Disaggregated modeling of mode choice by ANN - a case study of Ahmedabad City in Gujarat State

Traveling is an integral part of today's lifestyle for people across the world. The increased traveling has led to a number of serious problems like congestion, noise pollution, air pollution, greenhouse effect, etc. In transportation planning, the choice of a transportation mode is one of the most important parameters, and it is difficult to predict the same, as it depends on human behavior, which is very complex in nature. By far, most of the discrete mode choice models are based on the principle of "random utility maximization," derived from the Econometric theory. However, in the present study, the artificial intelligence technique is used for modeling of the mode choice behavior. Further, an attempt has been made to predict mode choice by using neural network technique. The present study is aimed at introducing a new modeling technique, artificial neural network (ANN). An ANN is inspired by biological neurons as it learns from the past. The ANN is best suited for the problems where input variables are complex in nature. The study provides guidelines in deciding network architecture for the behavior model. For efficient use of ANN technique, it is required to decide types of activation functions, the number of neuron/s in different layers and the amount of data used for the training. The data used for the present study were collected from the household travel survey conducted in the Ahmedabad city of Gujarat state for the Public Transportation System. In the study an attempt has been made to find out the sensitivity of the various parameters in the model. Same data is also analyzed by linear regression method to obtain utility function and finally the output of ANN model is compared with the regression model.

• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/issn/02580500
• Supplemental Notes:
  • Abstract reprinted with permission from the publisher.
• Authors:
  • Ramanuj, P S
  • Gundaliya, P J
• Publication Date: 2013

Language

• English

Media Info

• Media Type: Print
• Features: Figures; References; Tables;
• Pagination: pp 3-12
• Serial:
  • Journal of the Indian Roads Congress
  • Volume: 74
  • Issue Number: 1
  • Publisher: Indian Roads Congress
  • ISSN: 0258-0500

Subject/Index Terms

• TRT Terms: Case studies; Linear regression analysis; Mode choice; Neural networks; Public transit; Travel surveys
• Geographic Terms: Ahmedabad (India)
• Subject Areas: Planning and Forecasting; Public Transportation; I72: Traffic and Transport Planning;

Filing Info

• Accession Number: 01483063
• Record Type: Publication
• Files: TRIS
• Created Date: Jun 4 2013 11:07AM