Aref Darzi
Aref Darzi has a total of 35 co-authors. Here are the top co-authors by number of shared publications:
- Yixuan Pan
Graduate Research
8 shared publications - Mofeng Yang
Department of Civil and Environmental Engineering, University of Maryland, College Park, MD
7 shared publications - Chenfeng Xiong
Villanova University
6 shared publications - Sepehr Ghader
Social Sciences Maryland Transportation Institute (MTI) University of Maryland 20742 College Park MD
6 shared publications - Lei Zhang
Maryland Transportation Institute, University of Maryland, College Park, MD 20742, USA
5 shared publications - Aliakbar Kabiri
Department of Civil and Environmental Engineering, University of Maryland, College Park, MD
4 shared publications
Latest Publications
- Open Access
- 25 April 2025
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 20  (4) , e0321970
Examining trip-level errors in passively collected mobile device data for data quality assurance
Location-based service (LBS) data passively collected by mobile devices has been widely adopted in multiple fields for its advantages in revealing travel behaviors. Data quality assessments have always been important steps for analyses using the data, but the impact of trip-level errors has not been a focus of these assessments. We examine a newly emerged type of error present at trip-level in LBS datasets that violates the spatio-temporal consistency of such data by including trips on road segments where and when there should be no trips. We designed a distributed-computing workflow to quantify the errors by comparing the number of trips on closed road segments during road closures with time periods before and after. Using two real-world cases from 2023, we examined multiple datasets acquired from major vendors in the US, and several of the datasets contained a significant number of trip-level errors. These findings point to the errors being present in recent datasets that have not otherwise been processed for data quality and can significantly impact analyses by data users. Data users should consider conducting trip-level error data quality checks as part of their preprocessing steps.
- 25 April 2025
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 20  (4) , e0321970
Open AccessExamining trip-level errors in passively collected mobile device data for data quality assurance
Location-based service (LBS) data passively collected by mobile devices has been widely adopted in multiple fields for its advantages in revealing travel behaviors. Data quality assessments have always been important steps for analyses using the data, but the impact of trip-level errors has not been a focus of these assessments. We examine a newly emerged type of error present at trip-level in LBS datasets that violates the spatio-temporal consistency of such data by including trips on road segments where and when there should be no trips. We designed a distributed-computing workflow to quantify the errors by comparing the number of trips on closed road segments during road closures with time periods before and after. Using two real-world cases from 2023, we examined multiple datasets acquired from major vendors in the US, and several of the datasets contained a significant number of trip-level errors. These findings point to the errors being present in recent datasets that have not otherwise been processed for data quality and can significantly impact analyses by data users. Data users should consider conducting trip-level error data quality checks as part of their preprocessing steps.
- 3 March 2024
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
- Vol. 2678  (12) , 687-701
Using Geographically Weighted Models to Explore Temporal and Spatial Varying Impacts on Commute Trip Change Resulting from COVID-19
COVID-19 deeply affected people’s daily life and travel behaviors. Comprehending changes in travel behavior holds significant importance, making it imperative to investigate the influential factors of sociodemographics and socioeconomics on such behavior. This study used large-scale mobile device location data at the U.S. county level in the Washington, D.C., Maryland, and Virginia (DMV) area, U.S., to reveal the impacts of demographic and socioeconomic variables on commute trip change. The study investigated the impact of these variables on commuter trips over time and space. It reflected the short- and long-term impact of COVID-19 on travel behavior via linear regression and geographically weighted regression (GWR) models. The findings indicated that counties with a higher percentage of people using walking and biking (active mode) for commuting during the initial phase of COVID-19 experienced a greater reduction in their commute trips compared with others. Conversely, for the long-term effect of COVID-19 in November 2020, we can see the impact of using active mode on trip change is not significant any more and, instead, results showed people who were using bus and rail (public mode) for commuting decreased their trips more than others. Additionally, a positive correlation was observed between median income levels and the reduction in commute trips. On the other hand, sectors that necessitated ongoing outdoor operations during the pandemic, such as manufacturing, wholesale trade, and food services, showed a substantial negative correlation with trip change. Moreover, in the DMV area, counties with a higher proportion of Democrat voters experienced less trip reduction than others. Notably, by applying the GWR and multiscale GWR models, the local spatial relationships of variables and commuting behaviors were captured. The results showed the emergence of local correlations as the pandemic evolved, suggesting a geographical impact pattern. At the onset of COVID-19, the pandemic’s impact on commuting behaviors was global. However, as time passed, travel behavior became more influenced by spatial factors and started to show localized effects.
- 3 March 2024
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
- Vol. 2678  (12) , 687-701
Using Geographically Weighted Models to Explore Temporal and Spatial Varying Impacts on Commute Trip Change Resulting from COVID-19
COVID-19 deeply affected people’s daily life and travel behaviors. Comprehending changes in travel behavior holds significant importance, making it imperative to investigate the influential factors of sociodemographics and socioeconomics on such behavior. This study used large-scale mobile device location data at the U.S. county level in the Washington, D.C., Maryland, and Virginia (DMV) area, U.S., to reveal the impacts of demographic and socioeconomic variables on commute trip change. The study investigated the impact of these variables on commuter trips over time and space. It reflected the short- and long-term impact of COVID-19 on travel behavior via linear regression and geographically weighted regression (GWR) models. The findings indicated that counties with a higher percentage of people using walking and biking (active mode) for commuting during the initial phase of COVID-19 experienced a greater reduction in their commute trips compared with others. Conversely, for the long-term effect of COVID-19 in November 2020, we can see the impact of using active mode on trip change is not significant any more and, instead, results showed people who were using bus and rail (public mode) for commuting decreased their trips more than others. Additionally, a positive correlation was observed between median income levels and the reduction in commute trips. On the other hand, sectors that necessitated ongoing outdoor operations during the pandemic, such as manufacturing, wholesale trade, and food services, showed a substantial negative correlation with trip change. Moreover, in the DMV area, counties with a higher proportion of Democrat voters experienced less trip reduction than others. Notably, by applying the GWR and multiscale GWR models, the local spatial relationships of variables and commuting behaviors were captured. The results showed the emergence of local correlations as the pandemic evolved, suggesting a geographical impact pattern. At the onset of COVID-19, the pandemic’s impact on commuting behaviors was global. However, as time passed, travel behavior became more influenced by spatial factors and started to show localized effects.
- Open Access
Using Geographically Weighted Models to Explore Temporal and Spatial Varying Impacts on Commute Trip Change Due to Covid-19
COVID-19 has deeply affected daily life and travel behaviors. Understanding these changes is crucial, prompting an investigation into socio-demographic and socio-economic factors. This study used large-scale mobile device location data in Washington, D.C., Maryland, and Virginia (DMV area) to unveil the impacts of these variables on commute trip changes. It reflected short and long-term impacts through linear regression and geographically weighted regression models. Findings indicated that counties with a higher percentage of people using walking and biking during the initial phase of COVID-19 experienced greater reductions in commute trips. For the long-term effect in November, the impact of active modes became insignificant, and individuals using public modes showed more significant trip reductions. Positive correlations were observed between median income levels and reduced commute trips. Sectors requiring ongoing outdoor operations during the pandemic showed substantial negative correlations. In the DMV area, counties with a higher proportion of Democratic voters experienced less trip reduction. Applying Geographically Weighted Regression models captured local spatial relationships, showing the emergence of local correlations as the pandemic evolved, suggesting a geographical impact pattern. Initially global, the pandemic's impact on commuting behaviors became more influenced by spatial factors over time, showing localized effects.
Open AccessUsing Geographically Weighted Models to Explore Temporal and Spatial Varying Impacts on Commute Trip Change Due to Covid-19
COVID-19 has deeply affected daily life and travel behaviors. Understanding these changes is crucial, prompting an investigation into socio-demographic and socio-economic factors. This study used large-scale mobile device location data in Washington, D.C., Maryland, and Virginia (DMV area) to unveil the impacts of these variables on commute trip changes. It reflected short and long-term impacts through linear regression and geographically weighted regression models. Findings indicated that counties with a higher percentage of people using walking and biking during the initial phase of COVID-19 experienced greater reductions in commute trips. For the long-term effect in November, the impact of active modes became insignificant, and individuals using public modes showed more significant trip reductions. Positive correlations were observed between median income levels and reduced commute trips. Sectors requiring ongoing outdoor operations during the pandemic showed substantial negative correlations. In the DMV area, counties with a higher proportion of Democratic voters experienced less trip reduction. Applying Geographically Weighted Regression models captured local spatial relationships, showing the emergence of local correlations as the pandemic evolved, suggesting a geographical impact pattern. Initially global, the pandemic's impact on commuting behaviors became more influenced by spatial factors over time, showing localized effects.
- 17 November 2023
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
- Vol. 2678  (6) , 881-890
Elaborated Framework for Duplicate Device Detection from Multisourced Mobile Device Location Data
Mobile device location data (MDLD) have been popularly utilized in various fields. Yet large-scale applications are limited because of either biased or insufficient spatial coverage of the data from individual data vendors. One approach to improve the data coverage is to leverage the data from different data vendors and integrate them to build a more representative dataset. To extract reliable statistics from MDLD, certain data preprocessing steps are crucial to ensure the accuracy of the analysis. One of these steps is the development of a framework to remove duplicated devices or several devices that belong to the same data subject. This treatment is especially necessary when using a multiplicity of data sources, as the same device may be captured by more than one data provider. We propose a data integration methodology for multisourced data to investigate the feasibility of integrating data from several sources. By leveraging the uniqueness of travel pattern of each device, duplicate devices are identified. The proposed methodology is shown to be cost-effective through a national-level analysis. The method is successfully applied to a dataset from January 2020 consisting of more than 270 million raw devices nationwide. Our findings suggest that devices sharing the same imputed home location and the same top-five most-visited locations during a month can represent the same user in the MDLD. It is shown that more than 99.6% of the sample devices having the aforementioned attribute in common are observed at the same location simultaneously.
- 17 November 2023
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
- Vol. 2678  (6) , 881-890
Elaborated Framework for Duplicate Device Detection from Multisourced Mobile Device Location Data
Mobile device location data (MDLD) have been popularly utilized in various fields. Yet large-scale applications are limited because of either biased or insufficient spatial coverage of the data from individual data vendors. One approach to improve the data coverage is to leverage the data from different data vendors and integrate them to build a more representative dataset. To extract reliable statistics from MDLD, certain data preprocessing steps are crucial to ensure the accuracy of the analysis. One of these steps is the development of a framework to remove duplicated devices or several devices that belong to the same data subject. This treatment is especially necessary when using a multiplicity of data sources, as the same device may be captured by more than one data provider. We propose a data integration methodology for multisourced data to investigate the feasibility of integrating data from several sources. By leveraging the uniqueness of travel pattern of each device, duplicate devices are identified. The proposed methodology is shown to be cost-effective through a national-level analysis. The method is successfully applied to a dataset from January 2020 consisting of more than 270 million raw devices nationwide. Our findings suggest that devices sharing the same imputed home location and the same top-five most-visited locations during a month can represent the same user in the MDLD. It is shown that more than 99.6% of the sample devices having the aforementioned attribute in common are observed at the same location simultaneously.
- 19 August 2023
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
- Vol. 2678  (5) , 525-541
National-Level Multimodal Origin–Destination Estimation Based on Passively Collected Location Data and Machine Learning Methods
Along with the development of information and positioning technologies, there emerges passively collected location data that contain location observations with time information from various types of mobile devices. Passive location data are known for their large sample size and continuous behavior observations. However, they also require careful and comprehensive data processing and modeling algorithms for privacy protection and practical applications. In the meantime, the travel demand estimation of origin–destination (OD) tables is fundamental in transportation planning and analysis. There is a lack of national OD estimation that provides time-dependent travel behaviors for all travel modes. Passively collected location data appeal to researchers for their potential of serving as the data source for estimation and monitoring of large-scale multimodal travel demand. This research proposes a comprehensive set of methods for passive location data processing including data cleaning, activity location and purpose identification, trip-level information identification, social demographic imputation, sample weighting and expansion, and demand validation. For each task, the paper evaluates the state-of-the-practice and state-of-the-art algorithms and develops an applicable method jointly considering different features of various passive location data sources, imputation accuracy, and computation efficiency. The paper further examines the viability of the method kit in a national-level case study and successfully derives the multimodal national-level OD estimates with additional data products, such as trip rate and vehicle miles traveled, at different geographic levels and temporal resolutions.
- 19 August 2023
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
- Vol. 2678  (5) , 525-541
National-Level Multimodal Origin–Destination Estimation Based on Passively Collected Location Data and Machine Learning Methods
Along with the development of information and positioning technologies, there emerges passively collected location data that contain location observations with time information from various types of mobile devices. Passive location data are known for their large sample size and continuous behavior observations. However, they also require careful and comprehensive data processing and modeling algorithms for privacy protection and practical applications. In the meantime, the travel demand estimation of origin–destination (OD) tables is fundamental in transportation planning and analysis. There is a lack of national OD estimation that provides time-dependent travel behaviors for all travel modes. Passively collected location data appeal to researchers for their potential of serving as the data source for estimation and monitoring of large-scale multimodal travel demand. This research proposes a comprehensive set of methods for passive location data processing including data cleaning, activity location and purpose identification, trip-level information identification, social demographic imputation, sample weighting and expansion, and demand validation. For each task, the paper evaluates the state-of-the-practice and state-of-the-art algorithms and develops an applicable method jointly considering different features of various passive location data sources, imputation accuracy, and computation efficiency. The paper further examines the viability of the method kit in a national-level case study and successfully derives the multimodal national-level OD estimates with additional data products, such as trip rate and vehicle miles traveled, at different geographic levels and temporal resolutions.
- Open Access
An elaborated pattern-based method of identifying data oscillations from mobile device location data
In recent years, passively collected GPS data have been popularly applied in various transportation studies, such as highway performance monitoring, travel behavior analysis, and travel demand estimation. Despite multiple advantages, one of the issues is data oscillations (aka outliers or data jumps), which are unneglectable since they may distort mobility patterns and lead to wrongly or biased conclusions. For transportation studies driven by GPS data, assuring the data quality by removing noises caused by data oscillations is undoubtedly important. Most GPS-based studies simply remove oscillations by checking the high speed. However, this method can mistakenly identify normal points as oscillations. Some other studies specifically discuss the removal of outliers in GPS data, but they all have limitations and do not fit passively collected GPS data. Many studies are well developed for addressing the ping-pong phenomenon in cellular data, or cellular tower data, but the oscillations in passively collected GPS data are very different for having much more various and complicated patterns and being more uncertain. Current methods are insufficient and inapplicable to passively collected GPS data. This paper aims to address the oscillated points in passively collected GPS data. A set of heuristics are proposed by identifying the abnormal movement patterns of oscillations. The proposed heuristics well fit the features of passively collected GPS data and are adaptable to studies of different scales, which are also computationally cost-effective in comparison to current methods.
Open AccessAn elaborated pattern-based method of identifying data oscillations from mobile device location data
In recent years, passively collected GPS data have been popularly applied in various transportation studies, such as highway performance monitoring, travel behavior analysis, and travel demand estimation. Despite multiple advantages, one of the issues is data oscillations (aka outliers or data jumps), which are unneglectable since they may distort mobility patterns and lead to wrongly or biased conclusions. For transportation studies driven by GPS data, assuring the data quality by removing noises caused by data oscillations is undoubtedly important. Most GPS-based studies simply remove oscillations by checking the high speed. However, this method can mistakenly identify normal points as oscillations. Some other studies specifically discuss the removal of outliers in GPS data, but they all have limitations and do not fit passively collected GPS data. Many studies are well developed for addressing the ping-pong phenomenon in cellular data, or cellular tower data, but the oscillations in passively collected GPS data are very different for having much more various and complicated patterns and being more uncertain. Current methods are insufficient and inapplicable to passively collected GPS data. This paper aims to address the oscillated points in passively collected GPS data. A set of heuristics are proposed by identifying the abnormal movement patterns of oscillations. The proposed heuristics well fit the features of passively collected GPS data and are adaptable to studies of different scales, which are also computationally cost-effective in comparison to current methods.
- Open Access
A Big Data Driven Framework for Duplicate Device Detection from Multi-sourced Mobile Device Location Data
Mobile Device Location Data (MDLD) has been popularly utilized in various fields. Yet its large-scale applications are limited because of either biased or insufficient spatial coverage of the data from individual data vendors. One approach to improve the data coverage is to leverage the data from multiple data vendors and integrate them to build a more representative dataset. For data integration, further treatments on the multi-sourced dataset are required due to several reasons. First, the possibility of carrying more than one device could result in duplicated observations from the same data subject. Additionally, when utilizing multiple data sources, the same device might be captured by more than one data provider. Our paper proposes a data integration methodology for multi-sourced data to investigate the feasibility of integrating data from several sources without introducing additional biases to the data. By leveraging the uniqueness of travel pattern of each device, duplicate devices are identified. The proposed methodology is shown to be cost-effective while it achieves the desired accuracy level. Our findings suggest that devices sharing the same imputed home location and the top five most-visited locations during a month can represent the same user in the MDLD. It is shown that more than 99.6% of the sample devices having the aforementioned attribute in common are observed at the same location simultaneously. Finally, the proposed algorithm has been successfully applied to the national-level MDLD of 2020 to produce the national passenger origin-destination data for the NextGeneration National Household Travel Survey (NextGen NHTS) program.
Open AccessA Big Data Driven Framework for Duplicate Device Detection from Multi-sourced Mobile Device Location Data
Mobile Device Location Data (MDLD) has been popularly utilized in various fields. Yet its large-scale applications are limited because of either biased or insufficient spatial coverage of the data from individual data vendors. One approach to improve the data coverage is to leverage the data from multiple data vendors and integrate them to build a more representative dataset. For data integration, further treatments on the multi-sourced dataset are required due to several reasons. First, the possibility of carrying more than one device could result in duplicated observations from the same data subject. Additionally, when utilizing multiple data sources, the same device might be captured by more than one data provider. Our paper proposes a data integration methodology for multi-sourced data to investigate the feasibility of integrating data from several sources without introducing additional biases to the data. By leveraging the uniqueness of travel pattern of each device, duplicate devices are identified. The proposed methodology is shown to be cost-effective while it achieves the desired accuracy level. Our findings suggest that devices sharing the same imputed home location and the top five most-visited locations during a month can represent the same user in the MDLD. It is shown that more than 99.6% of the sample devices having the aforementioned attribute in common are observed at the same location simultaneously. Finally, the proposed algorithm has been successfully applied to the national-level MDLD of 2020 to produce the national passenger origin-destination data for the NextGeneration National Household Travel Survey (NextGen NHTS) program.
- Open Access
- 11 November 2021
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 16  (11) , e0259803
Do racial and ethnic disparities in following stay-at-home orders influence COVID-19 health outcomes? A mediation analysis approach
Racial/ethnic disparities are among the top-selective underlying determinants associated with the disproportional impact of the COVID-19 pandemic on human mobility and health outcomes. This study jointly examined county-level racial/ethnic differences in compliance with stay-at-home orders and COVID-19 health outcomes during 2020, leveraging two-year geo-tracking data of mobile devices across ~4.4 million point-of-interests (POIs) in the contiguous United States. Through a set of structural equation modeling, this study quantified how racial/ethnic differences in following stay-at-home orders could mediate COVID-19 health outcomes, controlling for state effects, socioeconomics, demographics, occupation, and partisanship. Results showed that counties with higher Asian populations decreased most in their travel, both in terms of reducing their overall POIs’ visiting and increasing their staying home percentage. Moreover, counties with higher White populations experienced the lowest infection rate, while counties with higher African American populations presented the highest case-fatality ratio. Additionally, control variables, particularly partisanship, median household income, percentage of elders, and urbanization, significantly accounted for the county differences in human mobility and COVID-19 health outcomes. Mediation analyses further revealed that human mobility only statistically influenced infection rate but not case-fatality ratio, and such mediation effects varied substantially among racial/ethnic compositions. Last, robustness check of racial gradient at census block group level documented consistent associations but greater magnitude. Taken together, these findings suggest that US residents’ responses to COVID-19 are subject to an entrenched and consequential racial/ethnic divide.
- 11 November 2021
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 16  (11) , e0259803
Open AccessDo racial and ethnic disparities in following stay-at-home orders influence COVID-19 health outcomes? A mediation analysis approach
Racial/ethnic disparities are among the top-selective underlying determinants associated with the disproportional impact of the COVID-19 pandemic on human mobility and health outcomes. This study jointly examined county-level racial/ethnic differences in compliance with stay-at-home orders and COVID-19 health outcomes during 2020, leveraging two-year geo-tracking data of mobile devices across ~4.4 million point-of-interests (POIs) in the contiguous United States. Through a set of structural equation modeling, this study quantified how racial/ethnic differences in following stay-at-home orders could mediate COVID-19 health outcomes, controlling for state effects, socioeconomics, demographics, occupation, and partisanship. Results showed that counties with higher Asian populations decreased most in their travel, both in terms of reducing their overall POIs’ visiting and increasing their staying home percentage. Moreover, counties with higher White populations experienced the lowest infection rate, while counties with higher African American populations presented the highest case-fatality ratio. Additionally, control variables, particularly partisanship, median household income, percentage of elders, and urbanization, significantly accounted for the county differences in human mobility and COVID-19 health outcomes. Mediation analyses further revealed that human mobility only statistically influenced infection rate but not case-fatality ratio, and such mediation effects varied substantially among racial/ethnic compositions. Last, robustness check of racial gradient at census block group level documented consistent associations but greater magnitude. Taken together, these findings suggest that US residents’ responses to COVID-19 are subject to an entrenched and consequential racial/ethnic divide.
- 18 September 2021
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
- Vol. 2677  (4) , 168-180
Interactive COVID-19 Mobility Impact and Social Distancing Analysis Platform
The research team has utilized privacy-protected mobile device location data, integrated with COVID-19 case data and census population data, to produce a COVID-19 impact analysis platform that can inform users about the effects of COVID-19 spread and government orders on mobility and social distancing. The platform is being updated daily, to continuously inform decision-makers about the impacts of COVID-19 on their communities, using an interactive analytical tool. The research team has processed anonymized mobile device location data to identify trips and produced a set of variables, including social distancing index, percentage of people staying at home, visits to work and non-work locations, out-of-town trips, and trip distance. The results are aggregated to county and state levels to protect privacy, and scaled to the entire population of each county and state. The research team is making their data and findings, which are updated daily and go back to January 1, 2020, for benchmarking, available to the public to help public officials make informed decisions. This paper presents a summary of the platform and describes the methodology used to process data and produce the platform metrics.
- 18 September 2021
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
- Vol. 2677  (4) , 168-180
Interactive COVID-19 Mobility Impact and Social Distancing Analysis Platform
The research team has utilized privacy-protected mobile device location data, integrated with COVID-19 case data and census population data, to produce a COVID-19 impact analysis platform that can inform users about the effects of COVID-19 spread and government orders on mobility and social distancing. The platform is being updated daily, to continuously inform decision-makers about the impacts of COVID-19 on their communities, using an interactive analytical tool. The research team has processed anonymized mobile device location data to identify trips and produced a set of variables, including social distancing index, percentage of people staying at home, visits to work and non-work locations, out-of-town trips, and trip distance. The results are aggregated to county and state levels to protect privacy, and scaled to the entire population of each county and state. The research team is making their data and findings, which are updated daily and go back to January 1, 2020, for benchmarking, available to the public to help public officials make informed decisions. This paper presents a summary of the platform and describes the methodology used to process data and produce the platform metrics.
- 12 August 2021
- Published by Springer Nature in Transportation
- Vol. 49  (5) , 1339-1383
A data-driven travel mode share estimation framework based on mobile device location data
Mobile device location data (MDLD) contains abundant travel behavior information to support travel demand analysis. Compared to traditional travel surveys, MDLD has larger spatiotemporal coverage of the population and its mobility. However, ground truth information such as trip origins and destinations, travel modes, and trip purposes are not included by default. Such important attributes must be imputed to maximize the usefulness of the data. This paper targets at studying the capability of MDLD on estimating travel mode share at aggregated levels. A data-driven framework is proposed to extract travel behavior information from MDLD. The proposed framework first identifies trip ends with a modified Spatiotemporal Density-based Spatial Clustering of Applications with Noise algorithm. Then three types of features are extracted for each trip to impute travel modes using machine learning models. A labeled MDLD dataset with ground truth information is used to train the proposed models, resulting in a 95% recall rate in identifying trip ends and over 93% tenfold cross-validation accuracy in imputing the five travel modes (drive, rail, bus, bike and walk) with a random forest (RF) classifier. The proposed framework is then applied to two large-scale MDLD datasets, covering the Baltimore-Washington metropolitan area and the United States, respectively. The estimated trip distance, trip time, trip rate distribution, and travel mode share are compared against travel surveys at different geographies. The results suggest that the proposed framework can be readily applied in different states and metropolitan regions with low cost in order to study multimodal travel demand, understand mobility trends, and support decision making.
- 12 August 2021
- Published by Springer Nature in Transportation
- Vol. 49  (5) , 1339-1383
A data-driven travel mode share estimation framework based on mobile device location data
Mobile device location data (MDLD) contains abundant travel behavior information to support travel demand analysis. Compared to traditional travel surveys, MDLD has larger spatiotemporal coverage of the population and its mobility. However, ground truth information such as trip origins and destinations, travel modes, and trip purposes are not included by default. Such important attributes must be imputed to maximize the usefulness of the data. This paper targets at studying the capability of MDLD on estimating travel mode share at aggregated levels. A data-driven framework is proposed to extract travel behavior information from MDLD. The proposed framework first identifies trip ends with a modified Spatiotemporal Density-based Spatial Clustering of Applications with Noise algorithm. Then three types of features are extracted for each trip to impute travel modes using machine learning models. A labeled MDLD dataset with ground truth information is used to train the proposed models, resulting in a 95% recall rate in identifying trip ends and over 93% tenfold cross-validation accuracy in imputing the five travel modes (drive, rail, bus, bike and walk) with a random forest (RF) classifier. The proposed framework is then applied to two large-scale MDLD datasets, covering the Baltimore-Washington metropolitan area and the United States, respectively. The estimated trip distance, trip time, trip rate distribution, and travel mode share are compared against travel surveys at different geographies. The results suggest that the proposed framework can be readily applied in different states and metropolitan regions with low cost in order to study multimodal travel demand, understand mobility trends, and support decision making.