Smart Card Transaction Data Research Articles

A data fusion approach with mobile phone data for updating travel survey-based mode split estimates

Up-to-date information on different modes of travel to monitor transport traffic and evaluate rapid urban transport planning interventions is often lacking. Transport systems typically rely on traditional data sources providing outdated mode-of-travel data due to their data latency, infrequent data collection and high cost. To address this issue, we propose a method that leverages mobile phone data as a cost-effective and rich source of geospatial information to capture current human mobility patterns at unprecedented spatiotemporal resolution. Our approach employs mobile phone application usage traces to infer modes of transportation that are challenging to identify (bikes and ride-hailing/taxi services) based on mobile phone location data. Using data fusion and matrix factorisation techniques, we integrate official data sources (household surveys and census data) with mobile phone application usage data. This integration enables us to reconstruct the official data and create an updated dataset that incorporates insights from digital footprint data from application usage. We illustrate our method using a case study focused on Santiago, Chile successfully inferring four modes of transportation: mass-transit (all public transportation), motorised (cars and similar vehicles), active (pedestrian and cycle trips), and taxi (traditional taxi and ride-hailing services). Our analysis revealed significant changes in transportation patterns between 2012 and 2020. We quantify a reduction in mass-transit usage across municipalities in Santiago, except where metro/rail lines have been more recently introduced, highlighting added resilience to the public transport network of these infrastructure enhancements. Additionally, we evidence an overall increase in motorised transport throughout Santiago, revealing persistent challenges in promoting urban sustainable transportation. Findings also point to a rise in the share of taxi usage, and a drop in active mobility, suggesting a modal shift towards less sustainable modes of travel. We validate our findings comparing our updated estimates with official smart card transaction data. The consistency of findings with expert domain knowledge from the literature and historical transport usage trends further support the robustness of our approach.

Longitudinal analysis of public transport usage by older people using a latent Markov model

ABSTRACT Understanding trends in public transport usage by older people over the years is important for the long-term planning and development of age-friendly public transport. Despite extensive research on travel patterns of older people, very limited studies have explored the evolution oftheir public transport usage over the years. Using six years of smartcard transaction data and anonymous cardholder membership data from Shizuoka, Japan, we develop a latent Markov model to account for unobserved heterogeneity and to analyze the evolution of public transport usage by older people. The results indicate that age, gender, and residential built environment influence trends in publictransport usage by older adults over the years. There are five latentstates for older men and women, respectively, demonstrating theexistence of unobserved heterogeneity. Propensity to use public transport affected by unobserved factors tends to retain the same, especially for older people in the latent states of very low, low, and very high propensities.

Metro Passenger-Flow Representation via Dynamic Mode Decomposition and Its Application.

Passenger-flow anomaly detection and prediction are essential tasks for intelligent operation of the metro system. Accurate passenger-flow representation is the foundation of them. However, spatiotemporal dependencies, complex dynamic changes, and anomalies of passenger-flow data bring great challenges to data representation. Taking advantage of the time-varying characteristics of data, we propose a novel passenger-flow representation model based on low-rank dynamic mode decomposition (DMD), which also integrates the global low-rank nature and sparsity to explore the spatiotemporal consistency of data and depict abrupt data, respectively. The model can detect anomalies and predict short-term passenger flow conveniently and flexibly. For anomaly detection, we further introduce a strong temporal Toeplitz regularization to characterize the temporal periodic change of data, so as to more accurately detect anomalies. We conduct experiments with smart card transaction data from the Beijing metro system to assess the performance of the model in two use cases. In terms of anomaly detection, the experimental results demonstrate that our method can detect anomalies efficiently, especially for time sequence anomalies. As for short-term prediction, our model is superior to other methods in most cases.

Extracting Bus Transit Boarding and Alighting Information Using Smart Card Transaction Data

The smart card–based automated fare collection (AFC) technologies provide new opportunities for transportation data collection since smart card data (SCD) contain a significant amount of archived information that can be gathered and leveraged to help estimate public transit origin-destination (OD) matrices. Both boarding and alighting location detection are important particularly when there is no automatic vehicle location (AVL) system or other information in the database. With the analysis of raw data using limited information in this paper, an algorithm for boarding stop identification is presented after the trip direction and boarding cluster identification. Alighting stops are then identified based on the relationship between records. Finally, the feasibility and practicality of the methodology are tested using the bus transit SCD collected in Guangzhou, China.

A knowledge graph-based method for epidemic contact tracing in public transportation

Contact tracing is an effective measure by which to prevent further infections in public transportation systems. Considering the large number of people infected during the COVID-19 pandemic, digital contact tracing is expected to be quicker and more effective than traditional manual contact tracing, which is slow and labor-intensive. In this study, we introduce a knowledge graph-based framework for fusing multi-source data from public transportation systems to construct contact networks, design algorithms to model epidemic spread, and verify the validity of an effective digital contact tracing method. In particular, we take advantage of the trip chaining model to integrate multi-source public transportation data to construct a knowledge graph. A contact network is then extracted from the constructed knowledge graph, and a breadth-first search algorithm is developed to efficiently trace infected passengers in the contact network. The proposed framework and algorithms are validated by a case study using smart card transaction data from transit systems in Xiamen, China. We show that the knowledge graph provides an efficient framework for contact tracing with the reconstructed contact network, and the average positive tracing rate is over 96%.

Inference of activity patterns from urban sensing data using conditional random fields

The study of human activity–travel patterns for urban planning has evolved a long way in theories, methodologies, and applications. However, the scarcity of data has become a major barrier for the advancement of research in the field. Recently, the proliferation of urban sensing and location-based devices generates voluminous streams of spatio-temporal registered information. In this study, we propose an approach using the linear-chain Conditional Random Fields (CRFs) model to learn the spatio-temporal correspondences of different types of activities and the inter-dependencies among sequential activities from training dataset such as the household travel or time use surveys, and to infer the hidden activity types associated with urban sensing data. The performance of the CRFs model is compared against the Random Forest (RF) model, which has been used in a number of existing studies. The results show that the linear-chain CRFs models generally outperform the RF counterparts with respect to classification accuracy of activity types, in particular for those travelers having more outdoor daily activities. The proposed methodology is demonstrated by reconstructing the activity landscape of the surrounding area of a major Mass Rail Transit station in Singapore using the transit smart card transaction data. The inferred activities from the transit smart card data are expected to complement the ground surveys and improve our understanding of the interactions of different components of activities/travels as well as the relationship between urban space and human activities.

Examining public transport usage by older adults with smart card data: A longitudinal study in Japan

Understanding public transport usage by older adults is necessary to develop senior-friendly public transport and improve the mobility of older persons. Although extensive literature has examined the travel patterns of older adults, very limited efforts have been invested to explore the longitudinal variability in public transport usage by different age groups of older adults. To address this limitation, we developed user-monthly profiles to explore the seasonal variability in public transport usage by older adults and defined user-based time slots of the day and geographical user areas to represent daily trip patterns and examine day-to-day variability. Using one-year smart card transaction data and an anonymous cardholder database from Shizuoka, Japan, we evaluated the seasonal and day-to-day variability in public transport usage by older adults. We also analyzed the role of age and living environment in travel pattern variability. The results indicate that older adults in highly developed areas and younger-old group (aged 65–74) are more likely to be characterized by high-frequency public transport usage and low seasonal variability. Additionally, the day-to-day variability in public transport usage by older adults is greater in more developed areas and appears to increase with age. This study enhances our understanding of public transport usage by older adults, which may contribute to the development of senior-friendly public transport policies and services.

Evaluation model of bus routes optimization scheme based on multi-source bus data

An effective public transport network is helpful to improve the quality of public transport service and save travel time of travelers. As an important part of urban transportation, the continuous optimization of bus network is necessary to improve the service level of bus transit. Aiming to evaluate the rationality and feasibility of bus optimization scheme, the paper discussed the factors about the bus service supply and demand, including the bus operation, passenger flow and service level. Firstly, based on multi-source data including bus smart card transaction data, bus location data and static attribute data of bus network, the evaluation indicator system of bus routes optimization effect is proposed. Secondly, a comprehensive evaluation model of bus route optimization is established by applying the analytic hierarchy process (AHP). This paper systematically analyzes the various levels bus optimization scheme including bus route level, station level, bus scheduling scheme, and constructs an evaluation model covering the single bus route optimization and multi-routes combination optimization. The quantitative grading evaluation method of bus routes optimization proposed in this paper can scientifically classify the performance of bus route optimization and provide valuable reference for the decision-making of bus route adjustment. Eventually, this study also selects Beijing bus routes optimization scheme for empirical analysis.

Exploring Travel Pattern Variability of Public Transport Users Through Smart Card Data: Role of Gender and Age

A better understanding of travel pattern variability is important for public transport (PT) authorities to improve passenger experience and service provision. Although many studies have examined the travel pattern variability of PT users, these studies are often limited to a short analysis period or to only one dimension of travel behavior. In addition, there is limited knowledge of how the demographic characteristics of PT users are associated with their travel pattern variability. To address these limitations, we develop a novel measure that simultaneously considers multiple dimensions of travel behavior to quantify the intrapersonal variability in weekly PT usage. Moreover, we examine interpersonal variability by identifying clusters of users who share similar weekly profiles. Based on smart card transaction data for 52 weeks and an anonymous cardholder database (including age and gender) from Shizuoka, Japan, we analyze the intrapersonal and interpersonal variability in weekly PT usage as well as the role of gender and age in travel pattern variability. The results indicate that gender and age play an important role in the travel pattern variability of PT users. Female users exhibit higher intrapersonal variability than their male counterparts. Weekly patterns are the most diverse for users aged 70 or over, followed by the users aged 65–69. Regarding interpersonal variability, we identify five clusters of users, each characterized by a distinct weekly profile and associated with certain age and gender.

Examining influential factors on the energy consumption of electric and diesel buses: A data-driven analysis of large-scale public transit network in Beijing

Switching from diesel-powered to battery-powered buses has been a global tendency. Traditional approaches rely on standard driving cycles or fuel economy data for energy consumption on a limited number of buses; thus, expanding to large bus fleets at the city level has become challenging. This study uses high-resolution GPS and smart card transaction data to generate each bus driving profile and weight dynamics in a large-scale transit network. Two vehicle activity-based energy consumption models are adopted and calibrated for diesel bus (DB) and electric bus by using the field data of 630 bus routes in Beijing. The average energy consumptions of DBs and electric buses are 43.5 and 14.1 L/100 km, respectively. A gradient boosting regression tree algorithm is presented to examine and rank distinct influential factors on the energy consumption of DBs and electric buses. Heterogenous behaviors are identified: the leading attributes affecting the energy consumption of electric buses and DBs are route characteristics and operational condition, respectively. After computing, the total energy conservation of electrifying all bus fleets is equivalent to 0.87% of daily electricity demand in Beijing. These findings set a good base for further studies on bus fleet replacement, charging infrastructure deployment, and electrified route prioritization.

A procedure for public transit OD matrix generation using smart card transaction data

Most fare collection systems are initially installed as single-purpose devices which are only used for collecting fare; however, many transit planners consider them as a rich source of data required for studying the passengers' trip trends. Although, usually, there is no transaction made at the destination stop, making some assumptions can help us infer the destination. In this study, we present an integrated procedure that can generate origin–destination matrices and passenger load profiles as essential tools for public transport planning processes. Moreover, this procedure can be used to detect and analyze trips that include transfers. In an attempt to employ the proposed algorithm in the Tehran bus rapid transit network, 52% of the transactions could be used to trace the trips in an origin–destination format. The trips that include transfers are recognized and analyzed further. Our detailed results of the method application indicate that the proposed algorithm is a productive and economical public transport planning method.

Frequent-pattern growth algorithm based association rule mining method of public transport travel stability

The accurate depiction and understanding of the travel behavior characteristics of public transport (PT) commuters is an important foundation for better improving PT service and encouraging car owners to use the sustainable and ecofriendly PT; and there are significant differences in the travel stability (TS) characteristic of PT commuters, developing methods for accurately measuring such differences is an issue. Therefore, smart card transaction data, line and stop data, and travel survey data from Beijing were collected, then individual travel chain information of commuting passengers was extracted using the associating and matching method. Thereafter, a multilevel characteristic indicator system including the number of nonhome activity points, commuting trip ratio, travel spatial equilibrium, time stability and departure time concentration was constructed to capture the individual TS. Moreover, an association rule mining model based on the frequent-pattern (FP) growth algorithm was developed by modeling the indicators as items and the PT-commuter TS as transactions. Thus, seven meaningful rules for revealing the internal relationships between individual travel characteristics and commuter TS were obtained, and PT commuters were classified into three groups according to the TS levels. Finally, a conceptual model of the mode shift to higher TS levels among commuters was developed, and some targeted measures for enhancing the TS levels of PT users were proposed. The findings are expected to provide new perspectives for travel behavior analysis and policy control to enhance and maintain passenger TS, also are conducive to increasing PT usage while reducing the usage of cars.

Mining commuting behavior of urban rail transit network by using association rules

Automated Fare Collection (AFC) systems in rail transit services collect enormous amounts of detailed data on on-board transactions. A better understanding of travelers’ commuting and transfer behavior based on those massive volumes of AFC data would enable the rail service operators to evaluate their service quality and optimize operation strategies. This paper proposes an efficient and effective data mining procedure to figure out the association rules, aiming to extract connectivity and correlation of passenger flow among different services lines in urban rail transit networks. A case study based on the Beijing Subway network is conducted to demonstrate the applicability of the proposed method. Using up to 28 million AFC smart card transaction data, we match and analyze travelers’ trip chains to investigate the commuting trip patterns in terms of spatio-temporal distribution characteristics. An innovational non-nigh-to-five commuting behavior and traditional nine-to-five commuting behavior are divided by the obtained associated rules to ensure a more nuanced description of commuting behaviors. Further, the results indicated by stronger association rules (2-frequent itemset and 3-frequent itemset) also provide a better understanding of transfer behaviors, like the frequent transfers among different service lines, and potentially vulnerable stations in the network. The research outcomes can be used to assist rail transit service operators in developing optimal operation strategies like timetabling design to enhance the transfer performance between different rail lines.

Understanding Daily Travel Patterns of Subway Users – An Example From the Beijing Subway

The daily travel patterns (DTPs) present short-term and timely characteristics of the users’ travel behaviour, and they are helpful for subway planners to better understand the travel choices and regularity of subway users (SUs) in details. While several well-known subway travel patterns have been detected, such as commuting modes and shopping modes, specific features of many patterns are still confused or omitted. Now, based on the automatic fare collection (AFC) system, a data-mining procedure to recognize DTPs of all SUs has become possible and effective. In this study, DTPs are identified by the station sequences (SSs), which are modelled from smart card transaction data of the AFC system. The data-mining procedure is applied to a large weekly sample from the Beijing Subway to understand DTPs. The results show that more than 93% SUs of the Beijing Subway travel in 7 DTPs, which are remarkably stable in share and distribution. Different DTPs have their own unique characteristics in terms of time distribution, activity duration and repeatability, which provide a wealth of information to calibrate different types of users and characterize their travel patterns.

User retention tendency of bus routes based on user behavior transition in an area with low mode share of public transport

Although both public transport and private modes have their own purposes and safety issues, most people are free to choose either way to make a trip. Previous research states that increasing the mode share of public transport is an important transportation policy to improve traffic safety, and it is a key outcome of demand management indeed. However, rather than merely focusing on increasing its ridership, a more reliable way to reach for the universality of a public transport system is through its customer retention tendency. Research on satisfaction with bus service often focuses on the influence of specific variables. However, numerous variables may influence users' decision-making. To ease their work, managers have no choice but to ignore some unknown variables. Therefore, we now propose a bottom-up procedure, which needs only smart card data, to obtain the odds ratio of usage of a specific bus route. Logistic regression models are calibrated based on four behavior groups, and the significant coefficients of route variables represent the odds ratios of the bus route usage. The calibration of odds ratio does not need any individual personal or individual socio-economical information, but only smart card transaction data. This method will dramatically decrease the cost and time for data collection. Further, the procedure proposed in this study can be encapsulated in software, which managers can then use to assist their planning.

Demand management of congested public transport systems: a conceptual framework and application using smart card data

Transportation demand management, long used to reduce car traffic, is receiving attention among public transport operators as a means to reduce congestion in crowded public transportation systems. Though far less studied, a more structured approach to public transport demand management (PTDM) can help agencies make informed decisions on the combination of PTDM and infrastructure investments that best manage crowding. Automated fare collection data, readily available in many public transport agencies, provide a unique platform to advance systematic approaches for the design and evaluation of PTDM strategies. The paper discusses the main steps for developing PTDM programs: (a) problem identification and formulation of program goals; (b) program design; (c) evaluation; and (d) monitoring. The problem identification phase examines bottlenecks in the system based on a spatiotemporal passenger flow analysis. The design phase identifies the main design parameters based on a categorization of potential interventions along spatial, temporal, modal, and targeted user group parameters. Evaluation takes place at the system, group, and individual levels, taking advantage of the detailed information obtained from smart card transaction data. The monitoring phase addresses the long-term sustainability of the intervention and informs potential changes to improve its effectiveness. A case study of a pre-peak fare discount policy in Hong Kong’s MTR network is used to illustrate the application of the various steps with focus on evaluation and analysis of the impacts from a behavioral point of view. Smart card data from before and after the implementation of the scheme from a panel of users was used to study policy-induced behavior shifts. A cluster analysis inferred customer groups relevant to the analysis based on their usage patterns. Users who shifted their behavior were identified based on a change point analysis and a logit model was estimated to identify the main factors that contribute to this change: the amount of time a user needed to shift his/her departure time, departure time variability, fare savings, and price sensitivity. User heterogeneity suggests that future incentives may be improved if they target specific groups.

Water transit and excess travel: discrete choice modelling of bus and ferry trips in Brisbane, Australia

ABSTRACTRecent investment in urban ferry transport has created interest in what value such systems provide in a public transport network. In some cases, ferry services are in direct competition with other land-based transport, and despite often longer travel times passengers still choose water transport. This paper seeks to identify a premium attached to urban water transit through an identification of excess travel patterns. A one-month sample of smart card transaction data for Brisbane, Australia, was used to compare bus and ferry origin–destination pairs between a selected suburban location and the central business district. Logistic regression of the data found that ferry travel tended towards longer travel times (OR = 2.282), suggesting passengers do derive positive utility from ferry journeys. The research suggests the further need to incorporate non-traditional measures other than travel time for deciding the value of water transit systems.

Extracting bus transit boarding stop information using smart card transaction data

The smart card-based automated fare collection (AFC) system has become the main method for collecting urban bus and rail transit fares in many cities worldwide. Such smart card technologies provide new opportunities for transportation data collection since the transaction data obtained through AFC system contains a significant amount of archived information which can be gathered and leveraged to help estimate public transit origin–destination matrices. Boarding location detection is an important step particularly when there is no automatic vehicle location (AVL) system or GPS information in the database in some cases. With the analysis of raw data without AVL information in this paper, an algorithm for trip direction detection is built and the directions for any bus in operation can be confirmed. The transaction interval between each adjacent record will also be analyzed to detect the boarding clusters for all trips in sequence. Boarding stops will then be distributed with the help of route information and operation schedules. Finally, the feasibility and practicality of the methodology are tested using the bus transit smart card data collected in Guangzhou, China.

Estimating the activity types of transit travelers using smart card transaction data: a case study of Singapore

Understanding individual daily activity patterns is essential for travel demand management and urban planning. This research introduces a new method to infer transit riders’ activities from their smart card transaction records. Using Singapore as an example, activity type classification models were built using household travel survey and a rich set of urban built environment measures to reveal the spatial and temporal correspondences that indicate the activity participation of transit riders. The calibrated model is then applied to the transit smart card dataset to extract the embedded activity information. The proposed approach enables to spatially and temporally quantify, visualize, and examine urban activity landscapes in a metropolitan area and provides real-time decision support for the city. This study also demonstrates the potential value of combining new ‘‘big data’’ such as transit smart card data and “small data” such as traditional travel surveys to create better insights of urban travel demand and activity dynamics.

FISS: function identification of subway stations based on semantics mining and functional clustering

In modern cities, the subway system plays an important role in carrying a large proportion of passenger transport. However, there still remain some issues on how to accurately identify the regional functions of subway stations. In this study, the authors propose an approach named FISS for identifying the functions of subway station regions based on semantics mining and functional clustering. First, they extract the passenger's travel patterns of each subway station based on the smart card transaction data and Shanghai subway network data and calculate the relative point of interest (POI) contents of each subway station by using Shanghai POI data, then feed the two above-mentioned results into latent Dirichlet allocation model for obtaining the mobile semantics and the location semantics separately. Furthermore, they carry out standardisation after combining the two semantics, then extract the functional characteristic vectors of subway stations by conducting sparse principal components analysis, and cluster these vectors by using the improved k -means algorithm. At last, they visualise the result after taking subway station's function identification by the interclass passenger flow transfer, the distribution of geographical function proportion and the similarity of inter-cluster. The results demonstrate the accuracy and efficiency of the proposed approach compared with other existing methods.