Treffer: Investigating the Users' Perception of Metro Rail Service Quality with Machine Learning Approach and Explainable Artificial Intelligence—Case Study: Dhaka, Bangladesh.

Title:
Investigating the Users' Perception of Metro Rail Service Quality with Machine Learning Approach and Explainable Artificial Intelligence—Case Study: Dhaka, Bangladesh.
Authors:
Fahim, Ashraf Uddin, Minami, Masaaki
Source:
Environment & Urbanization ASIA; Mar2025, Vol. 16 Issue 1, p27-40, 14p
Subject Terms:
MACHINE learning, ARTIFICIAL intelligence, QUALITY of service, STATISTICS, PUBLIC transit
Database:
Complementary Index

Weitere Informationen

Users of public transit have diverse views on service quality. In Dhaka, travelling by metro rail service has played a significant role in people's transportation ever since it was established in this city in 2022. This survey examined metro riders' satisfaction with station, ticketing, security, information, platform, employee and metro rail services. A purposeful sample strategy was utilized to collect 382 fully screened surveys on Dhaka Metro Users' satisfaction. Data were analysed using machine learning and explainable AI. IBM SPSS 25 statistical software performed some statistical analysis, like chi-square tests and frequencies, whereas Python and Google Colab used machine learning models. Metro rail services were found satisfactory by 46.7% men and 36.2% women. Machine learning models found that in terms of quality, station-related services ranked highest, whereas platform, ticketing and security services had the least impact. Maximum service quality prediction accuracy was 84% with the Random Forest model. To meet metro users' needs, the most essential variable will be used to improve services. This study shows how machine learning and explainable AI (using SHAP values) can discover transport service elements for improvement. [ABSTRACT FROM AUTHOR]

Copyright of Environment & Urbanization ASIA is the property of Sage Publications Inc. and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)