Pub Date : 2025-08-09DOI: 10.1080/19427867.2024.2436168
Wei Li , Ruicai Peng , Qin Luo
Optimizing the last train schedule is challenged by the elastic nature of passenger flow and the immutability of the schedule once announced to the public. The elastic passenger demand is first mathematically described, and an optimization method is proposed to minimize the adjustment to the last trains while maximizing the accessible passengers within the metro network. Linearization techniques are then employed to transform it into a linear programming form. The proposed method is verified through a real-world case study, which indicate an 8.4% increase in accessible passenger flow on a weekday with adjustments required for only 3 lines and 10 stations’ last train times. In a special event scenario, a 9.8% is achieved with adjustments to 2 lines and 8 stations’ last train times. The results suggest that for existing metro systems, judiciously adjusting the last train times can effectively enhance the overall level of nighttime operational service.
{"title":"Last train schedule optimization for metro systems considering minimum adjustment cost under elastic passenger demand","authors":"Wei Li , Ruicai Peng , Qin Luo","doi":"10.1080/19427867.2024.2436168","DOIUrl":"10.1080/19427867.2024.2436168","url":null,"abstract":"<div><div>Optimizing the last train schedule is challenged by the elastic nature of passenger flow and the immutability of the schedule once announced to the public. The elastic passenger demand is first mathematically described, and an optimization method is proposed to minimize the adjustment to the last trains while maximizing the accessible passengers within the metro network. Linearization techniques are then employed to transform it into a linear programming form. The proposed method is verified through a real-world case study, which indicate an 8.4% increase in accessible passenger flow on a weekday with adjustments required for only 3 lines and 10 stations’ last train times. In a special event scenario, a 9.8% is achieved with adjustments to 2 lines and 8 stations’ last train times. The results suggest that for existing metro systems, judiciously adjusting the last train times can effectively enhance the overall level of nighttime operational service.</div></div>","PeriodicalId":48974,"journal":{"name":"Transportation Letters-The International Journal of Transportation Research","volume":"17 7","pages":"Pages 1302-1319"},"PeriodicalIF":3.3,"publicationDate":"2025-08-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145108938","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-08-09DOI: 10.1080/19427867.2024.2437849
Kadir Akgol , Emre Demir , Ibrahim Aydogdu , Yetis Sazi Murat
Transit users generally travel between different pairs of stops on the transit line. Therefore, decision-makers should consider the route designs for each passenger’s benefit. In this study, the factor of the number of public transit passengers is combined with trip coefficients such as the route circuity coefficient used to determine the efficiency of public transportation (PT) vehicles. The flow direction performance (FDP), a novel criterion based on the circuity coefficient in PT, is developed. Accordingly, the flow direction method (FDM) a new hybrid metaheuristic optimization method that determines the optimum PT routes, is created. FDM calculates the optimum routes between several predetermined origin points and terminal points. Both sequences of stations and the transit route are optimized at the same time. We tested FDM’s performance on 2 benchmark and 3 real case network samples. Test results show the effectiveness of FDP and robustness of FDM on multi-route problems.
{"title":"Efficiency in public transportation: a new flow direction method for optimizing multi-route networks","authors":"Kadir Akgol , Emre Demir , Ibrahim Aydogdu , Yetis Sazi Murat","doi":"10.1080/19427867.2024.2437849","DOIUrl":"10.1080/19427867.2024.2437849","url":null,"abstract":"<div><div>Transit users generally travel between different pairs of stops on the transit line. Therefore, decision-makers should consider the route designs for each passenger’s benefit. In this study, the factor of the number of public transit passengers is combined with trip coefficients such as the route circuity coefficient used to determine the efficiency of public transportation (PT) vehicles. The flow direction performance (FDP), a novel criterion based on the circuity coefficient in PT, is developed. Accordingly, the flow direction method (FDM) a new hybrid metaheuristic optimization method that determines the optimum PT routes, is created. FDM calculates the optimum routes between several predetermined origin points and terminal points. Both sequences of stations and the transit route are optimized at the same time. We tested FDM’s performance on 2 benchmark and 3 real case network samples. Test results show the effectiveness of FDP and robustness of FDM on multi-route problems.</div></div>","PeriodicalId":48974,"journal":{"name":"Transportation Letters-The International Journal of Transportation Research","volume":"17 7","pages":"Pages 1320-1332"},"PeriodicalIF":3.3,"publicationDate":"2025-08-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145108931","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-08-09DOI: 10.1080/19427867.2024.2416314
Kai Liu , Yuan Xu
This study proposes a novel instrument for measuring private car owners’ travel-related lifestyle (PCTRL) based on a survey (N = 2045) conducted in China’s four first-tier cities and ten quasi first-tier cities. By employing principal component analysis (PCA) and cluster analysis, the study classifies electric vhehicle (EV) owners and internal combustion engine vehicle (ICEV) owners into five distinct PCTRL segments, respectively. The findings reveal that only 66% of EV owners have similar PCTRLs to 67% of ICEV owners, with the most notable disparity being the higher likelihood of EV owners choosing private cars for their daily commutes. Furthermore, a random forest model was developed to predict the daily activity-based travel mode choices behavior of both EV and ICEV owners. The findings offer valuable insights into understanding the behavior of EV and ICEV owners in their daily travel mode choices and help formulating more effective traffic management strategies.
{"title":"Exploring disparities and similarities in daily travel mode choices between electric vehicle owners and internal combustion engine vehicle owners","authors":"Kai Liu , Yuan Xu","doi":"10.1080/19427867.2024.2416314","DOIUrl":"10.1080/19427867.2024.2416314","url":null,"abstract":"<div><div>This study proposes a novel instrument for measuring private car owners’ travel-related lifestyle (PCTRL) based on a survey (<em>N</em> = 2045) conducted in China’s four first-tier cities and ten quasi first-tier cities. By employing principal component analysis (PCA) and cluster analysis, the study classifies electric vhehicle (EV) owners and internal combustion engine vehicle (ICEV) owners into five distinct PCTRL segments, respectively. The findings reveal that only 66% of EV owners have similar PCTRLs to 67% of ICEV owners, with the most notable disparity being the higher likelihood of EV owners choosing private cars for their daily commutes. Furthermore, a random forest model was developed to predict the daily activity-based travel mode choices behavior of both EV and ICEV owners. The findings offer valuable insights into understanding the behavior of EV and ICEV owners in their daily travel mode choices and help formulating more effective traffic management strategies.</div></div>","PeriodicalId":48974,"journal":{"name":"Transportation Letters-The International Journal of Transportation Research","volume":"17 7","pages":"Pages 1155-1170"},"PeriodicalIF":3.3,"publicationDate":"2025-08-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145108935","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-08-09DOI: 10.1080/19427867.2024.2417150
Kayo Kinjo , Akiyasu Tomoeda
Autonomous vehicles are essential to future transportation systems, potentially reducing traffic congestion. This study examines the impact of different vehicle control strategies on traffic flow through simulations. We propose a novel stochastic cellular automaton model, the controlled stochastic optimal velocity (CSOV) model, which incorporates vehicle control effects. Within the CSOV model, two control strategies are implemented: gap-based control (GC), which adjusts vehicle velocity to balance the gaps between adjacent vehicles, and flow-based control (FC), which aims to maintain a consistent local flow between the front and rear vehicles. Results show that both control strategies improve traffic flow. However, under weaker control, the GC sometimes resulted in lower flow compared to no control. In contrast, the FC consistently enhanced flow across control strengths, yielding more robust outcomes. Furthermore, when both strategies achieved comparable flow rates, the FC provided a more stable velocity distribution under varying traffic densities than the GC.
{"title":"Comparison of gap-based and flow-based control strategies using a new controlled stochastic cellular automaton model for traffic flow","authors":"Kayo Kinjo , Akiyasu Tomoeda","doi":"10.1080/19427867.2024.2417150","DOIUrl":"10.1080/19427867.2024.2417150","url":null,"abstract":"<div><div>Autonomous vehicles are essential to future transportation systems, potentially reducing traffic congestion. This study examines the impact of different vehicle control strategies on traffic flow through simulations. We propose a novel stochastic cellular automaton model, the controlled stochastic optimal velocity (CSOV) model, which incorporates vehicle control effects. Within the CSOV model, two control strategies are implemented: gap-based control (GC), which adjusts vehicle velocity to balance the gaps between adjacent vehicles, and flow-based control (FC), which aims to maintain a consistent local flow between the front and rear vehicles. Results show that both control strategies improve traffic flow. However, under weaker control, the GC sometimes resulted in lower flow compared to no control. In contrast, the FC consistently enhanced flow across control strengths, yielding more robust outcomes. Furthermore, when both strategies achieved comparable flow rates, the FC provided a more stable velocity distribution under varying traffic densities than the GC.</div></div>","PeriodicalId":48974,"journal":{"name":"Transportation Letters-The International Journal of Transportation Research","volume":"17 7","pages":"Pages 1171-1181"},"PeriodicalIF":3.3,"publicationDate":"2025-08-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145109002","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-08-09DOI: 10.1080/19427867.2024.2426795
Hao Tong , Chengcheng Xu , Qi Ai , Weilin Ren , Changshuai Wang , Chang Peng , Yanli Jiao
Jam-absorption driving (JAD) can effectively prevent the generation and propagation of traffic oscillation. To alleviate the traffic congestion in the signalized intersection with mixed traffic flow, including human driving vehicles (HDVs) and connected and automated vehicles (CAVs), this study provides a jam-absorption driving strategy based on the traffic delay prediction of the mixed platoon under traffic congestion. An online traffic congestion prediction method with the objective of JAD is proposed and focuses on the leaving state of the trajectory to achieve fast capture of congestion features. Then, with real-time status and prediction information, we develop a Jam-absorption driving strategy based on a deep reinforcement learning (DRL) model to improve adaptability to the mixed traffic environment. The results show that this strategy can suppress more than 70% of traffic oscillations with excellent execution efficiency, improving traffic safety and efficiency.
{"title":"Developing a jam-absorption strategy for mixed traffic flow at signalized intersections using deep reinforcement learning","authors":"Hao Tong , Chengcheng Xu , Qi Ai , Weilin Ren , Changshuai Wang , Chang Peng , Yanli Jiao","doi":"10.1080/19427867.2024.2426795","DOIUrl":"10.1080/19427867.2024.2426795","url":null,"abstract":"<div><div>Jam-absorption driving (JAD) can effectively prevent the generation and propagation of traffic oscillation. To alleviate the traffic congestion in the signalized intersection with mixed traffic flow, including human driving vehicles (HDVs) and connected and automated vehicles (CAVs), this study provides a jam-absorption driving strategy based on the traffic delay prediction of the mixed platoon under traffic congestion. An online traffic congestion prediction method with the objective of JAD is proposed and focuses on the leaving state of the trajectory to achieve fast capture of congestion features. Then, with real-time status and prediction information, we develop a Jam-absorption driving strategy based on a deep reinforcement learning (DRL) model to improve adaptability to the mixed traffic environment. The results show that this strategy can suppress more than 70% of traffic oscillations with excellent execution efficiency, improving traffic safety and efficiency.</div></div>","PeriodicalId":48974,"journal":{"name":"Transportation Letters-The International Journal of Transportation Research","volume":"17 7","pages":"Pages 1251-1262"},"PeriodicalIF":3.3,"publicationDate":"2025-08-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145108930","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-08-09DOI: 10.1080/19427867.2024.2439349
Qiyuan Jiang , Yizheng Wu , Jian Sun , Yuxin Wang , Shuochen Zhang , Lewen Wang , Weinan He
In 2023, Beijing adjusted dedicated bus lane (DBL) policies, allowing private car access during specific periods to improve road transport efficiency. This study proposes a comprehensive method to assess the operational and environmental impact of DBL adjustments. Traffic volumes are estimated using a localized traffic fundamental diagram model based on large-scale floating car speed data, while vehicle emissions are quantified employing a high-resolution road network emission inventory. Results show over 20% increased traffic volumes and 10% higher average speed on DBLs in central Beijing. Extended to regional road network, the adjustments enhance expressway capacity and improve traffic efficiency on main and minor arterial roads. While total emissions on expressways increase due to heightened traffic volumes, average vehicular emission intensities in urban areas decrease because of smoother traffic flow. These findings could provide valuable insights for decision-makers with the information needed to target reasonable DBL policies in metropolitan regions.
{"title":"Is a dedicated bus lane operationally and environmentally beneficial? A case study in Beijing","authors":"Qiyuan Jiang , Yizheng Wu , Jian Sun , Yuxin Wang , Shuochen Zhang , Lewen Wang , Weinan He","doi":"10.1080/19427867.2024.2439349","DOIUrl":"10.1080/19427867.2024.2439349","url":null,"abstract":"<div><div>In 2023, Beijing adjusted dedicated bus lane (DBL) policies, allowing private car access during specific periods to improve road transport efficiency. This study proposes a comprehensive method to assess the operational and environmental impact of DBL adjustments. Traffic volumes are estimated using a localized traffic fundamental diagram model based on large-scale floating car speed data, while vehicle emissions are quantified employing a high-resolution road network emission inventory. Results show over 20% increased traffic volumes and 10% higher average speed on DBLs in central Beijing. Extended to regional road network, the adjustments enhance expressway capacity and improve traffic efficiency on main and minor arterial roads. While total emissions on expressways increase due to heightened traffic volumes, average vehicular emission intensities in urban areas decrease because of smoother traffic flow. These findings could provide valuable insights for decision-makers with the information needed to target reasonable DBL policies in metropolitan regions.</div></div>","PeriodicalId":48974,"journal":{"name":"Transportation Letters-The International Journal of Transportation Research","volume":"17 7","pages":"Pages 1333-1347"},"PeriodicalIF":3.3,"publicationDate":"2025-08-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145108940","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-08-09DOI: 10.1080/19427867.2024.2430109
Md Istiak Jahan , Tanmoy Bhowmik , Sachraa G. Borjigin , Jiehong Lou , Nneoma M. Ugwu , Deb A. Niemeier , Naveen Eluru
The growing adoption of electric vehicles offers a potential opportunity to reduce transportation sector carbon footprint. In our research, we studied vehicle purchase behavior with emphasis on alternative fuel vehicles using the vehicle purchase dataset ‘MaritzCX New Vehicle Customer Study.’ This study consisted of a two-level modeling approach. In the first level, purchasing of a new car was estimated based on consumers socio-economic characteristics. In the second level, the vehicle purchase decision was examined with a two-dimensional dependent variable – vehicle type and fuel type. We employed an innovative data fusion approach that probabilistically links records from MaritzCX with records from National Household Travel Survey with the objective of identifying new independent variables affecting the decision process while maximizing data fit. The final model included a host of independent variables from four different categories: vehicle-, economic-, demographic-, and spatial characteristics. Finally, the model results were employed to conduct an elasticity analysis.
{"title":"A maximum log-likelihood based data fusion model for estimating household’s vehicle purchase decision","authors":"Md Istiak Jahan , Tanmoy Bhowmik , Sachraa G. Borjigin , Jiehong Lou , Nneoma M. Ugwu , Deb A. Niemeier , Naveen Eluru","doi":"10.1080/19427867.2024.2430109","DOIUrl":"10.1080/19427867.2024.2430109","url":null,"abstract":"<div><div>The growing adoption of electric vehicles offers a potential opportunity to reduce transportation sector carbon footprint. In our research, we studied vehicle purchase behavior with emphasis on alternative fuel vehicles using the vehicle purchase dataset ‘MaritzCX New Vehicle Customer Study.’ This study consisted of a two-level modeling approach. In the first level, purchasing of a new car was estimated based on consumers socio-economic characteristics. In the second level, the vehicle purchase decision was examined with a two-dimensional dependent variable – vehicle type and fuel type. We employed an innovative data fusion approach that probabilistically links records from MaritzCX with records from National Household Travel Survey with the objective of identifying new independent variables affecting the decision process while maximizing data fit. The final model included a host of independent variables from four different categories: vehicle-, economic-, demographic-, and spatial characteristics. Finally, the model results were employed to conduct an elasticity analysis.</div></div>","PeriodicalId":48974,"journal":{"name":"Transportation Letters-The International Journal of Transportation Research","volume":"17 7","pages":"Pages 1263-1279"},"PeriodicalIF":3.3,"publicationDate":"2025-08-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145108937","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-08-09DOI: 10.1080/19427867.2024.2433337
Mahsa Ghaffari Targhi , Mohammad Ansari Esfeh , Adam Weiss , Lina Kattan
This study investigates perceptions and attitudes toward autonomous vehicles (AVs) using an online stated preference (SP) survey conducted in Alberta, Canada. It explores the effect of different sociodemographic, external, and psychological factors on users’ attitudes toward AVs. Additionally, factors contributing to people’s willingness to pay for AVs were evaluated. The results indicate that sociodemographic factors, external factors, and people’s perceptions significantly affect people’s willingness to pay for automation. Level 3 of automation is shown to have a positive effect on the drivers’ utility of driving for commuting and non-commuting trips, while other levels of automation were found negatively affecting the utility of driving. Men were generally more willing to pay for AVs, particularly for commuting trips, while weather conditions, especially icy roads, posed significant concerns about AV reliability. Middle-aged drivers exhibited the highest willingness to pay (WTP) for higher levels of automation, emphasizing the potential early adoption among this group.
{"title":"Users’ perceptions toward autonomous vehicles: case study in Alberta, Canada","authors":"Mahsa Ghaffari Targhi , Mohammad Ansari Esfeh , Adam Weiss , Lina Kattan","doi":"10.1080/19427867.2024.2433337","DOIUrl":"10.1080/19427867.2024.2433337","url":null,"abstract":"<div><div>This study investigates perceptions and attitudes toward autonomous vehicles (AVs) using an online stated preference (SP) survey conducted in Alberta, Canada. It explores the effect of different sociodemographic, external, and psychological factors on users’ attitudes toward AVs. Additionally, factors contributing to people’s willingness to pay for AVs were evaluated. The results indicate that sociodemographic factors, external factors, and people’s perceptions significantly affect people’s willingness to pay for automation. Level 3 of automation is shown to have a positive effect on the drivers’ utility of driving for commuting and non-commuting trips, while other levels of automation were found negatively affecting the utility of driving. Men were generally more willing to pay for AVs, particularly for commuting trips, while weather conditions, especially icy roads, posed significant concerns about AV reliability. Middle-aged drivers exhibited the highest willingness to pay (WTP) for higher levels of automation, emphasizing the potential early adoption among this group.</div></div>","PeriodicalId":48974,"journal":{"name":"Transportation Letters-The International Journal of Transportation Research","volume":"17 7","pages":"Pages 1280-1301"},"PeriodicalIF":3.3,"publicationDate":"2025-08-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145108939","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-08-09DOI: 10.1080/19427867.2024.2422717
Kathleen Salazar-Serna , Sergio A. Barona , Isabel C. García , Lorena Cadavid , Carlos J. Franco
Overfitting poses a significant limitation in mode choice prediction using classification models, often worsened by the proliferation of features from encoding categorical variables. While dimensionality reduction techniques are widely utilized, their effects on travel-mode choice models’ performance have yet to be comparatively studied. This research compares the impact of dimensionality reduction methods (PCA, CATPCA, FAMD, LDA) on the performance of multinomial models and various supervised learning classifiers (XGBoost, Random Forest, Naive Bayes, K-Nearest Neighbors, Multinomial Logit) for predicting travel mode choice. Utilizing survey data from the Aburrá Valley in Colombia, we detail the process of analyzing derived dimensions and selecting optimal models for both overall and class-specific predictions. Results indicate that dimension reduction enhances predictive power, particularly for less common transport modes, providing a strategy to address class imbalance without modifying data distribution. This methodology deepens understanding of travel behavior, offering valuable insights for modelers and policymakers in developing regions with similar characteristics.
{"title":"Addressing overfitting in classification models for transport mode choice prediction: a practical application in the Aburrá Valley, Colombia","authors":"Kathleen Salazar-Serna , Sergio A. Barona , Isabel C. García , Lorena Cadavid , Carlos J. Franco","doi":"10.1080/19427867.2024.2422717","DOIUrl":"10.1080/19427867.2024.2422717","url":null,"abstract":"<div><div>Overfitting poses a significant limitation in mode choice prediction using classification models, often worsened by the proliferation of features from encoding categorical variables. While dimensionality reduction techniques are widely utilized, their effects on travel-mode choice models’ performance have yet to be comparatively studied. This research compares the impact of dimensionality reduction methods (PCA, CATPCA, FAMD, LDA) on the performance of multinomial models and various supervised learning classifiers (XGBoost, Random Forest, Naive Bayes, K-Nearest Neighbors, Multinomial Logit) for predicting travel mode choice. Utilizing survey data from the Aburrá Valley in Colombia, we detail the process of analyzing derived dimensions and selecting optimal models for both overall and class-specific predictions. Results indicate that dimension reduction enhances predictive power, particularly for less common transport modes, providing a strategy to address class imbalance without modifying data distribution. This methodology deepens understanding of travel behavior, offering valuable insights for modelers and policymakers in developing regions with similar characteristics.</div></div>","PeriodicalId":48974,"journal":{"name":"Transportation Letters-The International Journal of Transportation Research","volume":"17 7","pages":"Pages 1213-1230"},"PeriodicalIF":3.3,"publicationDate":"2025-08-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145109066","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-08-09DOI: 10.1080/19427867.2024.2422713
Jorge Diaz-Gutierrez , Andisheh Ranjbari
Transit agencies use direct demand models (DDM) to allocate services. Since the service supply – a crucial predictor in DDMs – is endogenous to demand, including it in the model might yield biased estimations. A widely used methodology that is believed to handle this issue is Fixed Effects (FE). However, the underlying assumptions of FE are valid only if service adjustments take a considerable amount of time. This study investigates the performance of FE for estimating transit ridership. We collected 2013–2019 data and constructed 16 DDMs, employing four methodologies with a shared set of variables. We found that FE has significant limitations in handling endogeneity and will result in parameter estimates that significantly differ from those produced by methodologies that are specifically designed to control for endogeneity (such as FE-IV). Moreover, the use of FE leads to the omission of certain predictors and inaccurate ridership predictions, misguiding agencies as to what changes to implement and potentially impacting revenue projections.
{"title":"How effective are fixed-effects models in fixing the transit supply–demand bidirectional interaction?","authors":"Jorge Diaz-Gutierrez , Andisheh Ranjbari","doi":"10.1080/19427867.2024.2422713","DOIUrl":"10.1080/19427867.2024.2422713","url":null,"abstract":"<div><div>Transit agencies use direct demand models (DDM) to allocate services. Since the service supply – a crucial predictor in DDMs – is endogenous to demand, including it in the model might yield biased estimations. A widely used methodology that is believed to handle this issue is Fixed Effects (FE). However, the underlying assumptions of FE are valid only if service adjustments take a considerable amount of time. This study investigates the performance of FE for estimating transit ridership. We collected 2013–2019 data and constructed 16 DDMs, employing four methodologies with a shared set of variables. We found that FE has significant limitations in handling endogeneity and will result in parameter estimates that significantly differ from those produced by methodologies that are specifically designed to control for endogeneity (such as FE-IV). Moreover, the use of FE leads to the omission of certain predictors and inaccurate ridership predictions, misguiding agencies as to what changes to implement and potentially impacting revenue projections.</div></div>","PeriodicalId":48974,"journal":{"name":"Transportation Letters-The International Journal of Transportation Research","volume":"17 7","pages":"Pages 1199-1212"},"PeriodicalIF":3.3,"publicationDate":"2025-08-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145108936","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号