Modeling and Analysis of Multimodal Travel Choice Behavior Considering User Portraits.

Saved in:

Bibliographic Details
Title:	Modeling and Analysis of Multimodal Travel Choice Behavior Considering User Portraits.
Authors:	Wei, Qian¹ wq09122022@163.com, He, Ruichun² Herc@mail.lzjtu.cn, Zhang, Shubin³ zhangshubin@mail.lzjtu.cn, Cao, Jiaying⁴ 1148399289@qq.com, Liu, Chenning¹ 1985164936@qq.com
Source:	IAENG International Journal of Applied Mathematics. Jun2026, Vol. 56 Issue 6, p2014-2025. 12p.
Subjects:	Discrete choice models, Consumer profiling, Clustering algorithms, Individuals' preferences, Transportation demand management
Geographic Terms:	Beijing (China)
Abstract:	This study examines preference heterogeneity in multimodal travel choice behavior by integrating user profiling with advanced discrete choice modeling. Using approximately 140,000 real-world travel records from Beijing, behavioral clustering methods are employed to construct user profiles that reflect differences in travel patterns and attribute sensitivities. Three model specifications--Multinomial Logit (MNL), an Extended MNL with interaction effects, and Mixed Logit (MXL)--are estimated and compared within a unified analytical framework. The extended MNL model incorporates interaction terms between user profiles and travel attributes, allowing systematic group-level differences in sensitivities to travel time, cost, and distance to be identified. The Mixed Logit model further accounts for continuous unobserved heterogeneity by specifying key coefficients as random parameters. The estimated standard deviations of these parameters are statistically significant, indicating substantial individual-level variation beyond observable segmentation. Model comparison results show that increasing behavioral flexibility leads to statistically significant improvements in model fit and explanatory power. Out-of-sample validation based on hold-out data further demonstrates that the MXL model provides more robust probabilistic predictions under unseen conditions. Overall, the results underscore the complementary roles of observable user profiling and continuous random-parameter modeling in capturing both discrete and continuous forms of heterogeneity in multimodal travel choice behavior. [ABSTRACT FROM AUTHOR]
	Copyright of IAENG International Journal of Applied Mathematics is the property of International Association of Engineers (IAENG) 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.)
Database:	Engineering Source

Description
Abstract:	This study examines preference heterogeneity in multimodal travel choice behavior by integrating user profiling with advanced discrete choice modeling. Using approximately 140,000 real-world travel records from Beijing, behavioral clustering methods are employed to construct user profiles that reflect differences in travel patterns and attribute sensitivities. Three model specifications--Multinomial Logit (MNL), an Extended MNL with interaction effects, and Mixed Logit (MXL)--are estimated and compared within a unified analytical framework. The extended MNL model incorporates interaction terms between user profiles and travel attributes, allowing systematic group-level differences in sensitivities to travel time, cost, and distance to be identified. The Mixed Logit model further accounts for continuous unobserved heterogeneity by specifying key coefficients as random parameters. The estimated standard deviations of these parameters are statistically significant, indicating substantial individual-level variation beyond observable segmentation. Model comparison results show that increasing behavioral flexibility leads to statistically significant improvements in model fit and explanatory power. Out-of-sample validation based on hold-out data further demonstrates that the MXL model provides more robust probabilistic predictions under unseen conditions. Overall, the results underscore the complementary roles of observable user profiling and continuous random-parameter modeling in capturing both discrete and continuous forms of heterogeneity in multimodal travel choice behavior. [ABSTRACT FROM AUTHOR]
ISSN:	19929978