LPLong Pan

Long Pan

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Long Pan has a total of 31 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Investigating the Role of POIs in the Relationship Between Bike‐Sharing and Public Transportation: Evidence From Beijing, China, Using Neural Networks and SHAP

    by Long Pan, Ziyu Liu, Enjian Yao, Rui Chen

    This study investigates how points of interest (POI) shape the interaction between dockless bike‐sharing (DBS) and public transportation. Using Beijing, China, as a case study, we propose a data‐driven analytical framework that delineates public transportation service buffers by incorporating station‐level passenger inflow and outflow characteristics. Based on these adaptive buffers, DBS trips are classified into competition, cooperation‐first‐mile, cooperation‐last‐mile and complement types. Grid‐level DBS trip volumes are then predicted using fully connected neural networks, and Shapley Additive Explanations are employed to interpret the influence of different POI categories. The results show that model performance varies across DBS trip types and improves with deeper neural network architectures. Competition between DBS and public transportation during peak hours is primarily associated with short‐distance, lifestyle‐oriented travel around residential and dining areas, whereas cooperation is strongly linked to access and egress functions related to daily activity chains. Parking‐intensive residential areas emerge as high‐potential zones for promoting DBS‐public transportation integration, indicating that DBS can effectively mitigate first‐ and last‐mile barriers in car‐oriented neighbourhoods. These findings provide insights for transit‐oriented planning and DBS deployment, suggesting that reallocating limited urban space toward integrated DBS‐public transportation facilities can support more sustainable urban mobility.

  • Deep Learning Model for Short-Term Origin–Destination Distribution Prediction in Urban Rail Transit Network Considering Destination Choice Behavior

    by Yue Wang, Enjian Yao, Yongsheng Zhang, Long Pan, He Hao

    Urban rail transit (URT) has emerged as a crucial mode of transportation in metropolitan areas. For the effective operation of expanding URT networks, accurate short-term origin–destination (OD) demand distribution predictions are essential. This study introduces a novel deep-learning-based model for predicting short-term OD distribution in extensive networks, taking destination choice behaviors into account. First, we perform a comprehensive analysis of station passenger flows and OD flows from both temporal and spatial dimensions. Then, we develop the origin–destination distribution prediction (ODDP) model, combining the destination choice model (DCM) with the deep learning model (DLM). The DCM aims to understand OD distribution patterns from a behavioral perspective by transforming real-time inflows into OD distributions. Meanwhile, the DLM, employing attention and convolution layers, effectively captures the intricate temporal and spatial dynamics of passenger flows. Our model is evaluated using data from the Guangzhou Metro network in China, showing significant enhancements in prediction accuracy, model interpretability, and overall robustness. The implementation of our model promises substantial benefits for the operational efficiency of URT systems.


  • Dynamic Passenger Route Guidance in the Multimodal Transit System With Graph Representation and Attention Based Deep Reinforcement Learning

    by He Hao, Enjian Yao, Rongsheng Chen, Long Pan, Yue Wang

    Recently, the limited capacity of the Urban Rail Transit (URT) has failed to meet passenger travel demands, especially in peak hours, which leads to crowded stations and oversaturated metro networks. Considering the diverse scenarios of the multimodal public transportation network and the need for rapid generation of strategies, this study proposes a deep reinforcement learning (RL) algorithm that guides passenger route selection in the multimodal transit network to operate better. The proposed RL algorithm, consisting of graph representation learning, convolution neural network, and self-attention mechanism, is used to generate the route guidance strategy for passengers, aiming to alleviate the congestion of the multimodal transit network, improve passengers’ travel experience, and reduce CO $_{2}$ emissions. Based on the multimodal transit network in Beijing, the simulation results demonstrate that the RL algorithm can well perceive the states and generate adaptive route guidance strategies that can decrease the section load rates, improve network-wide passengers’ travel experience, and reduce CO $_{2}$ emissions. Even in cases where passengers are not fully compliant with the route guidance, the proposed algorithm remains effective.


  • Open Access

    Investigating the Willingness of Shifting to MaaS in One-Trip Scenarios: Insights From Comparative Stated Surveys

    by Enjian Yao, He Hao, Long Pan, Rongsheng Chen, Yue Wang, Hui Xiao

    Considered as a potential solution to reduce carbon emissions, Mobility-as-a-Service (MaaS) is crucial for developing eco-friendly traffic systems. However, how to guide users to shift to MaaS at its early stage still needs more insights. Motivated by this, we conducted a stated preference survey in Beijing with a novel experiment manner, which provides respondents with comparative scenarios to ask them to focus on differences between MaaS options and their previous choice. Further, a latent-class-based model is developed, integrating travel habit and cross-scenario variables collected from the comparative SP experiments. According to the estimation results, we obtain two classes. The first represents respondents who tend to be familiar with MaaS and use shared mobility, and the other represents car enthusiasts. With respect to the preferences of these two classes, respondents in the former class may generate higher utility for all MaaS options. That said, they also tend to keep their mode choice in the scenario without MaaS when they choose private cars or taxis. On the contrary, respondents in the latter class may be negative about all MaaS options. Despite this, the previous mode choice of public transportation (PT) positively affects choosing MaaS options including PT. Further, we propose the shifting to MaaS index to quantify willingness to shift to MaaS. Our findings demonstrate that respondents in class 1 are much more likely to shift to MaaS and decreasing trip time does not have a strong effect on shifting to MaaS. Accordingly, some insights and suggestions are proposed for MaaS development, including cities suitable for MaaS, promotional strategies, and resource-optimization strategies.

  • Public Transit Itinerary Choice Analysis Considering Various Incentives

    by Long Pan, E.O.D. Waygood, Zachary Patterson

    Previous research on public transport user preferences has primarily focused on mode choice, transfer penalties, and preferences related to service qualities. Individuals are generally found to not like waiting, transfers, or overcrowding. However, from an operation perspective, it could be important to convince users to change to a different itinerary to reduce overcrowding, which can cause problems such as bunching. This would most likely require convincing the user to voluntarily accept a longer wait or to take a different route potentially with a transfer. One means of doing so might be through incentives or offering a shared taxi. Incentivizing transit users to switch to a less-crowded itinerary, which could potentially be a way to manage bus overcrowding, is not well-studied. Through a discrete choice experiment, this study thus filled this research gap by testing people’s willingness to change routes or to take a shared taxi in exchange for one of three distinct types of incentives. A mixed binary logit model and a mixed nested logit model were used to investigate how different factors influence public transit user itinerary choice behavior. The results suggested that some people would be willing to change to the proposed alternative. Heterogeneity within the responses was seen as transit users were found to have distinct preferences for different incentives, and the value of the incentives influenced users differently. Our models and results could be helpful for public transit agencies to improve efficiency and reduce costs by better balancing the travel demand.


  • Would You Wait? Bus Choice Behavior Analysis Considering Various Incentives

    by Long Pan, E. O. D. Waygood, Zachary Patterson

    During the peak hour, ridership is much higher, and this can lead to problems such as overcrowding for customers and vehicle bunching for the operator as dwell times increase for the lead bus. In this paper, we examined whether to avoid overcrowding people might be willing to wait for the next bus knowing there were seats and if they were offered an incentive. Three distinct types of incentive were offered, defined according to goal-framing theory. To obtain the choice data, a discrete choice experiment was developed and incorporated into an online survey that was distributed to public transport users. A binary logit model and a mixed multinomial logit model were used to investigate how different factors influence public transport users’ bus choice behavior. Results show several considerations that increase the likelihood of users agreeing to the request to wait. These include the weather being sunny/fine, not very cold, or both, the purpose being seeing a friend or shopping, longer in-vehicle time, shorter proposed bus waiting time, and the incentive. The results of the mixed multinomial logit model show variation among the respondents with regard to the incentives. We found that age, gender, work status, possession of a driver ID, and vehicle availability are significant predictors for incentive preference. In that our models target specific groups of users with tailored incentives, the results indicate how to persuade public transport users to avoid overcrowded buses.


  • Open Access

    Forecasting Passenger Flow Distribution on Holidays for Urban Rail Transit Based on Destination Choice Behavior Analysis

    by Enjian Yao, Junyi Hong, Long Pan, Binbin Li, Yang Yang, Dongbo Guo

    Passenger travel flows of urban rail transit during holidays usually show distinct characteristics different from normal days. To ensure efficient operation management, it is essential to accurately predict the distribution of holiday passenger flow. Based on Automatic Fare Collection (AFC) data, this paper explores the passengers’ destination choice differences between normal days and holidays, as well as one-way tickets and public transportation cards, which provides support for variable selection in modeling. Then, a forecasting model of holiday travel distribution is proposed, in which the destination choice model is established for representing local and nonlocal passengers. Meanwhile, explanatory variables such as land matching degree, scenic spot dummy, and level of service variables are introduced to deal with the particularity of holiday passengers’ travel behavior. The parameters calibrated by the improved weighted exogenous sampling maximum likelihood (WESML) method are applied to predict passenger flow distribution in different holiday cases with annual changes in the metro network, using the data collected from Guangzhou Metro, China. The results show that the proposed model is valid and performs better than the other comparable models in terms of forecasting accuracy. The proposed model has the capability to provide a more universal and accurate passenger flow distribution prediction method for urban rail transit in different holiday scenarios with network changes.