The prevailing disparity in the availability and demand for medical services necessitates an in-depth analysis and projection of the travel behavior patterns of medical travelers. First, this paper screens and processes data on travel to healthcare facilities from the Fifth Tokyo Metropolitan Area Resident Travel Survey. Second, the preliminary descriptive statistics reflected that the medical and health travelers in this dataset are elderly individuals, from smaller households, exhibiting high travel frequency, with a pronounced preference for morning travel. This paper used multinomial logit model, decision tree and random forest model to predict travelers’ choices of public transportation, private transportation, non-motorized vehicles and walking in daily healthcare travel scenarios. The models demonstrate robust fit, achieving commendable predictive accuracies of 63.3%, 65.41%, and 65.54%, respectively.
A Comparative Study of MNL and Machine Learning Methods for Travel Mode Choice of Medical Travel
Lecture Notes in Civil Engineering
International Conference on Traffic and Transportation Studies ; 2024 ; Lanzhou, China August 23, 2024 - August 25, 2024
The Proceedings of the 11th International Conference on Traffic and Transportation Studies ; Chapter : 35 ; 305-313
2024-11-14
9 pages
Article/Chapter (Book)
Electronic Resource
English
Modelling travel mode choice on combined data sources using machine learning methods
Elsevier | 2025
|Taylor & Francis Verlag | 2019
|Comparative mode choice analysis of university staff commuting travel preferences
DOAJ | 2022
|