Enjian Yao
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
Research Subjects
Research Interests
Enjian Yao has a total of 125 co-authors. Here are the top co-authors by number of shared publications:
Yang Yang
Beijing Jiaotong University
10 shared publications- Long Pan
Key Laboratory of Transport Industry of Big Data Application Technologies for Comprehensive Transport, Beijing Jiaotong University, Beijing, China
9 shared publications - Shasha Liu
Key Laboratory of Transport Industry of Big Data Application Technologies for Comprehensive Transport, Beijing Jiaotong University, Beijing 100044, China
8 shared publications Yongsheng Zhang
School of Transportation and Traffic, Key Laboratory of Transport Industry of Big Data Application Technologies for Comprehensive Transport, Beijing Jiaotong University, Beijing, China
8 shared publications- Rongsheng Chen
Key Laboratory of Transport Industry of Big Data Application Technologies for Comprehensive Transport, Beijing Jiaotong University, Beijing, China
6 shared publications - Yang Yang
Beijing Jiaotong University
6 shared publications
Latest Publications
- Open Access
- 25 August 2026
- Published by MDPI AG in Infrastructures
- Vol. 11 (9) , 296
A Timed Petri Net Method to Optimize the Scheduling of a Railway Hub Construction Project
The construction of large-scale buildings often faces extended production cycles due to inefficiencies in scheduling processes. To address this challenge, a timed Petri net model was developed to analyze and optimize construction scheduling. Based on the Petri net transition sequence, a scheduling optimization model was proposed. To solve the model efficiently, an improved brainstorming optimization (BSO) algorithm was introduced. Compared with the classical BSO, two targeted enhancements were introduced: a problem-specific encoding and decoding method for Petri net transition sequences to ensure solution feasibility and an embedded simulated annealing local search mechanism to prevent premature convergence in later iterations. The proposed methodology was validated using real-world data from a large high-speed railway hub foundation pit construction project. Results demonstrated a significant reduction of 531 working hours in the total scheduling time, representing a 15.47% improvement in scheduling efficiency compared to traditional sequential scheduling methods. This approach not only shortened the scheduling cycle and enhanced production efficiency but also offered an innovative solution to address scheduling issues in complex construction processes.
- 25 August 2026
- Published by MDPI AG in Infrastructures
- Vol. 11 (9) , 296
Open AccessA Timed Petri Net Method to Optimize the Scheduling of a Railway Hub Construction Project
The construction of large-scale buildings often faces extended production cycles due to inefficiencies in scheduling processes. To address this challenge, a timed Petri net model was developed to analyze and optimize construction scheduling. Based on the Petri net transition sequence, a scheduling optimization model was proposed. To solve the model efficiently, an improved brainstorming optimization (BSO) algorithm was introduced. Compared with the classical BSO, two targeted enhancements were introduced: a problem-specific encoding and decoding method for Petri net transition sequences to ensure solution feasibility and an embedded simulated annealing local search mechanism to prevent premature convergence in later iterations. The proposed methodology was validated using real-world data from a large high-speed railway hub foundation pit construction project. Results demonstrated a significant reduction of 531 working hours in the total scheduling time, representing a 15.47% improvement in scheduling efficiency compared to traditional sequential scheduling methods. This approach not only shortened the scheduling cycle and enhanced production efficiency but also offered an innovative solution to address scheduling issues in complex construction processes.
- 23 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. PP (15249050) , 1-15
AR2G-AIRL: A Context-Adaptive Game-Inverse Reinforcement Learning Algorithm for Modeling Pedestrian–Vehicle Interaction Considering Group Behaviors
Pedestrian-vehicle interactions are critical problems in the development of autonomous driving technologies. However, there is still a lack of comprehensive understanding of pedestrian behaviors during pedestrian-vehicle interactions, especially considering the impact of group behaviors. This study proposes an Actor-Relation-Graph-based Game-Adversarial Inverse Reinforcement Learning (AR2G-AIRL) algorithm to understand pedestrian behaviors during pedestrian-vehicle interactions, considering the influence of surrounding pedestrians and vehicles. Specifically, we first identify pedestrian-vehicle interactions from a naturalistic driving dataset as the data basis. Subsequently, the AR2G-AIRL algorithm is constructed with three components: Actor-Relation Graphs (ARGs) to capture the features within groups, evolutionary game theory to extract the features between groups, and the Adversarial Inverse Reinforcement Learning (AIRL) algorithm to recover the reward function. The experiment results illustrate that the AR2G-AIRL algorithm can accurately simulate pedestrian behaviors during pedestrian-vehicle interactions and outperform all the baseline methods and ablation studies. Meanwhile, the trajectory and ARG weight visualization results demonstrate the importance of surrounding pedestrians under diverse interactions. Moreover, the analysis of the recovered reward function validates the “safety in numbers” effect and reveals the basic patterns of pedestrians interacting with vehicles without surrounding pedestrians and the heterogeneous impact of the size and morphologies of pedestrian groups on pedestrian behaviors. These results highlight the importance of incorporating group behaviors, offering insights into enhancing road safety and the efficiency of autonomous systems.
- 23 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. PP (15249050) , 1-15
AR2G-AIRL: A Context-Adaptive Game-Inverse Reinforcement Learning Algorithm for Modeling Pedestrian–Vehicle Interaction Considering Group Behaviors
Pedestrian-vehicle interactions are critical problems in the development of autonomous driving technologies. However, there is still a lack of comprehensive understanding of pedestrian behaviors during pedestrian-vehicle interactions, especially considering the impact of group behaviors. This study proposes an Actor-Relation-Graph-based Game-Adversarial Inverse Reinforcement Learning (AR2G-AIRL) algorithm to understand pedestrian behaviors during pedestrian-vehicle interactions, considering the influence of surrounding pedestrians and vehicles. Specifically, we first identify pedestrian-vehicle interactions from a naturalistic driving dataset as the data basis. Subsequently, the AR2G-AIRL algorithm is constructed with three components: Actor-Relation Graphs (ARGs) to capture the features within groups, evolutionary game theory to extract the features between groups, and the Adversarial Inverse Reinforcement Learning (AIRL) algorithm to recover the reward function. The experiment results illustrate that the AR2G-AIRL algorithm can accurately simulate pedestrian behaviors during pedestrian-vehicle interactions and outperform all the baseline methods and ablation studies. Meanwhile, the trajectory and ARG weight visualization results demonstrate the importance of surrounding pedestrians under diverse interactions. Moreover, the analysis of the recovered reward function validates the “safety in numbers” effect and reveals the basic patterns of pedestrians interacting with vehicles without surrounding pedestrians and the heterogeneous impact of the size and morphologies of pedestrian groups on pedestrian behaviors. These results highlight the importance of incorporating group behaviors, offering insights into enhancing road safety and the efficiency of autonomous systems.
- 12 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 27 (2) , 2268-2280
Human-Like Decision Making: A Human–Machine Shared Steering Control System With Adaptive Authority Allocation
Aiming at the dynamic conflict and driving authority allocation problems in the human-automation shared steering control (SSC) system, this paper proposed a framework for decision-making and control integration that considering risk perception and game theory. Firstly, by introducing the risk perception field, the complex and variable driving risk is precisely quantified, and the drivers are divided into skilled, normal and unpracticed according to their driving skill. Taking the driving risk acceptance as the core element, combined with the driving skill, a two-layer architecture integrating decision-making and control is constructed. The upper path planning module efficiently generates candidate trajectories based on preview time and driving risk acceptance. The lower control module uses model predictive control (MPC) and non-cooperative Stackelberg game to realize steering interaction while ensuring the priority of driver’s intention. Meanwhile, a dynamic authority allocation strategy based on driving risk and driving skill is established. Finally, to fully verify the effectiveness of the proposed method, two double lane change (DLC) scenarios with the same and different reference trajectories were designed, and four evaluation indicators were proposed, including comfort and driving conflict. The experimental results show that compared with the Nash game, the Stackelberg-based method can guarantee the driver’s dominance, alleviate the human-automation conflict, reduce the driver’s workload and increase the driving smoothness.
- 12 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 27 (2) , 2268-2280
Human-Like Decision Making: A Human–Machine Shared Steering Control System With Adaptive Authority Allocation
Aiming at the dynamic conflict and driving authority allocation problems in the human-automation shared steering control (SSC) system, this paper proposed a framework for decision-making and control integration that considering risk perception and game theory. Firstly, by introducing the risk perception field, the complex and variable driving risk is precisely quantified, and the drivers are divided into skilled, normal and unpracticed according to their driving skill. Taking the driving risk acceptance as the core element, combined with the driving skill, a two-layer architecture integrating decision-making and control is constructed. The upper path planning module efficiently generates candidate trajectories based on preview time and driving risk acceptance. The lower control module uses model predictive control (MPC) and non-cooperative Stackelberg game to realize steering interaction while ensuring the priority of driver’s intention. Meanwhile, a dynamic authority allocation strategy based on driving risk and driving skill is established. Finally, to fully verify the effectiveness of the proposed method, two double lane change (DLC) scenarios with the same and different reference trajectories were designed, and four evaluation indicators were proposed, including comfort and driving conflict. The experimental results show that compared with the Nash game, the Stackelberg-based method can guarantee the driver’s dominance, alleviate the human-automation conflict, reduce the driver’s workload and increase the driving smoothness.
- 9 January 2026
- Published by Elsevier in Transportation Research Part D: Transport and Environment
- Vol. 153
Dynamic planning integrating life cycle assessment for taxi battery swapping stations
- 9 January 2026
- Published by Elsevier in Transportation Research Part D: Transport and Environment
- Vol. 153
Dynamic planning integrating life cycle assessment for taxi battery swapping stations
- Open Access
- 1 January 2026
- Published by Wiley in Journal of Advanced Transportation
- Vol. 2026 (1)
Dynamic Evolution of Multilevel Rail Transit Destination Choice Behavior Based on Disaggregate–Reinforcement Learning
To address the distribution of origin–destination (OD) passenger flow in multilevel rail transit complex network, we propose a simulation method based on travel behavior and reinforcement learning for travelers’ multilevel rail transit network destination choice behavior. First, we extended the node–place (NP) model into node–place–ridership (NPR) model as the ridership and the level of rail transit are introduced into the indicators. The place value is calculated based on the difference in the catchment area of stations at each level to construct an indicator system for stations. From the perspective of traveler behavior, a destination choice model is developed using the station’s NPR variables. Then, the Bush–Mosteller (BM) reinforcement learning model is employed to simulate the evolutionary mechanism of travelers’ destination choice behavior under the influence of group decision‐making characteristics, thereby designing an OD matrix convergence algorithm. The methodology’s validity is demonstrated through a case study involving a multilevel rail transit network in Guangzhou and Foshan, Guangdong Province. Results indicate that the approach effectively captures the choice behavior mechanisms of travelers within incomplete information contexts. The evolution of OD passenger flows between intracity, suburban city, and intercity stations exhibits distinct patterns; however, OD volumes between stations spanning different hierarchical levels remain consistently low (lower than 300) and stable. During the propagation and diffusion of travel‐related information throughout the network, the positive effect exerted by dominant stations demonstrates significantly greater transmission intensity than the negative effect associated with nondominant stations. The estimated result of the same‐level dummy in the destination choice model (0.527–0.825) indicates that the interchange barrier between interlevel OD is a major factor affecting the travel of multilevel rail users.
- 1 January 2026
- Published by Wiley in Journal of Advanced Transportation
- Vol. 2026 (1)
Open AccessDynamic Evolution of Multilevel Rail Transit Destination Choice Behavior Based on Disaggregate–Reinforcement Learning
To address the distribution of origin–destination (OD) passenger flow in multilevel rail transit complex network, we propose a simulation method based on travel behavior and reinforcement learning for travelers’ multilevel rail transit network destination choice behavior. First, we extended the node–place (NP) model into node–place–ridership (NPR) model as the ridership and the level of rail transit are introduced into the indicators. The place value is calculated based on the difference in the catchment area of stations at each level to construct an indicator system for stations. From the perspective of traveler behavior, a destination choice model is developed using the station’s NPR variables. Then, the Bush–Mosteller (BM) reinforcement learning model is employed to simulate the evolutionary mechanism of travelers’ destination choice behavior under the influence of group decision‐making characteristics, thereby designing an OD matrix convergence algorithm. The methodology’s validity is demonstrated through a case study involving a multilevel rail transit network in Guangzhou and Foshan, Guangdong Province. Results indicate that the approach effectively captures the choice behavior mechanisms of travelers within incomplete information contexts. The evolution of OD passenger flows between intracity, suburban city, and intercity stations exhibits distinct patterns; however, OD volumes between stations spanning different hierarchical levels remain consistently low (lower than 300) and stable. During the propagation and diffusion of travel‐related information throughout the network, the positive effect exerted by dominant stations demonstrates significantly greater transmission intensity than the negative effect associated with nondominant stations. The estimated result of the same‐level dummy in the destination choice model (0.527–0.825) indicates that the interchange barrier between interlevel OD is a major factor affecting the travel of multilevel rail users.
- Open Access
- 18 December 2025
- Published by MDPI AG in Urban Science
- Vol. 9 (12) , 546
Personalized Guidance Information and Travel Choice Behavior During Metro Service Disruptions: Evidence from Beijing, China
Guidance information plays an important role in influencing metro passengers’ travel choices and enhancing their travel experience during unplanned service disruptions. However, limited research has examined passengers’ behavioral responses to personalized guidance information in such contexts. This study aims to fill the gap and explore the impact of personalized guidance information on passengers’ travel choice behavior during unplanned metro service disruptions. First, we reconstruct the decision-making process of metro passengers under disruption scenarios and design personalized guidance strategies, followed by a stated preference survey to collect preference data. Using data from Beijing, China, a hybrid utility–regret model is developed to analyze how the content and frequency of personalized guidance information affect passengers’ travel choice preferences. The results show that recommended plans with explanatory information are more likely to be adopted, particularly when explanations are framed from the passenger’s perspective. A single notification serves as a timely reminder, whereas overly frequent messages may trigger annoyance and reduce effectiveness. These findings provide practical implications for the design of personalized guidance strategies, thereby mitigating the impacts of metro service disruptions.
- 18 December 2025
- Published by MDPI AG in Urban Science
- Vol. 9 (12) , 546
Open AccessPersonalized Guidance Information and Travel Choice Behavior During Metro Service Disruptions: Evidence from Beijing, China
Guidance information plays an important role in influencing metro passengers’ travel choices and enhancing their travel experience during unplanned service disruptions. However, limited research has examined passengers’ behavioral responses to personalized guidance information in such contexts. This study aims to fill the gap and explore the impact of personalized guidance information on passengers’ travel choice behavior during unplanned metro service disruptions. First, we reconstruct the decision-making process of metro passengers under disruption scenarios and design personalized guidance strategies, followed by a stated preference survey to collect preference data. Using data from Beijing, China, a hybrid utility–regret model is developed to analyze how the content and frequency of personalized guidance information affect passengers’ travel choice preferences. The results show that recommended plans with explanatory information are more likely to be adopted, particularly when explanations are framed from the passenger’s perspective. A single notification serves as a timely reminder, whereas overly frequent messages may trigger annoyance and reduce effectiveness. These findings provide practical implications for the design of personalized guidance strategies, thereby mitigating the impacts of metro service disruptions.