Bus arrival time prediction using adaptive neuro-fuzzy inference system

Articles in Press

Document Type : Scientific - Research

Authors

1 Department of Civil Engineering, Yazd University, Iran

2 Ph.D. candidate, University of Tehran

Abstract

The provision of accurate bus arrival travel time information is vital since it increases the quality of public transport services. Improving the quality of service increases the satisfaction of bus users and in turn attracts additional ridership. In this paper, a model based on adaptive neural-fuzzy inference system is developed to predict bus arrival time at bus stops. Real travel time data from a bus route located in Tehran, Iran, collected over three months is used for calibration and validation of the model. A regression-based model is also developed to predict bus arrival time. The results of this study show that the neuro-fuzzy model can predict arrival time in more than 86% of cases with a maximum error of 20%. The root-mean-squared error is employed as an index to compare the proposed model with the regression-based model. It is found that the neuro-fuzzy model outperforms the regression-based model in terms of prediction accuracy.

Keywords

Main Subjects

Quarterly Journal of Transportation Engineering

Articles in Press, Accepted Manuscript
Available Online from 09 July 2023

Files

History

  • Receive Date: 13 February 2023
  • Revise Date: 24 May 2023
  • Accept Date: 10 June 2023

Share

How to cite

Statistics