2024
Wang, Boqing; Yang, Min; Feng, Tao; Yang, Yuyuan; Yuan, Yalong
Heterogeneous choice of personalized Mobility-as-a-Service bundles and its impact on sustainable transportation Journal Article
In: Transportation Research Part D: Transport and Environment, vol. 131, pp. 104224, 2024, ISSN: 1361-9209.
Abstract | Links | BibTeX | Tags: Choice behavior, Heterogeneity, Mobility as a service (MaaS), Sustainable transportation, Travel bundle
@article{WANG2024104224,
title = {Heterogeneous choice of personalized Mobility-as-a-Service bundles and its impact on sustainable transportation},
author = {Boqing Wang and Min Yang and Tao Feng and Yuyuan Yang and Yalong Yuan},
url = {https://www.sciencedirect.com/science/article/pii/S1361920924001810},
doi = {https://doi.org/10.1016/j.trd.2024.104224},
issn = {1361-9209},
year = {2024},
date = {2024-01-01},
journal = {Transportation Research Part D: Transport and Environment},
volume = {131},
pages = {104224},
abstract = {The Mobility-as-a-service (MaaS) framework has the potential to induce changes in individual travel behavior towards more sustainable transportation alternatives by integrating multimodal travel services. However, insights into the behavioral responses of specific groups of people to MaaS, thus promoting green travel, are currently lacking. This paper intends to investigate the intergroup heterogeneous preferences on the choice of MaaS bundles and evaluate their contributions to eco-friendly travel. A stated choice experiment is designed, incorporating personalized MaaS service bundles complemented by different green travel-oriented ticketing strategies. The experiment caters to different groups of users related to public transport, cars, and other modes of transport. The results indicate a substantial improvement in the goodness-of-fit of the behavioral model after user classification, revealing significant heterogeneity in preferences among users regarding MaaS bundle choices. Notably, the impact of sub-tickets on nudging toward green transportation is more pronounced than that of monthly and discount tickets.},
keywords = {Choice behavior, Heterogeneity, Mobility as a service (MaaS), Sustainable transportation, Travel bundle},
pubstate = {published},
tppubtype = {article}
}
The Mobility-as-a-service (MaaS) framework has the potential to induce changes in individual travel behavior towards more sustainable transportation alternatives by integrating multimodal travel services. However, insights into the behavioral responses of specific groups of people to MaaS, thus promoting green travel, are currently lacking. This paper intends to investigate the intergroup heterogeneous preferences on the choice of MaaS bundles and evaluate their contributions to eco-friendly travel. A stated choice experiment is designed, incorporating personalized MaaS service bundles complemented by different green travel-oriented ticketing strategies. The experiment caters to different groups of users related to public transport, cars, and other modes of transport. The results indicate a substantial improvement in the goodness-of-fit of the behavioral model after user classification, revealing significant heterogeneity in preferences among users regarding MaaS bundle choices. Notably, the impact of sub-tickets on nudging toward green transportation is more pronounced than that of monthly and discount tickets.
2022
Li, Xiaodong; Feng, Tao; Rasouli, Soora
Exploring random taste heterogeneity in choice modelling using mixture density network Conference
7th International Choice Modelling Conference (ICMC), May 23-25, 2022. Reykjavik, Iceland, 2022.
Abstract | Links | BibTeX | Tags: Choice models, Heterogeneity, Machine learning
@conference{8ee855dfda764e948993b1c62df890a4,
title = {Exploring random taste heterogeneity in choice modelling using mixture density network},
author = {Xiaodong Li and Tao Feng and Soora Rasouli},
url = {http://www.icmconference.org.uk/2022-icmc-reykjavik.html},
year = {2022},
date = {2022-01-31},
urldate = {2022-01-31},
address = {Reykjavik, Iceland},
organization = {7th International Choice Modelling Conference (ICMC), May 23-25, 2022.},
abstract = {Capturing heterogeneity in subjects’ decision making process, as accurate as possible, plays an essential role in choice modeling research. In this paper, we investigate the random taste heterogeneity in travel behavior modeling which is an integral part of decision-making process. In contrast to previous works, we use the Mixture Density Network (MDN) which is built from Neural Network and mixture Gaussian model to identify the latent heterogeneity. We assume that the taste variation of individuals follows a series of distribution with certain mean and standard deviation which are dependent on individual social demographic characteristics. We integrated this machine learning method into the discrete choice model and jointly estimated the parameters. Using the stated preference data of Swissmetro, we applied our proposed model and discovered random taste variations which are highly interpretable. We also compared the model with traditional mixed logit model and found the superiority of the proposed model.},
keywords = {Choice models, Heterogeneity, Machine learning},
pubstate = {published},
tppubtype = {conference}
}
Capturing heterogeneity in subjects’ decision making process, as accurate as possible, plays an essential role in choice modeling research. In this paper, we investigate the random taste heterogeneity in travel behavior modeling which is an integral part of decision-making process. In contrast to previous works, we use the Mixture Density Network (MDN) which is built from Neural Network and mixture Gaussian model to identify the latent heterogeneity. We assume that the taste variation of individuals follows a series of distribution with certain mean and standard deviation which are dependent on individual social demographic characteristics. We integrated this machine learning method into the discrete choice model and jointly estimated the parameters. Using the stated preference data of Swissmetro, we applied our proposed model and discovered random taste variations which are highly interpretable. We also compared the model with traditional mixed logit model and found the superiority of the proposed model.
2021
Yuan, Yalong; Yang, Min; Feng, Tao; Rasouli, Soora; Ruan, Xinpei; Wang, Xinyu; Li, Yu
Analyzing heterogeneity in passenger satisfaction, loyalty, and complaints with air-rail integrated services Journal Article
In: TRANSPORTATION RESEARCH PART D-TRANSPORT AND ENVIRONMENT, vol. 97, 2021, ISSN: 1361-9209.
Abstract | Links | BibTeX | Tags: Air-Rail, Heterogeneity
@article{WOS:000682934900007,
title = {Analyzing heterogeneity in passenger satisfaction, loyalty, and complaints with air-rail integrated services},
author = {Yalong Yuan and Min Yang and Tao Feng and Soora Rasouli and Xinpei Ruan and Xinyu Wang and Yu Li},
url = {https://www.sciencedirect.com/science/article/pii/S1361920921002480},
doi = {10.1016/j.trd.2021.102950},
issn = {1361-9209},
year = {2021},
date = {2021-08-01},
urldate = {2021-08-01},
journal = {TRANSPORTATION RESEARCH PART D-TRANSPORT AND ENVIRONMENT},
volume = {97},
abstract = {To increase the understanding of passenger choice of air-rail integrated
services (ARIS), the present paper a) analyzes the profile of intermodal
passengers by classifying them based on their socioeconomic and travel
characteristics and (b) uncovers any heterogeneity in passenger
satisfaction with ARIS among different passenger groups. Through
collecting 1,345 passenger satisfaction questionnaires about ARIS,
surveyed in Shijiazhuang, China, we first apply latent class clustering
to classify passengers into three groups. Then, the key factors
affecting passenger satisfaction and its subsequent effect on loyalty
and complaints are identified using a partial least squares structural
equation model. Finally, path heterogeneities in different passenger
groups are explored using a multi-group analysis. The results show that
the three passenger groups attach similar importance to ticket services,
reliability, accessibility and comfort, while some noticeable
differences in perception are identified in personalized services,
information services, and connectivity.},
keywords = {Air-Rail, Heterogeneity},
pubstate = {published},
tppubtype = {article}
}
To increase the understanding of passenger choice of air-rail integrated
services (ARIS), the present paper a) analyzes the profile of intermodal
passengers by classifying them based on their socioeconomic and travel
characteristics and (b) uncovers any heterogeneity in passenger
satisfaction with ARIS among different passenger groups. Through
collecting 1,345 passenger satisfaction questionnaires about ARIS,
surveyed in Shijiazhuang, China, we first apply latent class clustering
to classify passengers into three groups. Then, the key factors
affecting passenger satisfaction and its subsequent effect on loyalty
and complaints are identified using a partial least squares structural
equation model. Finally, path heterogeneities in different passenger
groups are explored using a multi-group analysis. The results show that
the three passenger groups attach similar importance to ticket services,
reliability, accessibility and comfort, while some noticeable
differences in perception are identified in personalized services,
information services, and connectivity.
services (ARIS), the present paper a) analyzes the profile of intermodal
passengers by classifying them based on their socioeconomic and travel
characteristics and (b) uncovers any heterogeneity in passenger
satisfaction with ARIS among different passenger groups. Through
collecting 1,345 passenger satisfaction questionnaires about ARIS,
surveyed in Shijiazhuang, China, we first apply latent class clustering
to classify passengers into three groups. Then, the key factors
affecting passenger satisfaction and its subsequent effect on loyalty
and complaints are identified using a partial least squares structural
equation model. Finally, path heterogeneities in different passenger
groups are explored using a multi-group analysis. The results show that
the three passenger groups attach similar importance to ticket services,
reliability, accessibility and comfort, while some noticeable
differences in perception are identified in personalized services,
information services, and connectivity.
Peng, Y.; Feng, T.; Timmermans, H. J. P.
Heterogeneity in outdoor comfort assessment in urban public spaces Journal Article
In: Science of the Total Environment, vol. 790, 2021, ISSN: 18791026.
Abstract | Links | BibTeX | Tags: Heterogeneity, Latent class path model, Outdoor comfort, Urban public space
@article{Peng2021,
title = {Heterogeneity in outdoor comfort assessment in urban public spaces},
author = {Y. Peng and T. Feng and H. J. P. Timmermans},
doi = {10.1016/j.scitotenv.2021.147941},
issn = {18791026},
year = {2021},
date = {2021-01-01},
journal = {Science of the Total Environment},
volume = {790},
abstract = {The assessment of outdoor comfort can provide valuable insights into the quality of urban public spaces. Rational indices based on the heat-balance model have been extensively used for assessing thermal comfort in various outdoor environments for a long time. However, a growing body of literature is arguing the theoretical limitations of rational indices, including the lack of contextual considerations and the non-consideration of the active role of human being. Furthermore, the well-documented inconsistency in previous investigations suggests individuals' heterogeneous thermal expectations, preferences and adaptations typically depend on person-related and place-related contexts. Nevertheless, the study of heterogeneity in comfort assessment remains fragmentary. In an attempt to incorporate heterogeneity into a comprehensive conceptual framework of outdoor comfort, this study develops a latent class path model based on empirical data of 701 respondents from Eindhoven, the Netherlands. We identified two latent classes associated with different causal structures underlying the assessment of outdoor comfort. Meanwhile, the membership of latent classes is identified based on respondents' socio-demographic and behavioral covariates. The results show that, with respect to the influence on comfort assessment, the exogenous and endogenous variables are varying in effectiveness and strength between the two latent classes. Latent class 1 assesses the outdoor comfort mainly based on thermal sensation and expectations of thermal and wind conditions, whereas latent class 2 comprehensively considered both thermal and non-thermal influences, especially, the psychological acceptability and need satisfaction of outdoor activity. Our findings suggest that, in addition to the expanded framework of outdoor comfort study, understanding the heterogeneity in respondents' comfort perception is needed to comprehensively address the approaches to archive the comfortable urban public spaces in design, planning and management practices.},
keywords = {Heterogeneity, Latent class path model, Outdoor comfort, Urban public space},
pubstate = {published},
tppubtype = {article}
}
The assessment of outdoor comfort can provide valuable insights into the quality of urban public spaces. Rational indices based on the heat-balance model have been extensively used for assessing thermal comfort in various outdoor environments for a long time. However, a growing body of literature is arguing the theoretical limitations of rational indices, including the lack of contextual considerations and the non-consideration of the active role of human being. Furthermore, the well-documented inconsistency in previous investigations suggests individuals’ heterogeneous thermal expectations, preferences and adaptations typically depend on person-related and place-related contexts. Nevertheless, the study of heterogeneity in comfort assessment remains fragmentary. In an attempt to incorporate heterogeneity into a comprehensive conceptual framework of outdoor comfort, this study develops a latent class path model based on empirical data of 701 respondents from Eindhoven, the Netherlands. We identified two latent classes associated with different causal structures underlying the assessment of outdoor comfort. Meanwhile, the membership of latent classes is identified based on respondents’ socio-demographic and behavioral covariates. The results show that, with respect to the influence on comfort assessment, the exogenous and endogenous variables are varying in effectiveness and strength between the two latent classes. Latent class 1 assesses the outdoor comfort mainly based on thermal sensation and expectations of thermal and wind conditions, whereas latent class 2 comprehensively considered both thermal and non-thermal influences, especially, the psychological acceptability and need satisfaction of outdoor activity. Our findings suggest that, in addition to the expanded framework of outdoor comfort study, understanding the heterogeneity in respondents’ comfort perception is needed to comprehensively address the approaches to archive the comfortable urban public spaces in design, planning and management practices.
Yuan, Y.; Yang, M.; Feng, T.; Rasouli, S.; Ruan, X.; Wang, X.; Li, Y.
Analyzing heterogeneity in passenger satisfaction, loyalty, and complaints with air-rail integrated services Journal Article
In: Transportation Research Part D: Transport and Environment, vol. 97, 2021, ISSN: 13619209.
Abstract | Links | BibTeX | Tags: Air-rail integrated services, Heterogeneity, Latent class clustering, Multi-group analysis, Passenger loyalty, Passenger satisfaction
@article{Yuan2021,
title = {Analyzing heterogeneity in passenger satisfaction, loyalty, and complaints with air-rail integrated services},
author = {Y. Yuan and M. Yang and T. Feng and S. Rasouli and X. Ruan and X. Wang and Y. Li},
doi = {10.1016/j.trd.2021.102950},
issn = {13619209},
year = {2021},
date = {2021-01-01},
journal = {Transportation Research Part D: Transport and Environment},
volume = {97},
abstract = {To increase the understanding of passenger choice of air-rail integrated services (ARIS), the present paper a) analyzes the profile of intermodal passengers by classifying them based on their socioeconomic and travel characteristics and (b) uncovers any heterogeneity in passenger satisfaction with ARIS among different passenger groups. Through collecting 1,345 passenger satisfaction questionnaires about ARIS, surveyed in Shijiazhuang, China, we first apply latent class clustering to classify passengers into three groups. Then, the key factors affecting passenger satisfaction and its subsequent effect on loyalty and complaints are identified using a partial least squares structural equation model. Finally, path heterogeneities in different passenger groups are explored using a multi-group analysis. The results show that the three passenger groups attach similar importance to ticket services, reliability, accessibility and comfort, while some noticeable differences in perception are identified in personalized services, information services, and connectivity.},
keywords = {Air-rail integrated services, Heterogeneity, Latent class clustering, Multi-group analysis, Passenger loyalty, Passenger satisfaction},
pubstate = {published},
tppubtype = {article}
}
To increase the understanding of passenger choice of air-rail integrated services (ARIS), the present paper a) analyzes the profile of intermodal passengers by classifying them based on their socioeconomic and travel characteristics and (b) uncovers any heterogeneity in passenger satisfaction with ARIS among different passenger groups. Through collecting 1,345 passenger satisfaction questionnaires about ARIS, surveyed in Shijiazhuang, China, we first apply latent class clustering to classify passengers into three groups. Then, the key factors affecting passenger satisfaction and its subsequent effect on loyalty and complaints are identified using a partial least squares structural equation model. Finally, path heterogeneities in different passenger groups are explored using a multi-group analysis. The results show that the three passenger groups attach similar importance to ticket services, reliability, accessibility and comfort, while some noticeable differences in perception are identified in personalized services, information services, and connectivity.
Gu, G.; Feng, T.; Yang, D.; Timmermans, H.
Modeling dynamics in household car ownership over life courses: a latent class competing risks model Journal Article
In: Transportation, vol. 48, iss. 2, 2021, ISSN: 15729435.
Abstract | Links | BibTeX | Tags: Car ownership, Heterogeneity, Latent class competing risks model, Life events
@article{Gu2021b,
title = {Modeling dynamics in household car ownership over life courses: a latent class competing risks model},
author = {G. Gu and T. Feng and D. Yang and H. Timmermans},
doi = {10.1007/s11116-019-10078-8},
issn = {15729435},
year = {2021},
date = {2021-01-01},
journal = {Transportation},
volume = {48},
issue = {2},
abstract = {This study presents a latent class competing risks model to examine the influence of socio-demographics and life course events on car transaction behaviour. The types of car transaction and interval times between car transactions events are incorporated in a competing risk model. To capture unobserved behavioural heterogeneity across the population, the model classifies households into different segments. Results estimated based on retrospective survey data show significant heterogeneity exist in household car ownership decisions. The covariates are found to have different effects on car ownership decisions between different classes. Households in the class labelled “Young households without a car” are more sensitive to life course events related to household composition. Households labelled as “middle-aged and aged households with car(s)” are more sensitive to life course events related to job and house locations.},
keywords = {Car ownership, Heterogeneity, Latent class competing risks model, Life events},
pubstate = {published},
tppubtype = {article}
}
This study presents a latent class competing risks model to examine the influence of socio-demographics and life course events on car transaction behaviour. The types of car transaction and interval times between car transactions events are incorporated in a competing risk model. To capture unobserved behavioural heterogeneity across the population, the model classifies households into different segments. Results estimated based on retrospective survey data show significant heterogeneity exist in household car ownership decisions. The covariates are found to have different effects on car ownership decisions between different classes. Households in the class labelled “Young households without a car” are more sensitive to life course events related to household composition. Households labelled as “middle-aged and aged households with car(s)” are more sensitive to life course events related to job and house locations.
2020
Guo, J.; Feng, T.; Timmermans, H. J. P.
Modeling co-dependent choice of workplace, residence and commuting mode using an error component mixed logit model Journal Article
In: Transportation, vol. 47, iss. 2, 2020, ISSN: 15729435.
Abstract | Links | BibTeX | Tags: Co-dependent location, Error component logit model, Heterogeneity, Multidimensional choice, Panel data, Pivoted stated choice experiment
@article{Guo2020,
title = {Modeling co-dependent choice of workplace, residence and commuting mode using an error component mixed logit model},
author = {J. Guo and T. Feng and H. J. P. Timmermans},
doi = {10.1007/s11116-018-9927-y},
issn = {15729435},
year = {2020},
date = {2020-01-01},
journal = {Transportation},
volume = {47},
issue = {2},
abstract = {This paper develops an error component mixed logit model to analyze the multi-dimensional residential, work and transportation mode choice. It expanse previous studies based on life-trajectory theory which predominantly only considered two life domains. In contributing to this emerging field of research, we design an integrated pivoted stated choice experiment considering the multi-dimensional choice of job, residence and transportation mode for the journey to work. The results of the estimated error component mixed logit model with panel effects indicate that most selected attributes of the residential environment, job profile and transportation mode are significantly related to individual differences in multidimensional choices. Moreover, the estimation of various sources of unobserved heterogeneity signals significant unobserved heterogeneity in selected taste parameters, and choice dependent heteroscedasticity in error component variance.},
keywords = {Co-dependent location, Error component logit model, Heterogeneity, Multidimensional choice, Panel data, Pivoted stated choice experiment},
pubstate = {published},
tppubtype = {article}
}
This paper develops an error component mixed logit model to analyze the multi-dimensional residential, work and transportation mode choice. It expanse previous studies based on life-trajectory theory which predominantly only considered two life domains. In contributing to this emerging field of research, we design an integrated pivoted stated choice experiment considering the multi-dimensional choice of job, residence and transportation mode for the journey to work. The results of the estimated error component mixed logit model with panel effects indicate that most selected attributes of the residential environment, job profile and transportation mode are significantly related to individual differences in multidimensional choices. Moreover, the estimation of various sources of unobserved heterogeneity signals significant unobserved heterogeneity in selected taste parameters, and choice dependent heteroscedasticity in error component variance.
2019
Jiang, W.; Feng, T.; Timmermans, H.; Li, H.
Intention to move in renovated historical blocks in China Journal Article
In: Journal of Housing and the Built Environment, vol. 34, iss. 1, 2019, ISSN: 15737772.
Abstract | Links | BibTeX | Tags: China, Heterogeneity, Intention to move, Renovated historical block, Residential satisfaction
@article{Jiang2019,
title = {Intention to move in renovated historical blocks in China},
author = {W. Jiang and T. Feng and H. Timmermans and H. Li},
doi = {10.1007/s10901-018-9608-3},
issn = {15737772},
year = {2019},
date = {2019-01-01},
journal = {Journal of Housing and the Built Environment},
volume = {34},
issue = {1},
abstract = {Using data from 8 renovated historical blocks in China, the influence of social-demographic characteristics and residential satisfaction on the intention to move is analysed using multinomial logit models. To explore the observed and unobserved heterogeneity among individuals, a mixed logit model was adopted. The model results indicate that older residents and residents who live in blocks during the middle renovation stage are less interested in relocating. Higher residential satisfaction will lead to a lower intention to move house, and housing satisfaction and satisfaction with the living environment have a higher impact compared with satisfaction related to other residential dimensions. Moreover, the interaction between neighbourhood satisfaction and renovation stages was found influential, while the unobserved taste variation between residents was not identified.},
keywords = {China, Heterogeneity, Intention to move, Renovated historical block, Residential satisfaction},
pubstate = {published},
tppubtype = {article}
}
Using data from 8 renovated historical blocks in China, the influence of social-demographic characteristics and residential satisfaction on the intention to move is analysed using multinomial logit models. To explore the observed and unobserved heterogeneity among individuals, a mixed logit model was adopted. The model results indicate that older residents and residents who live in blocks during the middle renovation stage are less interested in relocating. Higher residential satisfaction will lead to a lower intention to move house, and housing satisfaction and satisfaction with the living environment have a higher impact compared with satisfaction related to other residential dimensions. Moreover, the interaction between neighbourhood satisfaction and renovation stages was found influential, while the unobserved taste variation between residents was not identified.
2013
Feng, T.; Arentze, T.; Timmermans, H.
Capturing preference heterogeneity of truck drivers’ route choice behavior with context effects using a latent class model Journal Article
In: European Journal of Transport and Infrastructure Research, vol. 13, iss. 4, 2013, ISSN: 15677141.
Abstract | Links | BibTeX | Tags: Context effects, Freight transport, Heterogeneity, Latent class model, Route choice behavior
@article{Feng2013,
title = {Capturing preference heterogeneity of truck drivers' route choice behavior with context effects using a latent class model},
author = {T. Feng and T. Arentze and H. Timmermans},
doi = {10.18757/ejtir.2013.13.4.3004},
issn = {15677141},
year = {2013},
date = {2013-01-01},
journal = {European Journal of Transport and Infrastructure Research},
volume = {13},
issue = {4},
abstract = {This paper investigates heterogeneity in truck drivers' route choice preferences. A latent class model is estimated to identify heterogeneous segments of drivers. A stated choice experiment designed for identifying route choice behavior of truck drivers provides the data for model estimation. The effects of road pricing and environmental bonus are examined considering context dependency. Results reveal that size of truck is a significant segmentation variable of preferences for route attributes. Drivers of light trucks care more about congestion than drivers of heavy trucks, and are highly sensitive to road pricing and slightly sensitive to a road bonus. Drivers of heavy trucks are more sensitive to road category and urban area than drivers of light trucks, and are insensitive to bonus and slightly sensitive to pricing.},
keywords = {Context effects, Freight transport, Heterogeneity, Latent class model, Route choice behavior},
pubstate = {published},
tppubtype = {article}
}
This paper investigates heterogeneity in truck drivers’ route choice preferences. A latent class model is estimated to identify heterogeneous segments of drivers. A stated choice experiment designed for identifying route choice behavior of truck drivers provides the data for model estimation. The effects of road pricing and environmental bonus are examined considering context dependency. Results reveal that size of truck is a significant segmentation variable of preferences for route attributes. Drivers of light trucks care more about congestion than drivers of heavy trucks, and are highly sensitive to road pricing and slightly sensitive to a road bonus. Drivers of heavy trucks are more sensitive to road category and urban area than drivers of light trucks, and are insensitive to bonus and slightly sensitive to pricing.