeconomic characteristics of consumers that may affect the travel demand of each selected. We use an artificial neural network to predict travel demand with characteristics selected from three different database sources: Census Summary Tape files, TIGER/Line files, and Federal Transit Administration's National Transit GIS database.

A neural network is an information processing system that is intensely parallel and neural networks are capable of learning how to classify and associate input/output patterns. This capability makes neural network a suitable approach for mode choice modeling for this study.

A neural network has two phases: the training and the testing. In the training phase, we find weights between inputs and outputs, and in the testing phase, neural network calculates outputs representing travel demand with weights from the training phase.





Rights Statement

Rights Statement

In Copyright. URI:
This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).