Search

• A public transit accessibility under heat exposure model is studied with time-dependent consideration. • A column generation method to urban rail-bus-walk service network integration is devised. • Experiments in Beijing show the impact of heat exposure on public transit accessibility. The evolution of urban transportation systems is increasingly driven by the integration of rail, bus, and walking pathways, forming a cohesive service network. This urban rail-bus-walk service network integration is pivotal in addressing urban challenges and ensuring seamless mobility. Nevertheless, the effectiveness of such integrated systems depends on their accessibility and the willingness of individuals to take them as their primary transportation mode. In the context of climate change, extreme heat events pose an increasing threat to passenger comfort and health, which in turn, affects individuals' decisions to use public transit. Additionally, walking to transit stops or waiting for transportation becomes more challenging during such extreme heat conditions. This paper tackles the challenge of comprehensively addressing the evolving issues related to public transit accessibility in response to rising temperatures, an area of research that has received limited attention until now. This paper introduced an approach aimed at optimizing the urban rail-bus-walk service network with a focus on both accessibility and heat exposure considerations. By calculating the transportation accessibility level under varying temperature conditions and identifying its vulnerability to high temperatures, this paper presents a methodology to enhance the transit service. Such improvement involves not only adjustments to the integration network but also considering of cooling facilities, such as shaded routes and shelters. Column generation is applied to solve this problem and the optimization result demonstrates improvement of accessibility under the extreme heat condition. [ABSTRACT FROM AUTHOR]

In many densely populated mega-cities around the world, metro systems are becoming overly saturated by the ever-increasing travel demand, which results in overcrowded stations, excessive delay, and unsafe anxieties among riders. The concept of extra-long trains (i.e., trains longer than station platforms) was recently proposed in the literature as a promising way to increase a metro line's capacity without additional infrastructure construction. This paper develops a general modeling framework to optimize train block configuration and operation protocols of a metro line so that extra-long trains can be used under varying demand distribution and infrastructure setting. The design problem is formulated as the integration of two coupled vehicle routing problems which simultaneously optimizes the train block configuration plan, door opening strategy, stop skipping strategy, and train dispatch schedule. This paper proposes two customized solution methods, including an adapted savings heuristic and a destroy-and-repair algorithm. A series of hypothetical examples are tested to demonstrate how the proposed solution approaches outperform an existing commercial solver even for small to moderate problem instances. Moreover, two real-world case studies, with very different demand patterns from two continents, are presented to test the effectiveness of using extra-long trains. Our results show that a metro line with extra-long train operations is capable of serving up to 20%–30% more passenger demand as compared to that with only regular trains, while at the same time producing a similar or lower passenger average travel time. • Design of extra-long trains (i.e., longer than station platforms) in a general setting. • Mixed-integer program to optimize train block configuration and operation protocols. • Customized solution methods (adapted savings and destroy-and-repair). • Extensive experiments with hypothetical and real-world data (from Beijing and Santiago). • Results show significantly increase in metro line capacity without infrastructure construction. [ABSTRACT FROM AUTHOR]

Shared Autonomous Electric Vehicles (SAEVs) are pivotal for future transportation, offering both promise and challenges upon integration with the power grid. This symbiosis augments power system flexibility, stability and reliability through Vehicle-to-Grid (V2G) services, and optimize transportation efficiency. However, it amplifies the demand for robust charging infrastructure and electricity power during peak periods. This paper proposes a framework employing a sequential receding horizon optimization approach to manage SAEV mobility and charging dynamics. Focused on maximizing transportation service quality while ensuring power grid stability, the model accommodates dynamic trip requests and electricity generation, utilizing a rolling horizon algorithm. Notably, the study explores the potential of SAEVs in fortifying the integration of renewable energy resources (RES) into the power grid. Our research strives to equip policymakers and system planners with a robust tool for crafting efficient and sustainable future urban transportation and energy systems. • A novel modeling and optimization framework for SAEV-PowerGrid coupled system. • Applied a rolling horizon algorithm to solve the problem with realistic sequential and dynamic input. • Case study applications for downtown Beijing. • Revealed SAEV's significant role in integrating renewable energy resources into the powergrid. [ABSTRACT FROM AUTHOR]

Integrating emerging shared mobility with traditional fixed-line public transport is a promising solution to the mismatch between supply and demand in urban transportation systems. The advent of modular vehicles (MVs) provides opportunities for more flexible and seamless intermodal transit. The MVs, which have been implemented, are comprised of automated modular units (MUs), and can dynamically change the number of MUs comprising them at different times and stops. However, this innovative intermodal urban transit brings with it a new level of dynamism and uncertainty. In this paper, we study the problem of jointly optimizing the timetable and the vehicle schedule within an intermodal urban transit network utilizing MVs within the context of distributionally robust optimization (DRO), which allows MVs to dynamically (de)couple at each stop and permits flexible circulations of MUs across different transportation modes. We propose a DRO formulation to explore the trade-off between operators and passengers, with the objective of minimizing the worst-case expectation of the weighted sum of passengers' and operating costs. Furthermore, to address the computational intractability of the proposed DRO model, we design a discrepancy-based ambiguity set to reformulate it into a mixed-integer linear programming model. In order to obtain high-quality solutionss of realistic instances, we develop a customized decomposition-based algorithm. Extensive numerical experiments demonstrate the effectiveness of the proposed approach. The computational results of real-world case studies based on the operational data of Beijing Bus Line illustrate that the proposed integrated timetabling and vehicle scheduling method reduces the expected value of passengers' and operating costs by about 6% in comparison with the practical timetable and fixed-capacity vehicles typically used in the Beijing bus system. • We address the timetabling and vehicle scheduling problem concerning modular units. • We formulate a distributionally robust model for the investigated problem. • We develop an algorithm combining ALNS and GUROBI to generate high-quality solutions. • We verify the effectiveness of our proposed approaches by numerical experiments. [ABSTRACT FROM AUTHOR]

• A multitask learning model for short-term inflow prediction of multi-traffic modes is proposed. • The adaptive matrix and attention mechanism are used to extract the spatiotemporal features of multi-traffic modes. • We analyze the interaction mechanism in the multi-traffic modes system. • The research highlights the significance of jointly predicting the inflow of multi-traffic modes. Network-wide short-term inflow prediction is important in efficiently managing the urban transportation system. Nowadays, all kinds of traffic modes gradually become interconnected and form a complex multi-traffic modes system, while extensive studies focus on the single-traffic mode and ignore the correlations among different traffic modes. There exist some challenges for short-term inflow prediction of multi-traffic modes: (1) the interaction mechanism among multi-traffic modes is difficult to learn and few studies explore the mechanism, (2) the data of multi-traffic modes are usually heterogenous due to the different spatial units of different traffic modes, and (3) it is challenging to extract the complex and dynamic features of the multi-traffic modes and most existing methods apply static spatiotemporal correlations among multi-traffic modes, while the genuine correlations among different traffic modes might be missing. To tackle these challenges, this study proposed a multitask-learning-based model called M ulti M ode-former (M2-former) with the encoder-decoder structure for network-wide short-term inflow prediction of the multi-traffic modes system. Specifically, the encoder is designed to learn and capture the complex and dynamic spatiotemporal correlations of multi-traffic modes, and the decoder is designed to extract the features of the target traffic mode and share knowledge among multi-traffic modes. Extensive experiments are conducted based on the real-world multi-traffic modes system data of Beijing, China. Results prove the superiority of the M2-former. In addition, the spatial and temporal information interaction mechanisms among multi-traffic modes are also explored. This paper can provide a reliable method and critical insights for the management and understanding of a multi-traffic modes system. [ABSTRACT FROM AUTHOR]

• A two-population 3-strategy evolutionary game model for subway network operation extension. • Theoretical analysis and numerical simulation of Evolutionary Equilibria for the replicator dynamics system. • The best long-term strategies of networks and key factors (motivations) analysis. • An extended game model for investigating the impact of adjustment minutes. With the growing demand for night travel and the development of the night-time economy, extending the operation time of metropolitan subway networks has become essential. This paper proposes a 2-population 3-strategy evolutionary game model for subway network operation extension decisions, which considers the interests of subway companies, passengers, and the government while addressing transfer failure between lines due to uncoordinated subway operation time adjustments. The study resolves the Evolutionary Equilibria (EEs) of the replicator dynamics system and identifies key factors influencing the evolutionary path using theoretical calculations and a ternary phase diagram. Additionally, an extended game model investigates the impact of adjustment minutes. To verify the models, a case study based on the Beijing subway network is conducted. We show that the rule of maximum total fitness ensures the priority of EE strategies for all lines, selecting Nash equilibrium strategies for the lines which cannot select EE in a vastly reduced network strategy profile space (from 729 to 2). The study concludes that combining subsidy measures can eliminate multiple equilibria produced by implementing only penalty measures, and combining penalty measures can have the effect of making small subsidies as effective as large subsidies. The variation of the satisfaction/loyalty affects the system equilibrium only when both game players value passenger satisfaction/loyalty. Finally, we find that the proper adjustment minutes not only can make operation extension a long-term best strategy but also increase the payoff (106,158 RMB/year) and enhance the occupancy rate (20 %) of the last train. The study provides a theoretical foundation for the government to regulate the operation extension of subway companies from multiple perspectives. [ABSTRACT FROM AUTHOR]

Carefully coordinating train timetables of different operating lines can help reduce transfer delays, which in turn reduces station crowding and improves overall service quality. This paper explores the optimization to train timetable and skip-stop plans that aims to minimize the total passenger waiting time and station crowding. The problem is formulated as a mixed-integer non-linear programming model. To effectively address the complexity of the model, a decomposition and approximate dynamic programming approach is designed to convert the original network-level problem into a series of small-scale subproblems, one for each operating line, to be solved quickly in a distributed manner. The effectiveness and practicability of the model and algorithm are demonstrated on two case networks: a small-scale synthetic network of three metro lines and a real-world network based on Beijing metro. The computational results illustrate that the proposed strategy to generate train timetables and skip-stop plans can effectively reduce passenger waiting time and station crowing. The proposed decomposition and approximate dynamic programming approach is also shown to perform more efficiently than traditional heuristic algorithms, such as genetic algorithm and simulated annealing algorithm for large-scale networks. • The optimization of train timetable and skip-stop plans of metro networks is considered. • A mixed-integer nonlinear programming model is formulated for this problem. • A decomposition and approximate dynamic programming approach is designed. • The optimized strategy reduces passenger waiting time and station crowing. [ABSTRACT FROM AUTHOR]

• A network-based passenger flow control method is proposed for the oversaturated metro network. • A bi-objective integer linear programming model is formulated. • The equivalent single objective linear integer programming model is deduced. • Numerical examples demonstrate the effectiveness of the proposed approaches. The oversaturated situation is now very common in the metro system of some megacities due to large commuting demands in peak hours, which leads to passenger accumulation on platforms and causes potential accident risks. To improve the transport efficiency and passenger accumulation safety at each station, this paper proposes a cooperative passenger flow control optimization in a specific metro network, in which each passenger can freely switch trains between different metro lines with a single ticket. Through considering the dynamic characteristics and transfer behaviors of passengers, an effective bi-objective integer linear programming model is formulated to characterize the passenger control process, in which the objectives are to minimize the total passenger waiting time and passenger accumulation risks at all the involved stations. To solve the proposed model conveniently, the above model is transformed into a single-objective model through reformulating the risk objective as a threshold-based constraint. Finally, two sets of numerical experiments, including a small-scale case and a real-world instance with operation data of the Beijing metro system, are implemented to demonstrate the performance and effectiveness of the proposed approaches. [ABSTRACT FROM AUTHOR]

• A two-layer deep learning structure based on the LSTM and the GRU was developed to capture the spatio-temporal features of the travel speed of lanes. • Real-world traffic speed data of different lanes in Beijing was used to validate the proposed FDL model. • The proposed FDL model outperforms both the traditional and the state-of-the-art benchmark algorithms, including two deep learning methods. • The predicted speeds of the FDL model are most robust during the 24 h among NN-based benchmark models. Accurate and robust short-term traffic prediction is an important part of advanced traveler information systems. With the development of intelligent navigation and autonomous driving, it is necessary to explore the lane-level predictions of traffic speeds. However, most existing traffic prediction models concentrate on forecasting the traffic flow characteristics of the entire road sections rather than those of certain lanes. This paper proposes a fusion deep learning (FDL) model to predict lane-level traffic speed. First, the entropy-based grey relation analysis is introduced to choose lane sections that are strongly correlated with the lane section to be predicted. Second, a two-layer deep learning framework is established by combining the long short-term memory (LSTM) neural network and the gated recurrent unit (GRU) neural network. Third, the ground-truth data of several lane sections captured by remote traffic microwave sensors (RTMS) on the 2nd Ring Road of Beijing are utilized to examine the FDL model and compare it with several benchmark models. The experimental result indicates that in addition to capturing the fluctuations of traffic speed at the lane level, the FDL model has better performance than the benchmark models in terms of prediction accuracy and stability. [ABSTRACT FROM AUTHOR]

• Construct a unified model for integrating timetable, bus formation and vehicle scheduling problems. • Schedule autonomous modular vehicles (AMVs) in both directions simultaneously. • Design a decomposition-based iterative solution framework based on Alternating Direction Method of Multipliers (ADMM) algorithm to solve the origin model efficiently. • Solve the subproblem efficiently with a customized time-dependent shortest path algorithm. • Introduce AMVs into public transport systems can reduce overall system costs and increase vehicle utilization. This paper presents a joint optimization of the timetable, bus formation, and vehicle scheduling in a flexible public transport (PT) system that utilizes autonomous modular vehicles (AMVs). In this system, AMVs have the capability to detach or join with each other at intermediate stops along the route to dynamically adjust the bus formation (capacity). To increase vehicle utilization, a flexible scheduling strategy is proposed that allows AMVs to detach from one modular bus and join another modular bus in either direction of a bidirectional line. In particular, the penalty cost for each detachment or joining operation, as well as the limited number of available AMVs is explicitly considered. We formulate a unified model for the integrated optimization of the modular bus service (timetable and bus formation) and vehicle scheduling by introducing two types of decision variables. The objective is to minimize overall system costs, including passenger waiting time cost, operational costs, and detachment/joining penalty costs. The two types of decision variables are coupled by a vehicle resource consistency constraint, ensuring the conformity of the modular bus service and vehicle scheduling decisions. To tackle the complexity of our model, the Alternating Direction Method of Multipliers (ADMM) is employed to decompose it into two subproblems, which can be efficiently solved using a customized forward dynamic programming algorithm and a commercial solver. The model is validated using illustrative examples and a real-world instance from the Beijing Public Transport system, and it is compared with two benchmark models. Our results demonstrate the efficiency of the ADMM-based solution framework for solving the integrated optimization model. Furthermore, our findings indicate that the use of AMVs in PT systems can lead to reduced overall system costs and increased vehicle utilization. [ABSTRACT FROM AUTHOR]

• Robust scheduling strategies of electric buses. • Adverse impacts caused by the trip time stochasticity. • Static and dynamic models. • Branch-and-price framework. • Realistic operations data of the buses in Beijing. The electric bus scheduling problem requires not only satisfying timetable constraints but also considering battery range limitation and vehicles' recharging plans. This paper is devoted to proposing robust scheduling strategies of electric buses to tackle the challenge brought by the stochasticity of urban traffic conditions. To avoid en-route breakdown of electric buses, reduce delay costs, and achieve robustness, we propose both static and dynamic scheduling models. The static model introduces a buffer-distance strategy to tackle the adverse impacts caused by trip time stochasticity, whereas the dynamic model takes advantage of continuously-updated road traffic conditions and periodically reschedules an electric bus fleet during a day's operations. A branch-and-price framework is extended to effectively solve both models. Using the realistic operations data of bus lines in Beijing, we conduct numerical examples to simulate the performances of the proposed models and derive some important insights. As suggested by the numerical results, the proposed models can effectively avoid en-route breakdown, while maintaining cost efficiency. [ABSTRACT FROM AUTHOR]

Congestion in Terminal Maneuvering Area (TMA) in hub airports is the main problem in Chinese air transportation. In this paper we propose a new system to integrated sequence and merge aircraft to parallel runways at Beijing Capital International Airport (BCIA). This system is based on the advanced avionics capabilities. Our methodology integrates a Multi-Level Point Merge (ML-PM) system, an economical descent approaches procedure, and a tailored heuristic algorithm to find a good, systematic, operationally-acceptable solution. First, Receding Horizontal Control (RHC) technique is applied to divide the entire 24 h of traffic into several sub-problems. Then in each sub-problem, it is optimized on given objectives (conflict, deviation from Estimated Time of Arrival (ETA) on the runway and makespan of the arrival flow). Four decision variables are designed to control the trajectory: the entry time, the entry speed, the turning time on the sequencing leg, and the landing runway allocation. Based on these variables, the real time trajectories are generated by the simulation module. Simulated Annealing (SA) algorithm is used to search the best solution for aircraft to execute. Finally, the conflict-free, least-delay, and user-preferred trajectories from the entry point of TMA to the landing runway are defined. Numerical results show that our optimization system has very stable de-conflict performance to handle continuously dense arrivals in transition airspace. It can also provide the decision support to assist flow controllers to handle the asymmetric arrival flows on different runways with less fuel consumption, and to assist tactical controllers to easily re-sequence aircraft with more relaxed position shifting. Moreover, our system can provide the fuel consumption prediction, and runway assignment information to assist airport and airlines managers for optimal decision making. Theoretically, it realizes an automated, cooperative and green control of routine arrival flows. Although the methodology defined here is applied to the airport BCIA, it could also be applied to other airports in the world. [ABSTRACT FROM AUTHOR]

The energy-efficient driving of automatic train operation (ATO) system in high-speed railway systems has attracted more and more attention. Different from the energy-efficient automatic train control problem for metro trains, the traction/braking forces given by high-speed ATO systems are currently not continuous but discrete values corresponding to traction/braking notches due to the limited communication interface between ATO systems and rolling stocks. In addition, since high-speed trains are powered by single-phase alternating current, there is no power supply for train traction in neutral sections, which is introduced to separate two catenaries powered by currents with different phases. In this paper, we propose a discrete-space based mathematical formulation for the energy-efficient automatic train control problem for high-speed trains, where binary variables are introduced to indicate the traction/braking notch selections and to indicate whether a discrete space interval is a neutral section or not. The mathematical formulation is then transformed into a mixed integer linear programming (MILP) model, which can be solved by existing solvers. However, due to the discretization of the space the discrete traction/braking notches, the resulting MILP problem could be infeasible and the optimal driving strategy cannot be obtained. Hence, we present a novel relaxation method, i.e., the final braking notch relaxation method, to relax a few constraints in the final braking phase and to generate a feasible speed profile. Comprehensive experiments are investigated based on Beijing–Harbin high-speed railway line, where the energy consumption of the speed profiles with discrete (i.e., notch-based) traction/braking forces is slightly higher than that with continuous traction/braking forces. Moreover, the impacts of the neutral section locations and the effects of different numbers of traction/braking notches are deeply analyzed. • Energy-efficient automatic train operation for high speed railway lines. • A new optimal train control model with discrete notches and neutral sections. • Practically inspired final braking notch relaxation method to eliminate infeasibility. • The relaxation of discrete notches enhances computational efficiency. • Our model and method were tested based on data of Chinese high-speed railways. [ABSTRACT FROM AUTHOR]

• Developed a macro framework optimized transportation network utility. • Proposed a general passenger utility maximization model considered efficiency and proportional fairness. • Proved the existence of optimal passenger allocation with maximum passenger utility. • Jointly optimize passenger allocation and vehicle schedules by maximizing transportation network utility. • Numerical tests verified the improvement of passengers' fairness and operation's cost. To deal with the contradiction between oversaturated demand and the restricted transport infrastructure and capacity during peak periods in a rail transit system, several passenger flow control methods that include soft pricing incentive, rigorous physical barrier and so on are widely used in current scenarios. However, the traffic disorderliness at the bottleneck and the control unfairness for the flows on specific stations may occur with those methods in the rail transit corridor. To alleviate those problems, this paper focus on the passenger allocation and matching with the corresponding schedules by considering the fairness of passengers and cost of operation from a macro perspective, where the passenger trip plan can be collected via a reservation platform in advance. A passenger utility maximization model that programs passenger allocation is proposed firstly, and we theoretically prove that there is an optimal traffic flow allocation with maximum utility under the oversaturated flow condition for any network. The model is solved by the Newton iteration algorithm and further proven to effectively make a tradeoff between efficiency and fairness. Furthermore, the transportation network utility maximization model is constructed by jointly optimizing the passenger allocation and vehicle schedules, thereby saving the cost of vehicle operation while guaranteeing the fairness of passenger allocation macroscopically. The model is solved by Multi-tree depth-first search algorithm based on Lagrangian relaxation with a subgradient in an iterative manner. A real-world case from the Batong subway line in Beijing is studied to validate the proposed models and algorithms. The results show that the model can jointly optimize the passenger allocation and vehicle schedules effectively in both an economical and fair manner, and the passengers can pass through the bottleneck in an orderly way. [ABSTRACT FROM AUTHOR]

• Flexible lane usage as on-street parking spaces in connected and automated environments. • Time-dependent design of optimal locations, capacities and opening durations of on-street parking lots. • Balancing the reduction of vehicle miles traveled and the compromise of network mobility. • 10%–27% reduction of total vehicle miles traveled with numerical experiments. Parking remains a serious difficulty in metropolitan transport mainly because of the supply shortage in downtown areas. The parking experience will be improved with the support of connected and automated vehicle (CAV) technologies, enabled by remote parking functionality. Nevertheless, remote parking of CAVs not only results in additional surface traffic but also increases the total vehicle miles traveled (VMT). On the other hand, it is possible to manage the overall traffic by a centralized system with the assistance of CAV technology, which brings the potential for using road resources more flexibly. Inspired by the above ideas, this paper investigates a novel scheme to transfer some lanes to provide on-street parking spaces for CAVs within a necessary period to reduce "cruise-for-parking" travel; as such transfer compromises road capacity, a methodological framework is required to optimally plan the scheme for maximizing certain social goals. We formulate the on-street planning problem as a multi-period mixed-integer linear program with inter-period constraints, which determines the locations and opening durations of on-street parking lots, as well as the assignment of all parking demands across the network. Approximate approaches are adopted to solve the problem for large-scale instances. Numerical experiments are conducted on a real-world network in Wangjing District, Beijing. The results indicate that appropriate time-dependent planning of on-street parking can reduce the total VMT by 10%–27% under various demand levels. [ABSTRACT FROM AUTHOR]

In this paper, we present a case study on planning the locations of public electric vehicle (EV) charging stations in Beijing, China. Our objectives are to incorporate the local constraints of supply and demand on public EV charging stations into facility location models and to compare the optimal locations from three different location models. On the supply side, we analyse the institutional and spatial constraints in public charging infrastructure construction to select the potential sites. On the demand side, interviews with stakeholders are conducted and the ranking-type Delphi method is used when estimating the EV demand with aggregate data from municipal statistical yearbooks and the national census. With the estimated EV demand, we compare three classic facility location models – the set covering model, the maximal covering location model, and the p -median model – and we aim to provide policy-makers with a comprehensive analysis to better understand the effectiveness of these traditional models for locating EV charging facilities. Our results show that the p -median solutions are more effective than the other two models in the sense that the charging stations are closer to the communities with higher EV demand, and, therefore, the majority of EV users have more convenient access to the charging facilities. From the experiments of comparing only the p -median and the maximal covering location models, our results suggest that (1) the p -median model outperforms the maximal covering location model in terms of satisfying the other’s objective, and (2) when the number of charging stations to be built is large, or when minor change is required, the solutions to both models are more stable as p increases. [ABSTRACT FROM AUTHOR]

Aiming to alleviate traffic jams, many traffic management strategies/policies are adopted to nudge travelers to re-arrange their departure time or switch from driving to public transit or non-motorized mode. Analytical travel behavior model is needed to predict travelers’ departure time choice and mode switch under such strategies. In this paper, we developed an agent-based model for travellers’ choices of mode and departure time. Departing from the traditional utility maximization theory, this model focuses on the decision-making process based on imperfect information, bounded and distinctive rationalities. In the modeling framework, travelers accumulate experiences and update their spatial and temporal knowledge through a Bayesian learning process. Before making a trip, travelers decide whether to search for alternative departure time and/or travel mode according to their expected search gain and cost. When an additional search happens, travelers decide whether or not to switch to the new departure time and travel mode according to a series of decision conditions. Both the search and decision processes are represented by production (if–then) rules derived from a joint revealed/stated-preference survey data collected in Beijing. Then the agent-based model is applied to evaluate congestion charge policies with various demand scenarios in the 2nd ring road of Beijing. Results suggest that the model can display the peak spreading and mode switch process practically. [ABSTRACT FROM AUTHOR]

• Propose a physics-informed transfer learning method for network-scale traffic prediction. • Integrate transfer learning algorithms with network MFD properties and physical invariants. • Develop novel domain adaptation mechanism to transfer domain-invariant traffic flow features. • Alleviate existing data insufficiency, dataset shift, and heavy computational cost problems. • Explore network traffic flow pattern transferability and its relation to network traffic properties. Network traffic flow prediction on a fine-grained spatio-temporal scale is essential for intelligent transportation systems, and extensive studies have been carried out in this area. However, existing methods are mostly data-driven, with stringent requirements on the amount and quality of data. The collected network-scale traffic data are expected to be complete, sufficient, and representative, containing most traffic flow patterns in the road network. Unfortunately, it is very rare that sufficient and representative traffic data across the whole road network in several consecutive weeks are available for model calibration. In real-world applications, data insufficiency and dataset shift problems are prevalent, resulting in the 'cold start' issue in traffic prediction. To deal with the challenges above, this paper develops a two-stage physics-informed transfer learning method for network-scale link-wise traffic flow knowledge transfer under MFD-based physical constraints. In the first stage, the road network is partitioned and similar traffic regions are identified according to the physical invariants and MFD characteristics. In this way, the network-scale link-wise traffic flow pattern transfer between similar regions can be initiated under the assumption that regions with similar aggregated traffic flow patterns are more likely to share comparable link-wise traffic flow features. In the second stage, we propose our knowledge transfer architecture Deep Tensor Adaptation Network (DTAN) to bridge traffic flow knowledge in source and target regions via the parallel Siamese network structure, and further reduce domain discrepancy by imposing two distribution adaptation regularizations. A real-world traffic dataset on the urban expressway network of Beijing is used for numerical tests. The experiment results show that the proposed framework can leverage the trade-off between specific regression task performance in a single region and generalized domain adaptation capacity across multiple regions. The data insufficiency, dataset shift, and heavy computational cost problems are alleviated by improving model transferability. Finally, extensive empirical analysis is carried out to explore traffic flow pattern transferability and its relation to network traffic properties. [ABSTRACT FROM AUTHOR]

• Bayesian inference and deep learning is integrated for traffic congestion modeling; • A dynamic Bayesian graph convolutional network (DBGCN) is proposed. • The DBGCN outperforms the state-of-the-art prediction models. • The DBGCN can simulate congestion evolution via dynamic adjacency matrix. • The change of congestion source location leads to different congestion patterns. Congestion, whether recurrent or non-recurrent, propagates through the road network. The process of congestion propagation from a particular road to its neighbors can be regarded as a kind of message passing with a directed relationship. Existing methods have created a solid foundation for characterizing congestion propagation; however, they are either built upon simplified assumptions in traffic flow theory or predefined relationships among road sections, which would lead to downgraded accuracy in practice. This paper proposes a dynamic Bayesian graph convolutional network (DBGCN), which integrates Bayesian inference into a deep learning framework. Therefore, the rules of congestion propagation in the network can be actively learned from the observed data instead of predefining them based on prior knowledge. Experimental results on 971 testbeds in a regional road network in Beijing demonstrate that DBGCN outperforms the state-of-the-art models in inferring the congestion propagation spatiotemporal coverage and reveals variations in congestion propagation patterns according to the road network structure. Furthermore, the proposed model can simulate the congestion propagation process in customized scenarios by learning the latent congestion propagation rules. The results in different scenarios show that the change of congestion source location leads to distinct congestion magnitude, and the propagation of congestion will eventually stop at the road sections with strong shunting effect. [ABSTRACT FROM AUTHOR]

• Passengers' behavioural responses toward passenger flow control strategies. • NL model for departure time and mode shift decisions. • Mathematical programming with NL-based SUE constraints. • Trade-off between operational efficiency and service fairness. • Empirical case studies of Guangzhou and Beijing metros. This paper presents a methodology for developing demand-responsive passenger flow control strategies for oversaturated metro networks. A stated preference-off-revealed preference survey was conducted to capture passengers' behavioural responses to the passenger flow control strategies, and revealed significant behavioural change intentions in terms of departure times and mode choices. Currently, such behavioural responses are often ignored, leading to deviations in estimating the size of a target group and performance of a proposed strategy. To address this issue, in this study, a mathematical programming with equilibrium constraints (MPEC) approach was developed to optimise demand-responsive passenger flow control strategies. The proposed MPEC model highlights the importance of balancing the operational efficiency of a metro system with service fairness perceived by passengers. In particular, a nested logit-based stochastic user equilibrium problem was included to accommodate potential changes in demand patterns driven by candidate passenger flow control strategies. Two empirical cases based on the Guangzhou and Beijing metros were used to demonstrate the effectiveness of the proposed model and solution algorithm. The results show that the inclusion of service fairness considerations does not contradict the pursuit of efficiency. Instead, dual emphases on service fairness and passengers' behavioural responses help create a win–win situation for both metro operators and passengers. In the Guangzhou and Beijing metros, the highest section load rate decreased from 126.6% to 105.7% and from 122.98% to 106.87%, respectively, under the optimal passenger flow control strategies. The corresponding network load Gini coefficients improved from 0.278 to 0.259 and from 0.269 to 0.248, demonstrating the remarkable performance of this approach in regards to peak cutting and load balancing. [ABSTRACT FROM AUTHOR]

• We propose a novel approach to modelling spatial heterogeneity of COVID-19 spreading. • We focus on the effects of transportation networks on disease propagation heterogeneity. • The method combines the geographical weighted regression and the deep neural network. • The efficiency of several kinds of policy is analyzed in terms of pandemic containment. • The model has the potential to help contain the spread of COVID-19 or future epidemic. In this paper, we propose a novel approach to model spatial heterogeneity for epidemic spreading, which combines the relevance of transport proximity in human movement and the excellent estimation accuracy of deep neural network. We apply this model to investigate the effects of various transportation networks on the heterogeneous propagation of COVID-19 in China. We further apply it to predict the development of COVID-19 in China in two scenarios, i.e., i) assuming that different types of traffic restriction policies are conducted and ii) assuming that the epicenter of the COVID-19 outbreak is in Beijing, so as to illustrate the potential usage of the model in generating various policy insights to help the containment of the further spread of COVID-19. We find that the most effective way to prevent the coronavirus from spreading quickly and extensively is to control the routes linked to the epicenter at the beginning of the pandemic. But if the virus has been widely spread, setting restrictions on hub cities would be much more efficient than imposing the same travel ban across the whole country. We also show that a comprehensive consideration of the epicenter location is necessary for disease control. [ABSTRACT FROM AUTHOR]

• A method to solve train rescheduling and passenger reassignment simultaneously. • An integrated optimization model is formulated with a bi-objective function. • Our method can facilitate real-time rescheduling and re-ticketing efficiently. • Non-dominated train rescheduling and passenger reassignment solutions are obtained. During railway operations, unexpected events may influence normal traffic flows. This paper focuses on a train rescheduling problem for handling large disruptions, such as a rolling stock breakdown leading to a cancelled train service, where passenger reassignment strategies have to be considered. A novel mixed-integer linear programming formulation is established with consideration of train retiming, reordering, rerouting, and reservicing (addition of extra stops). The proposed mathematical formulation considers planning extra stops for non-canceled trains in order to transport the disrupted passengers, which were supposed to travel on the canceled train, to their pre-planned destination stations. Other constraints deal with limited seat capacity and track capacity, and mapping train rescheduling with passenger reassignment. A bi-objective function is optimized by a weighted-sum method to maximize the number of disrupted passengers reaching their destination stations and to minimize the weighted total train delay for all non-canceled trains at their destinations. A series of numerical experiments based on a part of the Beijing-Shanghai high-speed railway line is carried out to verify the effectiveness and efficiency of the proposed model and to perform a sensitivity analysis of various performance factors. The results show that an optimal reassignment plan of disrupted passengers is important to achieve real-time efficiency of traffic and re-ticketing. The impact of passenger reassignment on train rescheduling is influenced by the weights for objectives, duration of disruption, allowed additional dwell and running times, and relationship between passenger demand and total available train capacity. [ABSTRACT FROM AUTHOR]

This paper proposes a hierarchical Bayesian network model (BNM) to quantitatively evaluate the resilience of urban transportation systems. Based on systemic thinking and taking a sustainability perspective, we investigate the long-term resilience of the road transportation systems in four cities in China from 1998 to 2017, namely Beijing, Tianjin, Shanghai, and Chongqing, respectively. The model takes into account various factors collected from multi-source data platforms involved in stages of design, construction, operation, management, and innovation in road transportation systems. We test the model with the forward inference, sensitivity analysis, and backward inference. The result shows that the overall resilience scores of all four cities' transportation systems are within a moderate range with values between 49% to 59%. Although they all have an ever-increasing economic level, the levels of transportation resilience in Beijing and Tianjin decrease first and then gradually increase in a long run, which indicates a strong multi-dimensional, dynamic, and non-linear characteristic in resilience-economic coupling effect. Additionally, the results obtained from the sensitivity analysis and backward inference suggest that decision makers should pay more attention to the capabilities of quickly rebuilding and making changes to cope with future disturbances. As an exploratory study, this study clarifies the concepts of long-term multi-dimensional resilience and specific hazard-related resilience and provides an effective decision-support tool for stakeholders when building resilient infrastructure. • A novel BNM is proposed to evaluate multi-faceted transportation resilience. • Forward inference, sensitivity tests, and backward inference are conducted. • Long-term transportation resilience is highly dynamic in the tested cities. • The coupling effects of resilience with economic level is not always linear. • Abilities to rebuild and adapt through reconfiguration are of importance. [ABSTRACT FROM AUTHOR]