Method for analysis and prediction of dwell times at stops in local bus transportation

Abstract

The punctuality of local public bus transportation contributes to the service quality and has impact on mode choice. Appropriate planning and control measures as well as adequate passenger information require efficient analysis and prediction methods. Advanced fleet tracking systems provide enough data for these research purposes. The main factors that cause schedule deviation has been identified through the analysis of the data. As important time elements of the journey time are dwell times at stops, the research focused on it; however, the elaborated database structure is adequate also for analysis of the other time elements, which coincides with our further research intentions too. Innovative methods based on the historical data have been elaborated for the prediction of dwell times. The essence of the method: multi-variate analysis of the dwell times by exploration of the significant influencing factors, and then prediction of times based on the factors describing the certain situations. Soundness of the methods has been verified with examples, and the results well approximated the real values.

Keywords:

• data analysis
• database structure
• bus transport
• prediction
• dwell times