Search

Highlights: [•] A unified estimator for aggregate excess journey time in any incidence behavior. [•] The proposed estimator was applied to London Overground using Oyster smartcard data. [•] Significant variations in EJT between lines and time periods were observed. [Copyright &y& Elsevier]

• Use behavioral knowledge to design a new DNN architecture with alternative-specific utility (ASU-DNN). • ASU-DNN provides a more regular substitution pattern of travel mode choices. • ASU-DNN improves both the predictive power and interpretability. • Behavioral knowledge can function as an effective domain-knowledge-based regularization. Whereas deep neural network (DNN) is increasingly applied to choice analysis, it is challenging to reconcile domain-specific behavioral knowledge with generic-purpose DNN, to improve DNN's interpretability and predictive power, and to identify effective regularization methods for specific tasks. To address these challenges, this study demonstrates the use of behavioral knowledge for designing a particular DNN architecture with alternative-specific utility functions (ASU-DNN) and thereby improving both the predictive power and interpretability. Unlike a fully connected DNN (F-DNN), which computes the utility value of an alternative k by using the attributes of all the alternatives, ASU-DNN computes it by using only k 's own attributes. Theoretically, ASU-DNN can substantially reduce the estimation error of F-DNN because of its lighter architecture and sparser connectivity, although the constraint of alternative-specific utility can cause ASU-DNN to exhibit a larger approximation error. Empirically, ASU-DNN has 2–3% higher prediction accuracy than F-DNN over the whole hyperparameter space in a private dataset collected in Singapore and a public dataset available in the R mlogit package. The alternative-specific connectivity is associated with the independence of irrelevant alternative (IIA) constraint, which as a domain-knowledge-based regularization method is more effective than the most popular generic-purpose explicit and implicit regularization methods and architectural hyperparameters. ASU-DNN provides a more regular substitution pattern of travel mode choices than F-DNN does, rendering ASU-DNN more interpretable. The comparison between ASU-DNN and F-DNN also aids in testing behavioral knowledge. Our results reveal that individuals are more likely to compute utility by using an alternative's own attributes, supporting the long-standing practice in choice modeling. Overall, this study demonstrates that behavioral knowledge can guide the architecture design of DNN, function as an effective domain-knowledge-based regularization method, and improve both the interpretability and predictive power of DNN in choice analysis. Future studies can explore the generalizability of ASU-DNN and other possibilities of using utility theory to design DNN architectures. [ABSTRACT FROM AUTHOR]

Highlights • Introduces application Dynamic Factor Models for short-term demand prediction. • Correlation clustering of stations to identify co-behaving stations. • Proposes a simple methodology for predicting the impact of planned events on demand. • Extensive case study of the Central Line stations of London Underground. Abstract Demand for public transportation is highly affected by passengers' experience and the level of service provided. Thus, it is vital for transit agencies to deploy adaptive strategies to respond to changes in demand or supply in a timely manner, and prevent unwanted deterioration in service quality. In this paper, a real time prediction methodology, based on univariate and multivariate state-space models, is developed to predict the short-term passenger arrivals at transit stations. A univariate state-space model is developed at the station level. Through a hierarchical clustering algorithm with correlation distance, stations with similar demand patterns are identified. A dynamic factor model is proposed for each cluster, capturing station interdependencies through a set of common factors. Both approaches can model the effect of exogenous events (such as football games). Ensemble predictions are then obtained by combining the outputs from the two models, based on their respective accuracy. We evaluate these models using data from the 32 stations on the Central line of the London Underground (LU), operated by Transport for London (TfL). The results indicate that the proposed methodology performs well in predicting short-term station arrivals for the set of test days. For most stations, ensemble prediction has the lowest mean error, as well as the smallest range of error, and exhibits more robust performance across the test days. [ABSTRACT FROM AUTHOR]

For intelligent urban transportation systems, the ability to predict individual mobility is crucial for personalized traveler information, targeted demand management, and dynamic system operations. Whereas existing methods focus on predicting the next location of users, little is known regarding the prediction of the next trip. The paper develops a methodology for predicting daily individual mobility represented as a chain of trips (including the null set, no travel), each defined as a combination of the trip start time t , origin o , and destination d . To predict individual mobility, we first predict whether the user will travel ( trip making prediction ), and then, if so, predict the attributes of the next trip ( t , o , d ) ( trip attribute prediction ). Each of the two problems can be further decomposed into two subproblems based on the triggering event. For trip attribute prediction, we propose a new model, based on the Bayesian n -gram model used in language modeling, to estimate the probability distribution of the next trip conditional on the previous one. The proposed methodology is tested using the pseudonymized transit smart card records from more than 10,000 users in London, U.K. over two years. Based on regularized logistic regression, our trip making prediction models achieve median accuracy levels of over 80%. The prediction accuracy for trip attributes varies by the attribute considered—around 40% for t , 70–80% for o and 60–70% for d . Relatively, the first trip of the day is more difficult to predict. Significant variations are found across individuals in terms of the model performance, implying diverse travel behavior patterns. [ABSTRACT FROM AUTHOR]

The public transport networks of dense cities such as London serve passengers with widely different travel patterns. In line with the diverse lives of urban dwellers, activities and journeys are combined within days and across days in diverse sequences. From personalized customer information, to improved travel demand models, understanding this type of heterogeneity among transit users is relevant to a number of applications core to public transport agencies’ function. In this study, passenger heterogeneity is investigated based on a longitudinal representation of each user’s multi-week activity sequence derived from smart card data. We propose a methodology leveraging this representation to identify clusters of users with similar activity sequence structure. The methodology is applied to a large sample ( n = 33,026) from London’s public transport network, in which each passenger is represented by a continuous 4-week activity sequence. The application reveals 11 clusters, each characterized by a distinct sequence structure. Socio-demographic information available for a small sample of users ( n = 1973) is combined to smart card transactions to analyze associations between the identified patterns and demographic attributes including passenger age, occupation, household composition and income, and vehicle ownership. The analysis reveals that significant connections exist between the demographic attributes of users and activity patterns identified exclusively from fare transactions. [ABSTRACT FROM AUTHOR]

• Investigated computational fairness in travel behavior modeling. • Operationalized computational fairness by equality of opportunity. • Revealed prediction disparity for race, income, gender, health and region. • Adopted a bias mitigation method to improve fairness in travel behavior prediction. • Demonstrated the accuracy-fairness tradeoff. Although researchers increasingly adopt machine learning to model travel behavior, they predominantly focus on prediction accuracy, ignoring the ethical challenges embedded in machine learning algorithms. This study introduces an important missing dimension - computational fairness - to travel behavior analysis. It highlights the accuracy-fairness tradeoff instead of the single dimensional focus on prediction accuracy in the contexts of deep neural network (DNN) and discrete choice models (DCM). We first operationalize computational fairness by equality of opportunity , then differentiate between the bias inherent in data and the bias introduced by modeling. The models inheriting the inherent biases can risk perpetuating the existing inequality in the data structure, and the biases in modeling can further exacerbate it. We then demonstrate the prediction disparities in travel behavior modeling using the 2017 National Household Travel Survey (NHTS) and the 2018–2019 My Daily Travel Survey in Chicago. Empirically, DNN and DCM reveal consistent prediction disparities across multiple social groups: both over-predict the false negative rate of frequent driving for the ethnic minorities, the low-income and the disabled populations, and falsely predict a higher travel burden of the socially disadvantaged groups and the rural populations than reality. Comparing DNN with DCM, we find that DNN can outperform DCM in prediction disparities because of DNN's smaller misspecification error. To mitigate prediction disparities, this study introduces an absolute correlation regularization method, which is evaluated with synthetic and real-world data. The results demonstrate the prevalence of prediction disparities in travel behavior modeling, and the disparities still persist regarding a variety of model specifics such as the number of DNN layers, batch size and weight initialization. Since these prediction disparities can exacerbate social inequity if prediction results without fairness adjustment are used for transportation policy making, we advocate for careful consideration of the fairness problem in travel behavior modeling, and the use of bias mitigation algorithms for fair transport decisions. [ABSTRACT FROM AUTHOR]

• We propose a real-time predictive decision support platform which addresses both, operations control and customer information needs. • The system provides crowding predictions on trains and platforms, communicates this information to passengers • Using a fixed-point formulation, the methodology accounts for the expected passenger behavior in response to the predictive crowding information • It is demonstrated through a case study that providing predictive information to passengers can potentially reduce denied boarding and lead to better utilization of train capacity. Demand for public transport has witnessed a steady growth over the last decade in many densely populated cities around the world. However, capacity has not always matched this increased demand. As such, passengers experience long waiting times and are denied boarding during the peak hours. Crowded platforms and the subsequent customer dissatisfaction and safety issues have become a serious concern. The COVID-19 pandemic has dramatically reduced passengers' willingness to board crowded trains, causing a surge in demand for real-time crowding information. In this paper, we propose a real-time predictive decision support platform which addresses both, operations control and customer information needs. The system provides crowding predictions on trains and platforms, communicates this information to passengers, and takes into account their response to it. It is demonstrated through a case study that providing predictive information to passengers can potentially reduce denied boarding and lead to better utilization of train capacity. [ABSTRACT FROM AUTHOR]

In the governance of the shared mobility market of a city or of a metropolitan area, there are two conflicting principles: (1) the healthy competition between multiple platforms, such as between Uber and Lyft in the United States, and (2) economies of network scale, which leads to higher chances for trips to be matched, and thus higher operation efficiency, but which also implies monopoly. The current shared mobility markets, as observed in different cities in the world, are either monopolistic, or largely segmented by multiple platforms, the latter with significant efficiency loss. How to keep the competition between platforms, but to reduce the efficiency loss due to segmentation with new market designs is the focus of this paper. We first proposed a theoretical framework of shared mobility market segmentation and then proposed four market structure designs thereupon. The framework and four designs were first discussed as an abstract model, without losing generality, thus not constrained to any specific city. High-level perspectives and detailed mechanisms for each proposed market structure were both examined. Then, to assess the real-world performance of these market structure designs, we used a ride-sharing simulator with real-world ride-hailing trip data from New York City to simulate. The proposed market designs can reduce the total vehicle-miles traveled (VMT) by 6% while serving 2.9% more customers with 8.4% fewer total number of trips. In the meantime, customers receive better services with on-average 5.4% shorter waiting time. At the end of the paper, the feasibility of implementation for each proposed market structure was discussed. • Proposed a unified framework for describing the shared mobility market structure. • Described two status-quo shared mobility markets with the proposed unified framework. • Designed four shared mobility market structures to dissolve market segmentation. • Evaluated the performance of proposed shared mobility markets using real-world data. [ABSTRACT FROM AUTHOR]

• Extract economic information from DNN for choice analysis. • Introduce both function-based and gradient-based interpretations. • Highlight three challenges associated with the automatic learning capacity of DNN. • Compare economic information from DNN with those from discrete choice models. • Economic information aggregated either over trainings or population is more reliable. While deep neural networks (DNNs) have been increasingly applied to choice analysis showing high predictive power, it is unclear to what extent researchers can interpret economic information from DNNs. This paper demonstrates that DNNs can provide economic information as complete as classical discrete choice models (DCMs). The economic information from DNNs includes choice predictions, choice probabilities, market shares, substitution patterns of alternatives, social welfare, probability derivatives, elasticities, marginal rates of substitution, and heterogeneous values of time. Unlike DCMs, DNNs can automatically learn utility functions and reveal behavioral patterns that are not prespecified by domain experts, particularly when the sample size is large. However, the economic information obtained from DNNs can be unreliable when the sample size is small, because of three challenges associated with the automatic learning capacity: high sensitivity to hyperparameters, model non-identification, and local irregularity. The first challenge is related to the statistical challenge of balancing approximation and estimation errors of DNNs, the second to the optimization challenge of identifying the global optimum in the DNN training, and the third to the robustness challenge of mitigating locally irregular patterns of estimated functions. To demonstrate the strength and challenges, we estimated the DNNs using a stated preference survey from Singapore and a revealed preference data from London, extracted the full list of economic information from the DNNs, and compared them with those from the DCMs. We found that the economic information either aggregated over trainings or population is more reliable than the disaggregate information of the individual observations or trainings, and that larger sample size, hyperparameter searching, model ensemble, and effective regularization can significantly improve the reliability of the economic information extracted from the DNNs. Future studies should investigate the requirement of sample size, better ensemble mechanisms, other regularizations and DNN architectures, better optimization algorithms, and robust DNN training methods to address DNNs three challenges to provide more reliable economic information for DNN-based choice models. [ABSTRACT FROM AUTHOR]

• The paper develops a spatiotemporal topic model for human activity discovery. • Each topic is a distribution over space and time that corresponds to an activity. • The model accounts for a mixture of discrete and continuous travel attributes. • The model fits the data significantly better than heuristic approaches. • The number of topics controls the granularity of discovered activity patterns. Although automatically collected human travel records can accurately capture the time and location of human movements, they do not directly explain the hidden semantic structures behind the data, e.g., activity types. This work proposes a probabilistic topic model, adapted from Latent Dirichlet Allocation (LDA), to discover representative and interpretable activity categorization from individual-level spatiotemporal data in an unsupervised manner. Specifically, the activity-travel episodes of an individual user are treated as words in a document, and each topic is a distribution over space and time that corresponds to certain type of activity. The model accounts for a mixture of discrete and continuous attributes—the location, start time of day, start day of week, and duration of each activity episode. The proposed methodology is demonstrated using pseudonymized transit smart card data from London, U.K. The results show that the model can successfully distinguish the three most basic types of activities—home, work, and other. As the specified number of activity categories increases, more specific subpatterns for home and work emerge, and both the goodness of fit and predictive performance for travel behavior improve. This work makes it possible to enrich human mobility data with representative and interpretable activity patterns without relying on predefined activity categories or heuristic rules. [ABSTRACT FROM AUTHOR]

• A novel method to estimate the sun glare using Google Street View panoramas. • Deep learning algorithm was used to estimate and map sun glare occurrence. • With consideration all types of obstructions of sunlight in urban environment. The sun glare is one of the major environmental hazards that cause traffic accidents. Every year many traffic accidents are caused by sun glare in the United States. Providing accurate information about when and where sun glare happens would be helpful to prevent sun glare caused traffic accidents. In this study, we proposed to use the publicly accessible Google Street View (GSV) panorama images to estimate and predict the occurrence of sun glare. GSV images have view sight similar to drivers, which make GSV images suitable for estimating the visibility of sun glare to drivers. A recently developed convolutional neural network algorithm was used to segment GSV images and predict obstructions on sun glare. Based on the predicted obstructions for given locations, we further estimated the time windows of sun glare by calculating the sun positions and the relative angles between drivers and the sun for those locations. We conducted a case study in Cambridge, Massachusetts, USA. Results show that the method can predict the occurrence of sun glare precisely. The proposed method provides an important tool for people to deal with the sun glare and reduce the potential traffic accidents caused by the sun glare. [ABSTRACT FROM AUTHOR]

The bus control problem that combines holding and stop-skipping strategies is formulated as a multi-agent reinforcement learning (MARL) problem. Traditional MARL methods, designed for settings with joint action-taking, are incompatible with the asynchronous nature of at-stop control tasks. On the other hand, using a fully decentralized approach leads to environment non-stationarity, since the state transition of an individual agent may be distorted by the actions of other agents. To address it, we propose a design of the state and reward function that increases the observability of the impact of agents' actions during training. An event-based mesoscopic simulation model is built to train the agents. We evaluate the proposed approach in a case study with a complex route from the Chicago transit network. The proposed method is compared to a standard headway-based control and a policy trained with MARL but with no cooperative learning. The results show that the proposed method not only improves level of service but it is also more robust towards uncertainties in operations such as travel times and operator compliance with the recommended action. • A novel framework is proposed for joint holding and stop-skipping controls. • The framework is decentralized and considers nearby agents for evaluation. • Simulation experiments are conducted for a busy route operation in Chicago. • Metrics relevant to riders and agencies show the superiority of the proposed method. • The method shows robustness to reduced operator compliance. [ABSTRACT FROM AUTHOR]

Freelance drivers in the shared mobility market frequently switch or work for multiple platforms, affecting driver labor supply. Due to the importance of driver labor supply for the shared mobility market, understanding drivers' switching and multi-homing behavior is vital to managing service quality on – and effective regulation of – mobility platforms. However, a lack of individual-level data on driver behavior has thus far impeded a deeper understanding. This paper taxonomizes and estimates perceived switching and multi-homing frictions on mobility platforms. Based on a structural model of driver labor supply, we estimate switching and multi-homing costs in a platform duopoly using public and limited high-level survey data. Estimated costs are sizeable, and reductions in multi-homing and switching costs significantly affect platform market shares and driver welfare. Driver labor supply elasticity with respect to platform wage is also discussed considering both multi-homing and switching frictions. • We categorize different types of multi-homing and switching costs faced by platform drivers. • We propose a structural model of driver labor supply in a TNC oligopoly. • We estimate perceived multi-homing and switching costs using limited survey data. • We evaluate counterfactuals on driver welfare, market share, and supply elasticities. [ABSTRACT FROM AUTHOR]

Highlights • Propose an approach to design, simulate, and evaluate integrated AV and PT system. • Model the interaction between demand and supply through an iterative procedure. • Develop scenarios with various fleet and vehicle sizes, and fare and hailing strategies. • Evaluate scenarios for the passengers, AV and PT operators, and mobility system. Abstract Autonomous vehicles (AVs) represent potentially disruptive and innovative changes to public transportation (PT) systems. However, the exact interplay between AV and PT is understudied in existing research. This paper proposes a systematic approach to the design, simulation, and evaluation of integrated autonomous vehicle and public transportation (AV + PT) systems. Two features distinguish this research from the state of the art in the literature: the first is the transit-oriented AV operation with the purpose of supporting existing PT modes; the second is the explicit modeling of the interaction between demand and supply. We highlight the transit-orientation by identifying the synergistic opportunities between AV and PT, which makes AVs more acceptable to all the stakeholders and respects the social-purpose considerations such as maintaining service availability and ensuring equity. Specifically, AV is designed to serve first-mile connections to rail stations and provide efficient shared mobility in low-density suburban areas. The interaction between demand and supply is modeled using a set of system dynamics equations and solved as a fixed-point problem through an iterative simulation procedure. We develop an agent-based simulation platform of service and a discrete choice model of demand as two subproblems. Using a feedback loop between supply and demand, we capture the interaction between the decisions of the service operator and those of the travelers and model the choices of both parties. Considering uncertainties in demand prediction and stochasticity in simulation, we also evaluate the robustness of our fixed-point solution and demonstrate the convergence of the proposed method empirically. We test our approach in a major European city, simulating scenarios with various fleet sizes, vehicle capacities, fare schemes, and hailing strategies such as in-advance requests. Scenarios are evaluated from the perspectives of passengers, AV operators, PT operators, and urban mobility system. Results show the trade off between the level of service and the operational cost, providing insight for fleet sizing to reach the optimal balance. Our simulated experiments show that encouraging ride-sharing, allowing in-advance requests, and combining fare with transit help enable service integration and encourage sustainable travel. Both the transit-oriented AV operation and the demand-supply interaction are essential components for defining and assessing the roles of the AV technology in our future transportation systems, especially those with ample and robust transit networks. [ABSTRACT FROM AUTHOR]

• A novel train following model inspired from traffic flow theory is developed. • The proposed model is calibrated with the data from the MBTA Red Line. • The model accurately represents train operations under both normal and disrupted conditions. • The model can serve as a quick response tool both in real-time and offline. In this paper we introduce a mesoscopic Train Following Model which accurately captures train interactions and predicts delays based on spacing between consecutive trains. The Train Following Model is applied recursively block by block estimating train trajectories given initial conditions (i.e. the trajectory of an initial train and dispatching headways of following trains from the terminal station). We validate the proposed model using data from the Red Line of the Massachusetts Bay Transportation Authority (MBTA). The results indicate that it accurately represents train operations under both normal and disrupted conditions. Based on the model developed, the impacts of factors such as service frequency, headway variations, passenger demand, and initial train delays on line performance (i.e. line throughput and train knock-on delays) are explored. The proposed Train Following Model is generic and can be developed based on readily available historical train tracking data. It is not as resource intensive as micro simulation models, while it can efficiently address the drawbacks of macro-scale analytical models and complex discrete algebraic models. The proposed model can be used to predict system performance either off-line or in real-time. [ABSTRACT FROM AUTHOR]

• Model how subjective evaluation of existing modes influence autonomous mobility-on-demand (AMOD) adoption. • Model impact of inertia from existing travel modes on AMOD choice. • Find that subjective evaluations and inertia both predict mode choice. • Particularly, positive evaluations and current use of ridehailing are strongly predictive of AMOD choice. As autonomous vehicle (AV) technology advances, it is important to understand its potential demand and user characteristics. Literature from stated preference surveys find that attitudes and current travel behavior are as or more important than demographics in determining intention to purchase or use AVs. Yet to date no study has looked at how attitudes and use of existing modes both simultaneously affect AV adoption. In this study, we conduct a stated preference survey in Singapore to investigate how the subjective evaluation of existing travel modes (attitudes) and inertia based on previous use of existing modes affect the adoption of an autonomous mobility-on-demand service (AMOD). Using a sample size of 2,003 individuals and 11,613 choice observations, we estimate a mixed logit discrete choice model incorporating latent variables capturing subjective evaluations of existing travel modes (determined through confirmatory factor analysis), a two-part formulation of modal inertia, and other trip-specific and socio-demographic variables. Results show that subjective evaluation and use of existing modes both affect the adoption of AMOD. Specifically, people with a positive evaluation of ridehailing and those who are current ridehailing users are more likely to choose AMOD. Additionally, those who are current car drivers are more likely to choose AMOD, while users of public transit were less likely to choose AMOD. Given that ridehailing is the closest existing mode to our hypothetical AMOD service, our results might suggest that how AVs are implemented and their similarity to existing modes may be critical to the formation of attitudes and direction of inertia impacting adoption. Our research provides insights on the potential relationship between AVs and existing modes that could valuable in AV network design and service planning. [ABSTRACT FROM AUTHOR]

• Competition between AV and PT in a first-mile market is evaluated. • Five scenarios with different regulation levels are tested. • Supplies of AV and PT tend to concentrate both spatially and temporally. • AV and PT competition results in higher profits for both operators. • Passengers' generalized costs and system VKT are reduced after competition. Emerging autonomous vehicles (AV) can either supplement the public transportation (PT) system or compete with it. This study examines the competitive perspective where both AV and PT operators are profit-oriented with dynamic adjustable supply strategies under five regulatory structures regarding whether the AV operator is allowed to change the fleet size and whether the PT operator is allowed to adjust headway. Four out of the five scenarios are constrained competition while the other one focuses on unconstrained competition to find the Nash Equilibrium. We evaluate the competition process as well as the system performance from the standpoints of four stakeholders—the AV operator, the PT operator, passengers, and the transport authority. We also examine the impact of PT subsidies on the competition results including both demand-based and supply-based subsidies. A heuristic algorithm is proposed to update supply strategies for AV and PT based on the operators' historical actions and profits. An agent-based simulation model is implemented in the first-mile scenario in Tampines, Singapore. We find that the competition can result in higher profits and higher system efficiency for both operators compared to the status quo. After the supply updates, the PT services are spatially concentrated to shorter routes feeding directly to the subway station and temporally concentrated to peak hours. On average, the competition reduces the travel time of passengers but increases their travel costs. Nonetheless, the generalized travel cost is reduced when incorporating the value of time. With respect to the system efficiency, the bus supply adjustment increases the average vehicle load and reduces the total vehicle kilometers traveled measured by the passenger car equivalent (PCE), while the AV supply adjustment does the opposite. The results suggest that PT should be allowed to optimize its supply strategies under specific operation goals and constraints, and AV operations should be regulated to reduce their system impacts, including potentially limiting the number of licenses, operation time, and service areas, which makes AV operate in a manner more complementary to the PT system. Providing subsidies to PT results in higher PT supply, profit, and market share, lower AV supply, profit, and market share, and increased passenger's generalized cost and total system PCE. [ABSTRACT FROM AUTHOR]

• A time-varying weighted encounter network to model epidemic spreading in PT. • A scalable and lightweight theoretical framework to solve the problem. • Various public health and transportation-related control policies are evaluated. • Partial closure of bus routes cannot fully contain the spreading of epidemics. • Isolating influential passengers" at an early stage can reduce the spreading. Passenger contact in public transit (PT) networks can be a key mediate in the spreading of infectious diseases. This paper proposes a time-varying weighted PT encounter network to model the spreading of infectious diseases through the PT systems. Social activity contacts at both local and global levels are also considered. We select the epidemiological characteristics of coronavirus disease 2019 (COVID-19) as a case study along with smart card data from Singapore to illustrate the model at the metropolitan level. A scalable and lightweight theoretical framework is derived to capture the time-varying and heterogeneous network structures, which enables to solve the problem at the whole population level with low computational costs. Different control policies from both the public health side and the transportation side are evaluated. We find that people's preventative behavior is one of the most effective measures to control the spreading of epidemics. From the transportation side, partial closure of bus routes helps to slow down but cannot fully contain the spreading of epidemics. Identifying "influential passengers" using the smart card data and isolating them at an early stage can also effectively reduce the epidemic spreading. [ABSTRACT FROM AUTHOR]