Search

Future traffic will be composed of both human-driven vehicles (HDVs) and automated vehicles (AVs). To accurately predict the performance of mixed traffic, an important aspect is describing HDV behavior when interacting with AVs. A few exploratory studies show that HDVs change their behavior when interacting with AVs, being influenced by factors such as recognizability and driving style of AVs. Unsignalized priority intersections can significantly affect traffic flow efficiency and safety of the road network. To understand HDV behavior in mixed traffic at unsignalized priority T-intersections, a driving simulator experiment was set up in which 95 drivers took part in it. The route in the driving simulator included three T-intersections where the drivers had to give priority to traffic on the major road. The participants drove different scenarios which varied in whether the AVs were recognizable or not, and in their driving style (Aggressive or Defensive). The results showed that in mixed traffic having recognizable aggressive AVs, drivers accepted significantly larger gaps (and had larger critical gaps) when merging in front of AVs as compared to mixed traffic having either recognizable defensive AVs or recognizable mixed AVs (composed of both aggressive and defensive). This was not the case when merging in front of an HDV in the same scenarios. Drivers had significantly smaller critical gaps when driving in traffic having non-recognizable aggressive AVs compared to non-recognizable defensive AVs. The findings suggest that human drivers change their gap acceptance behavior in mixed traffic depending on the combined effect of recognizability and driving style of AVs, including accepting shorter gaps in front of non-recognizable aggressive AVs and changing their original driving behavior. This could have implications for traffic efficiency and safety at such priority intersections. Decision makers must carefully consider such behavioral adaptations before implementing any policy changes related to AVs and the infrastructure.

Connected and automated vehicles (CAVs) are expected to enhance traffic efficiency by driving at shorter time headways, and traffic safety by shorter reaction times. However, one of the main concerns regarding their deployment is the mixed traffic situation, in which CAVs and manually driven vehicles (MVs) share the same road. This study investigates the behavioural adaptation of MV drivers in car-following and lane changing behaviour when they drive next to a dedicated lane (DL) for CAVs and compares that to a mixed traffic situation. The expectation is that in a mixed traffic situation, the behavioural adaptation of MV drivers is negligible due to lower exposure time and scarce platoons, while concentrating the CAVs on one dedicated lane may cause significant behavioural adaptation of MV drivers due to a higher exposure time and conspicuity of CAV platoons. Fifty-one participants were asked to drive an MV on a 3-lane motorway in three different traffic scenarios, in a fixed-base driving simulator: (1) Base, only MVs were present in traffic, (2) Mixed, platoons of 2–3 CAVs driving on any lane and mixed with MVs, (3) DL, platoons of 2–3 CAVs driving only on a DL. The DL was recognizable by road signs and a buffer demarcation which separated the DL from the other lanes. A moderate penetration rate of 43% was assumed for CAVs. During the drives, the car following headways and the accepted merging gaps by participants were collected and used for comparisons of driving behaviour in different scenarios. Based on the results, we conclude that there is no significant difference in the driving behaviour between Base and Mixed scenarios at tested penetration rate, confirming our research expectation. However, in DL scenario, MV drivers drove closer to their leaders specially when driving on the middle lane next to the platoons and accepted shorter gaps (up to 12.7% shorter at on-ramps) in lane changing manoeuvres. Dedicating a lane to CAVs increases the density of CAV platoons on one lane and consequently their conspicuity becomes higher. As a result, MV drivers are influenced by CAV platoons on a DL and imitate their behaviour. The literature suggests that dedicating a lane to CAVs improves the traffic efficiency by providing more possibilities for platooning. This study shows that implementing such a solution will affect the driving behaviour of human drivers. This should be taken into consideration when evaluating the impacts of dedicated lanes on traffic efficiency and traffic safety.

We present a transit model that, based on automated fare collection and train tracking data, describes pedestrian movements in train stations and vehicle-specific train ridership distributions. Our approach differs from existing models in that we describe on-board passenger dynamics and pedestrian dynamics at stations in a joint framework. We assume that travelers first decide on the train(s) they will take, and then pursue their journey through the network by successively choosing pedestrian paths, waiting positions on platforms, and specific train cars. Travelers explicitly maximize their travel utility. We model how crowding influences their walking speeds, and how it affects travel utility both at stations and on-board. To illustrate the framework, we present a real case study of a major Dutch rail corridor, for which we collect a rich set of passenger, pedestrian and train operation data. We observe a good agreement of model estimates with empirical observations, and discuss the use of the model for various transit-related problems including level-of-service assessment, crowding estimation, transit optimization, and integrated investment appraisal.

Mobility as a Service (MaaS) is expected to significantly change mobility patterns, yet it is still not clear who will embrace this new mobility paradigm and how MaaS will impact passengers’ transportation. In the paper, we identify factors relevant for MaaS adoption based on a survey comprised of over thousand respondents in the Netherlands. We find five clusters in relation to individuals’ inclinations to adopt MaaS in the context of urban mobility. We characterize each of the clusters, allowing for the examining of different customer segments regarding MaaS. The cluster with the highest inclination for future MaaS adoption is also the largest cluster (representing one third of respondents). Individuals in this cluster have multimodal weekly mobility patterns. On the contrary, current unimodal car users are the least likely to adopt MaaS. We identify high (mobility) ownership need and low technology adoption (present in three of the five clusters) as the main barriers that can hinder MaaS adoption. Policies that directly address these two barriers can stimulate MaaS adoption.

This paper develops a model-based hierarchical control method for coordinated ramp metering on freeway networks with multiple bottlenecks and on- and off-ramps. The controller consists of two levels where at the upper level, a Model Predictive Control (MPC) approach is developed to optimize total network travel time by manipulating total inflow from on-ramps to the freeway network. The lower level controller distributes the optimal total inflows to each on-ramp of the freeway based on local traffic state feedback. The control method is based on a parsimonious aggregated traffic model that relates the freeway total outflow to the number of vehicles on the freeway sections. Studies on aggregated traffic modeling of networks have shown the existence of a well-defined and low-scatter Macroscopic Fundamental Diagram (MFD) for urban networks. The MFD links network aggregated flow and density (accumulation). However, the MFD of freeway networks typically exhibits high scatter and hysteresis loops that challenge the control performance of MFD-based controllers for freeways. This paper addresses these challenges by modelling the effect of density heterogeneity along the freeway and capacity drop on characteristics of freeway MFD using field traffic data. In addition, we introduce a model to predict the evolution of density heterogeneity that is essential to reproduce the dynamics of freeway MFD accurately. The proposed model is integrated as the prediction model of the MPC in the hierarchical control method. The proposed coordinated ramp metering method shows desirable performance to reduce the vehicles total time spent and eliminate congestion. The control approach is compared with other coordinated ramp metering controllers based on the MPC framework with different traffic prediction models (e.g. CTM and METANET). The outcomes of numerical experiments highlight that the MFD-based hierarchical controller (i) is better able to overcome the modeling mismatch between the prediction model and the plant (process model) in the MPC framework and (ii) requires less computation effort than other nonlinear controllers.

Vessel traffic is a key element determining port safety and capacity. The growth of port calls and cargo can have implications in port operations. Decision makers need to take decisions to anticipate any future capacity drop or increase in nautical risks. In this research, a multi-criteria decision making methodology is developed to evaluate the trade-off between safety and capacity of vessel traffic in ports, as well as other assessment indicators. The methodology first identifies the most relevant risk and capacity assessment criteria and the input required. A simulation model is used to calculate the criteria for a variety of different scenarios. The outcome results for the criteria from the simulations are used into a decision-making method that includes unknown decision makers’ preferences. The use of this methodology provides the best scenario for port vessel traffic, when strategic and operational decisions should be taken, and it can be used as a framework for port assessment by decision makers for future changes in traffic management strategies or changes in the port infrastructure, such as port expansions.

The importance of maritime transport keeps increasing with the trade globalization. With the growing demand for waterborne transport, vessel traffic flows are also expected to increase. This paper reviews maritime traffic models from the vessel behavior modeling perspective. The maritime traffic models include the models for vessel traffic both at sea and in confined water area. The aim of this paper is to analyze the underlying modeling paradigms and to assess the extent in which maritime traffic models can represent vessel behavior. Focusing on vessel behavior modeling, this paper provides a broad overview of the current literature on maritime traffic models of the last decades. The commercial models are not included due to the limit of information. To compare the capabilities of models in capturing the vessel behavior characteristics, the considered models are assessed from different aspects of vessel behavior representation, external impact modeling, and model applicability. The assessment shows that none of the existing models describe all dynamic kinetic information in detail for different vessels and consider the impacts from a full range of external factors, which is possibly due to the specific purpose when the models were developed. The models developed for specific vessels in specific situations ignore the irrespective behavioral details in other possible scenarios. Models without proper calibration and validation limit the applicability in other cases. It also indicates that few models can accurately simulate the different vessel behavior at a microscopic level. To investigate the possible potential and limitations, the models have been assessed and discussed to indicate the underlying modeling paradigms based on the modeling characteristics. Future developments can focus on the behavior of different vessels in different types of water areas and the corresponding impacts from external conditions (e.g. visibility, wind, current), vessel encounters and traffic rules. Through calibration and validation, future models should be able to fit the vessel behavior in real-life situations.

Intelligent vehicle technologies are opening new possibilities for decentralized vehicle routing systems, suitable for regulating large traffic networks, and at the same time, capable of providing customized advice to individual vehicles. In this study, we perform a rigorous simulation-based analysis of an in-vehicle routing strategy that aims to achieve a user-equilibrium distribution in traffic. Novel features of the approach include: a mechanism based on forward propagation of individual vehicle decisions to anticipate future traffic dynamics; time-dependent prediction of route travel times with neural network-based link predictors; and a stochastic routing policy for in-vehicle decision-making based on predicted travel times. However, for an effective application of the approach, design choices need to be made regarding the accuracy of the link predictors, and some control settings. These choices may depend on the network size and structure. We investigate the impact of two important design aspects: sequentially using link-level predictors for route travel time estimation, and the control parameter values, on the equilibrium performance at the network-level. The results suggest functional scalability of the approach, in terms of the prediction model accuracy and routing performance. Overall, the work contributes to a qualitative and quantitative understanding of emergent performance from the given routing approach.

This paper presents a car-following control strategy of connected automated vehicles (CAVs) to stabilize a mixed vehicular platoon consisting of CAVs and human-driven vehicles. This study first establishes a string stability criterion for a mixed vehicular platoon. Specifically, a mixed vehicular platoon is decomposed into “subsystems” that are all possible sequential subsets of the platoon. String stability is then defined as the “head-to-tail” string stability for all subsystems: the magnitude of a disturbance is not amplified from the first vehicle to the last vehicle of each subsystem. Based on this definition, distributed frequency-domain-based CAV control is proposed to increase the number of head-to-tail string stable subsystems and consequently dampen stop-and-go disturbances drastically. Specifically, an H-infinity control problem is formulated, where the maximum disturbance “damping ratios” in each subsystem is minimized within the predominant acceleration frequency boundaries of human-driven vehicles. Simulation experiments, embedded with real human-driven vehicle trajectories, were conducted, and results show that the proposed control can effectively dampen stop-and-go disturbances.

Minimizing passenger transfer times through public transport (PT) transfer synchronization is important during tactical planning and real-time control. However, there are computational challenges for solving this Timetable Synchronization Problem (TSP) for large, real-world urban PT networks. Hence, in this study we propose a data-driven, passenger-oriented methodology as a preparatory selection stage to reduce problem dimensionality by (1) determining the significant transfer hubs in the network, and (2) identifying subsets of lines within these hubs that need to be prioritized for transfer synchronization. In the first phase of our methodology we determine the spatial boundaries of transfer locations, using a clustering technique based on the passenger transfer flow matrix inferred from smartcard data. After that, a subset of hubs to be prioritized for synchronization is selected. In the second phase, we characterize the transfer patterns within the hubs based on a topological representation. Based on these topological graphs, the line bundles that need to be prioritized within the hubs are further identified using a modularity-based community detection technique. We apply our methodology to a real-world case study, i.e. the PT network of The Hague, the Netherlands. For this case study, our approach allows for prioritizing 70% of all transfers within identified transfer locations while only requiring 0.9% of these transfer locations, thus reducing the complexity of solving the TSP substantially at a relatively low cost. Our method supports public transport operators during timetable design and real-time control in determining where and which lines to prioritize when devising measures for improving transfer experience and synchronization.

Virtual Reality (VR) provides the possibility to study pedestrian wayfinding behaviour in multi-level buildings. Although VR has been applied increasingly to study pedestrian behaviour, it has remained unclear how different VR technology would affect behavioural outcomes in a multi-level building. The study compares the adoption of different VR technologies for pedestrian wayfinding studies, via investigating the difference in pedestrian wayfinding behaviour and user experience. Wayfinding experiments with two groups of participants were conducted using either HMD VR or Desktop VR. Pedestrian movement trajectory data were collected via the VR system and user experience data were recorded via a questionnaire. These data allow for direct comparison and detailed analysis of pedestrian behaviour and user experience between the adoption of two VR technologies. The results showed that technological differences have a significant impact on wayfinding task performance and head rotation change. However, the route choice, exit choice and user experience were overall similar between the two groups. These results provide empirical evidence supporting researchers to choose between immersive and non-immersive VR when study pedestrian wayfinding behaviour.

Specifying the choice set for travel behaviour analysis is a non-trivial task. Its size and composition are known to influence the results of model estimation and prediction. Most studies specify the choice set using choice set generation algorithms. These methods can introduce two types of errors to the specified choice set: false negative (not generating observed routes) and false positive (including irrelevant routes). Due to increased availability of revealed preference data, like GPS, it is now possible to identify the choice set using a data-driven approach. The data-driven path identification approach (DDPI) combines all unique routes that are observed for one origin-destination pair into a choice set. This paper evaluates this DDPI approach by comparing it to two commonly used choice set generation methods (breadth-first search on link elimination and labelling). The evaluation considers the three main purposes of choice sets: analysis of alternatives in the choice set, model estimation and prediction. The conclusion is that the DDPI approach is a useful addition to the current choice set identification methods. The findings indicate that in analysing alternatives in the choice set, the DDPI approach is most suitable, as it reflects the observed behaviour. For model estimation the DDPI approach provides a useful addition to the current choice set generation methods, as it provides insights into the preferences of individuals without requiring network-data for additional information or generating routes. In terms of prediction, the DDPI approach is not suitable, as it is not able to perform well with out-of-sample data.

This paper validates the prediction model embedded in a model predictive controller (MPC) of variable speed limits (VSLs). The MPC controller was designed based on an extended discrete first-order model with a triangular fundamental diagram. In our previous work, the extended discrete first-order model was designed to reproduce the capacity drop and the propagation of jam waves, and it was validated with reasonable accuracy without the presence of VSLs. As VSLs influence traffic dynamics, the dynamics including VSLs needs to be validated, before it can be applied as a prediction model in MPC. For conceptual illustrations, we use two synthetic examples to show how the model reproduces the key mechanisms of VSLs that are applied by existing VSL control approaches. Furthermore, the model is calibrated by use of real traffic data from Dutch freeway A12, where the field test of a speed limit control algorithm (SPECIALIST) was conducted. In the calibration, the original model is extended by using a quadrangular fundamental diagram which keeps the linear feature of the model and represents traffic states at the under-critical branch more accurately. The resulting model is validated using various traffic data sets. The accuracy of the model is compared with a second-order traffic flow model. The performance of two models is comparable: both models reproduce accurate results matching with real data. Flow errors of the calibration and validation are around 10%. The extended discrete first-order model-based MPC controller has been demonstrated to resolve freeway jam waves efficiently by synthetic cases. It has a higher computation speed comparing to the second-order model-based MPC.

Circuity of transit networks, defined as the ratio of network to Euclidean distance traveled from origin to destination stop, has been known to influence travel behavior. In addition to the longer time spent in travel, for networks where fare is based on distance traveled, higher circuity also means higher fare for the same Euclidean distance. This makes circuity relevant from an equity perspective. Using a case study of the urban transit network of Amsterdam in the Netherlands, this study explores the role of transit circuity on the disparity in distance traveled by travelers' income profile and its implications on travel times and costs for networks with distance-based fares. The analysis is based on travel patterns from smart card data for bus, tram, and metro modes, combined with neighborhood level income data. Results reveal that in Amsterdam, the higher the share of high income people living in proximity to a transit stop, the lower the circuity of journeys from the stop, when controlled for the Euclidean distance covered and spatial auto-correlation. The uneven distribution of circuity exacerbates the disparity in distance traveled, and hence fare paid between the income groups. However, the travel time per Euclidean distance favors the low income group, possibly due to the circuitous routes serving these areas being compensated by higher travel speeds. This study highlights the role of transit network design in determining its equity outcomes and emphasizes the importance of considering equity during route and fare planning. The process followed can be adapted to examine equity for other urban networks.

In recent years numerous pedestrian simulation tools have been developed that can support crowd managers and government officials in their tasks. New technologies to monitor pedestrian flows are in dire need of models that allow for rapid state-estimation. Many contemporary pedestrian simulation tools model the movements of pedestrians at a microscopic level, which does not provide an exact solution. Macroscopic models capture the fundamental characteristics of the traffic state at a more aggregate level, and generally have a closed form solution which is necessary for rapid state estimation for traffic management purposes. This contribution presents a next step in the calibration and validation of the macroscopic continuum model detailed in Hoogendoorn et al. (2014). The influence of global and local route choice on the development of crowd movement phenomena, such as dissipation, lane-formation and stripe-formation, is studied. This study shows that most self-organization phenomena and behavioural trends only develop under very specific conditions, and as such can only be simulated using specific parameter sets. Moreover, all crowd movement phenomena can be reproduced by means of the continuum model using one parameter set. This study concludes that the incorporation of local route choice behaviour and the balancing of the aptitude of pedestrians with respect to their own class and other classes are both essential in the correct prediction of crowd movement dynamics.

We propose in this paper a decentralized traffic signal control policy for urban road networks. Our policy is an adaptation of a so-called BackPressure scheme which has been widely recognized in data network as an optimal throughput control policy. We have formally proved that our proposed BackPressure scheme, with fixed cycle time and cyclic phases, stabilizes the network for any feasible traffic demands. Simulation has been conducted to compare our BackPressure policy against other existing distributed control policies in various traffic and network scenarios. Numerical results suggest that the proposed policy can surpass other policies both in terms of network throughput and congestion., Public technical report

This contribution puts forward a receding horizon control approach for automated driving systems, where tactical-level lane change decisions and control-level accelerations are jointly evaluated under a central mathematical framework. The key idea is that controlled vehicles predictively determine discrete desired lane sequences and continuous accelerations to minimise a cost function reflecting undesirable future situations. The interactions between controlled vehicles and surrounding vehicles are captured in the cost function. The approach is flexible in terms of application to controller design for both non-cooperative control systems where controlled vehicles only optimise their own cost and cooperative control systems where controlled vehicles coordinate their decisions to optimise the collective cost. To determine the controller behaviour, the problem is formulated as a differential game where controlled vehicles make decisions based on the expected behaviour of other vehicles. The control decisions are updated at regular frequency, using the newest information regarding the state of controlled vehicles and surrounding vehicles available. A problem decomposition technique is employed to reduce the dimensionality of the original problem by introducing a finite number of sub-problems and an iterative algorithm based on Pontryagin’s Principle is used to solve sub-problems efficiently. The proposed controller performance is demonstrated via numerical examples. The results show that the proposed approach can produce efficient lane-changing manoeuvres while obeying safety and comfort requirements. Particularly, the approach generates optimal lane change decisions in the predicted future, including strategic overtaking, cooperative merging and selecting a safe gap.

Congestion pricing schemes have been traditionally derived based on analytical representations of travel demand and traffic flows, such as in bottleneck models. A major limitation of these models, especially when applied to urban networks, is the inconsistency with traffic dynamics and related phenomena such as hysteresis and the capacity drop. In this study we propose a new method to derive time-varying tolling schemes using the concept of the Network Fundamental Diagram (NFD). The adopted method is based on marginal cost pricing, while it also enables to account realistically for the dynamics of large and heterogeneous traffic networks. We derive two alternative cordon tolls using network-aggregated traffic flow conditions: a step toll that neglects the spatial distribution of traffic by simply associating the marginal costs of any decrease in production within the NFD to the surplus of traffic; and a step toll that explicitly accounts for how network performance is also influenced by the spatial variance in a 3D-NFD. This pricing framework is implemented in the agent-based simulation model MATSim and applied to a case study of the city of Zurich. The tolling schemes are compared with a uniform toll, and they highlight how the inhomogeneous distribution of traffic may compromise the effectiveness of cordon tolls.

Within the realm of pedestrian research numerous measures have been proposed to estimate the level of crowdedness experienced by pedestrians. However, within the field of pedestrian traffic flow modelling there does not seem to be consensus on the question which of these measures performs best. This paper shows that the shape and scatter within the resulting fundamental diagrams differs a lot depending on the measure of crowdedness used. The main aim of the paper is to establish the advantages and disadvantages of the currently existing measures to quantify crowdedness in order to evaluate which measures provide both accurate and consistent results. The assessment is not only based on the theoretical differences, but also on the qualitative and quantitative differences between the resulting fundamental diagrams computed using the crowdedness measures on one and the same data set. The qualitative and quantitative functioning of the classical Grid-based measure is compared to with the X-T measure, an Exponentially Weighted Distance measure, and a Voronoi-Diagram measure. The consistency of relating these measures for crowdedness to the two macroscopic flow variables velocity and flow, the computational efficiency and the amount of scatter present within the fundamental diagrams produced by the implementation of the different measures are reviewed. It is found that the Voronoi-Diagram and X-T measure are the most efficient and consistent measures for crowdedness.

This contribution puts forward a novel multi-class continuum model that captures some of the key dynamic features of pedestrian flows. It considers route choice behaviour on both the strategic (pre-trip) and tactical (en-route) level. To achieve this, we put forward a class-specific equilibrium direction relation of the pedestrians, which is governed by two parts: one part describing the global route choice, which is pre-determined based on the expectations of the pedestrians, and one part describing the local route choice, which is a density-gradient dependent term that reflects local adaptations based on prevailing flow conditions. Including the local route choice term in the multi-class model causes first of all dispersion of the flow: pedestrians will move away from high density areas in order to reduce their overall walking costs. Second of all, for the crossing flow and bi-directional flow cases, local route choice causes well known self-organised patterns to emerge (i.e. diagonal stripes and bi-directional lanes). We study under which demand conditions self-organisation occurs and fails, as well as what the impact is of the choices of the different model parameters. In particular, the differences in the weights reflecting the impact of the own and the other classes appear to have a very strong impact on the self-organisation process.

Our understanding of crowd movements has rapidly improved over the course of the last decades. This study shows that the empirical research has not kept up with the pace of the simulation studies. Based on an extensive literature review, this study proposes a conceptual model describing the movements of individuals within a crowd. The model features the relationships between the macroscopic flow variables and characteristics of the pedestrians, their physiologic environment and the surrounding infrastructure in which they reside. Using trajectory data sets gathered during large-scale pedestrian events in the Netherlands, the conceptual model has been statistically tested. The results show that the proposed framework is capable of explaining trends featured by the macroscopic flow variables based on the before mentioned characteristics.

In this contribution a new model to include stochastic vehicle specific behaviour and interaction in traffic flow modelling is presented. The First Order Model with Stochastic Advection (FOMSA) is presented as a first order macroscopic kinematic wave model in a platoon-based Lagrangian coordinate system. The use of Lagrangian coordinates allows characteristics of specific vehicles or vehicle-groups to propagate along with the traffic flow using a vehicle specific invariant. The invariant reflects how vehicle or platoon specific characteristics propagate with the vehicles and influence the local behaviour of a vehicle or platoon on a macroscopic level and in interaction with other surrounding vehicles. It represents a local vehicle specific adjustment to the critical density and makes use of two parameters: a stochastic boundary parameter and a transition parameter. These parameters indicate the extent of differences between vehicles or platoons. A case study is also presented in which a demonstration of the model is given and the face validity and sensitivity of the parameters are shown. Previously, similar approaches have made use of second order model descriptions. The formulation of this model as a first order model makes use of the advantages of first order models and also applies the improved accuracy of Lagrangian coordinates over the Eulerian coordinate system in time-stepping.

The dynamics of pedestrian flows can be captured in a continuum modelling framework. However, compared to vehicular flow, this is a much more challenging task. In particular the integration of flow propagation and path choice are known to be problematic. Furthermore, pedestrian flow is characterised by different self-organised phenomena, such as the formation of dynamic lanes and diagonal stripes, which have not yet been captured in a continuum modelling framework. This contribution puts forward a novel multi-class continuum model that captures some of the key features of pedestrian flows. It considers path choice behaviour on both the strategic (pre-trip) and tactical (en-route) level. To achieve this, we present a methodology to derive a continuum model from a microscopic walker model, in this case the social forces model. In doing so, we show that the interaction term present in the social forces model introduces a local path choice component in the equilibrium velocity. Having derived the model, we analyse its properties both by means of mathematical analyses and simulation studies. This reveals the general behaviour of the model, as well as the ability of the model to reproduce self-organised structures, and phase transitions. To the best of our knowledge, this is the first continuum model that is able to reproduce these self-organised structures.

In probabilistic traffic models, consideration of stochasticity in the dynamics of traffic gives a closer representation of a traffic system in comparison to that of a deterministic approach. Monte Carlo simulation is a broadly accepted method to consider variations in traffic within modelling. In this contribution, the possibility of increasing the efficiency of probabilistic traffic flow models using Monte Carlo simulation is analysed using variance reduction techniques and sequencing, for varied capacity and traffic demand values. The techniques of Importance Sampling, Latin Hypercube Sampling and Quasi-Random Sequencing are compared in a dynamic macroscopic traffic model to demonstrate the effectiveness of these techniques for reduction of the computational load when considering multiple input variations. Demonstration of their efficiency in traffic modelling is expected to lead to a wider application of the methods in practice.

This contribution furthers the control framework for driver assistance systems in Part I to cooperative systems, where equipped vehicles can exchange relevant information via vehicle-to-vehicle communication to improve the awareness of the ambient situation (cooperative sensing) and to manoeuvre together under a common goal (cooperative control). To operationalize the cooperative sensing strategy, the framework is applied to the development of a multi-anticipative controller, where an equipped vehicle uses information from its direct predecessor to predict the behaviour of its pre-predecessor. To operationalize the cooperative control strategy, we design cooperative controllers for sequential equipped vehicles in a platoon, where they collaborate to optimise a joint objective. The cooperative control strategy is not restricted to cooperation between equipped vehicles. When followed by a human-driven vehicle, equipped vehicles can still exhibit cooperative behaviour by predicting the behaviour of the human-driven follower, even if the prediction is not perfect. The performance of the proposed controllers are assessed by simulating a platoon of 11 vehicles with reference to the non-cooperative controller proposed in Part I. Evaluations show that the multi-anticipative controller generates smoother behaviour in accelerating phase. By a careful choice of the running cost specification, cooperative controllers lead to smoother decelerating behaviour and more responsive and agile accelerating behaviour compared to the non-cooperative controller. The dynamic characteristics of the proposed controllers provide new insights into the potential impact of cooperative systems on traffic flow operations, particularly at the congestion head and tail.

There are few studies investigating crowd dynamics in panic situations. They used measures such as exit flow rate to explore the exit performance in evacuation scenarios. However, there is limited research exploring the relationship of exit flow rate and density behind the exit for panic scenarios. This study presents a macro level analysis to investigate the relationship of exit flow and exit jam characteristics. Animal group behavior (i.e. panicked woodlice experiments) is utilized for data analysis. The results reveal that change in woodlice escaping behavior cause an increasing trend in exit capacity as the jam behind the exit increases.

Whether pedestrian movements do or do not follow similar patterns as vehicular traffic while experiencing congestion is not entirely understood. Using data gathered during bottleneck experiments under laboratory conditions, the phenomenon of anticipation before entering congestion is studied. This paper provides proof that the movement behavior of pedestrians upstream and downstream of bottleneck severely differs severely. As such, this paper concludes that a fundamental diagram of pedestrian movement dynamics can only be provided for each separate situation.

This paper introduces a topic which is of growing interest in the traffic control and management community, namely the relation between traffic data quality and the efficiency of traffic management. After some explanatory background and state of the art, it will become clear that no standard approach is available for traffic engineers to determine which accuracy of traffic data is needed for a certain traffic management application. In this paper we will show how the effect of inaccurate measurements on network level can be quantified by determining information utility functions. This approach is illustrated by presenting the results of a simulation study into the effect of inaccurate loop detector measurements on a ramp metering system. The relation between different accuracies of flow and speed measurements as input for a ramp metering system and the corresponding effect on the network performance is studied with a micro simulation model. The results of the simulation are used in a cost-benefit analysis for a case comparing loop detectors with camera measurements, where loop detectors are considered more accurate but also more expensive. We will show that higher measurement accuracy will lead to a better performance of the ramp metering system, resulting in a higher average network speed, and also quantify this relationship. © IFAC.

This paper presents details of the developments of the Nomad model after being introduced more than 12 years ago. The model is derived from a normative theory of pedestrian behavior making it unique under microscopic models. Nomad has been successfully applied in several cases indicating that it fulfils requirements for accuracy, scope and computational efficiency. In this paper we introduce the components of the model by relating them with behavioral assumptions providing an explanatory character to Nomad. The link between the model and the pedestrian theory is the main contribution and could improve the development of pedestrian models in general.

Pedestrian dynamics has been being a popular research area. In this paper, breakdown probability in the bidirectional flow that includes pedestrians with heterogeneities is investigated. Reaction time appears to have the most pronounced breakdown proba- bility, followed by desired velocity and then body size. The relationship between the start time of breakdown and inflow reflects the decreasing trend of the start time of breakdown with increasing inflow. Furthermore, the smaller the value of the start time of breakdown for the same inflow, the larger the effect of the heterogeneity in a parameter of the social force model on breakdown by cross-comparisons.

Currently, pedestrian simulation models are used to predict where, when and why hazardous high density crowd movements arise. However, it is questionable whether models developed for low density situations can be used to simulate high density crowd movements. The objective of this paper is to assess the existent pedestrian simulation models with respect to known crowd phenomena in order to ascertain whether these models can indeed be used for the simulation of high density crowds and to indicate any gaps in the field of pedestrian simulation modeling research. This paper provides a broad, but not exhaustive overview of the crowd motion simulation models of the last decades. It is argued that any model used for crowd simulation should be able to simulate most of the phenomena indicated in this paper. In the paper cellular automata, social force models, velocity-based models, continuum models, hybrid models, behavioral models and network models are discussed. The comparison shows that the models can roughly be divided into slow but highly precise microscopic modeling attempts and very fast but behaviorally questionable macroscopic modeling attempts. Both sets of models have their use, which is highly dependent on the application the model has originally been developed for. Yet, for practical applications, that need both precision and speed, the current pedestrian simulation models are inadequate.

In this article we demonstrate how network components can be modeled using the kinematic wave model in the Lagrangian formulation. This includes modeling nodes (or discontinuities) such as inflow and outflow boundaries, merges and bifurcations (e.g. ramps) and nonhomogeneous roads. Nodes are usually fixed in space. This makes their implementation in Lagrangian coordinates where the coordinates move with the vehicle more complex than in Eulerian coordinates where the coordinates are fixed in space. To this end we derive an analytical node model. The article then discusses how to implement such sink and source terms in a discretized version of the kinematic wave model in Lagrangian coordinates. In this implementation several choices have to be made. Test results show that even with the most simple choices (discretization based on full vehicle groups and discrete time steps) accurate and plausible results are obtained. We conclude that the Lagrangian formulation can successfully be applied for simulation of networks of nonhomogeneous roads.

Driver assistance systems support drivers in operating vehicles in a safe, comfortable and efficient way, and thus may induce changes in traffic flow characteristics. This paper puts forward a receding horizon control framework to model driver assistance and cooperative systems. The accelerations of automated vehicles are controlled to optimise a cost function, assuming other vehicles driving at stationary conditions over a prediction horizon. The flexibility of the framework is demonstrated with controller design of Adaptive Cruise Control (ACC) and Cooperative ACC (C-ACC) systems. The proposed ACC and C-ACC model characteristics are investigated analytically, with focus on equilibrium solutions and stability properties. The proposed ACC model produces plausible human car-following behaviour and is unconditionally locally stable. By careful tuning of parameters, the ACC model generates similar stability characteristics as human driver models. The proposed C-ACC model results in convective downstream and absolute string instability, but not convective upstream string instability observed in human-driven traffic and in the ACC model. The control framework and analytical results provide insights into the influences of ACC and C-ACC systems on traffic flow operations., Comment: Submitted to Transportation Research Part C: Emerging Technologies

Advancements in the efficiency, quality and manufacturability of sensing and communication systems are driving the field of intelligent transport systems (ITS) into the twenty first century. One key aspect of ITS is the need for efficient and robust integrated network management of urban traffic networks. This paper presents a general model predictive control framework for both centralized traffic signal and route guidance systems aiming to minimize network congestion. Our novel model explicitly captures both non-zero travel time and spill-back constraints while remaining linear and thus generally tractable with quadratic costs. The end result is a central control scheme that may be realized for large urban networks containing thousands of sensors and actuators.We demonstrate the essences of our model and controller through a detailed mathematical description coupled with simulation results of specific scenarios. We show that using a central scheme such as ours may reduce the congestion inside the network by up to half while still achieving better throughput compared to that of other conventional control schemes.

The use of Prospect Theory to model route choice has increased in the past decades. The main application issue is how to define the reference point, i.e., the value that travellers use as a reference to distinguish gains and losses in the experienced travel times. Moreover, the question can be asked whether all travellers have the same reference point or whether heterogeneity in their behaviour plays an important role. This paper aims to (i) provide a behavioural interpretation of the reference point, (ii) investigate the role of heterogeneity in the reference point and (iii) discuss how to take heterogeneity into account. These aspects are discussed with the aid of an empirical route choice experiment and a model specification in which travel time is the main variable. Two model frameworks are proposed, one accounting for heterogeneity and another considering no heterogeneity in travellers' behaviour, and their outcomes are compared. Results show improvements in the ability of Prospect Theory to predict route choice behaviour by accounting for heterogeneity in the reference point. This is particularly the case when the reference point reflects travelers' route preferences. Statistical analyses show the significance of accounting for heterogeneity in travellers' behaviour. Thus, we cannot reject the hypothesis that heterogeneity leads to improvements in the prediction ability of Prospect Theory.

It is computationally expensive to find out where vulnerable parts in a network are. In literature a variety of methods were introduced that use simple indicators (measured in real-life or calculated in a traffic simulator) to pre-determine the seriousness of the delays caused by the blocking of that link and thereafter perform a more detailed analysis. This article reviews the indicators proposed in the literature and assesses the quality of these indicators. Furthermore, a multi-linear fit of the indicators is made to find a better, combined, indicator to rank the links according to their vulnerability. The article shows that different indicators assess different links to be vulnerable. Also combined they cannot predict the vulnerability of a link. Therefore, it is concluded that to find vulnerable links, one has to look further than link-based indicators. © 2012 Elsevier Ltd.

The aim of this paper is to gain insights into the level of heterogeneity in car-following behavior in real traffic. We use a large sample of trajectory observations collected by means of a helicopter to identify differences between the car-following behaviors of: (1) passenger car drivers, (2) passenger car drivers and truck drivers and (3) passenger car drivers following a passenger car and passenger car drivers following a truck. We thereto calibrate eight car-following models making different assumptions about the way in which drivers follow their leader(s) on the same lane. We show that considerable behavioral differences exist between passenger car drivers. Different passenger car drivers do not only consider different stimuli (like speed difference(s) with the leading car(s) and distance headway(s) to leading car(s)) but also the extents to which these stimuli influence their behavior differ. Truck drivers turn furthermore out to adopt in general a more robust car-following behavior than passenger car drivers. Their speeds show, for example, less variation over time. We also find indications that the desired headways of passenger car drivers are lower when following a truck than when following a passenger car.

In this paper, we show how the capacity of evacuation doors is affected by the evacuation door width, population composition, the presence of an open door and evacuation conditions. For this, laboratory experiments have been performed. Varying door opening widths showed that only the experiment with the widest door opening (275 cm) resulted in a capacity lower than the threshold capacity from the design guidelines (2.25 P/m/s). The average observed capacities are for all widths lowest for the lowest stress level and highest for the highest stress level. The population with a greater part of children has the highest capacity, while the lowest capacity is, as expected, found for the experiment with 5% disabled participants. The presence of a door opened in the escape direction in an angle of 90 degrees for a door opening of 85 cm results in a 20% capacity reduction.

This paper investigates to what extent travellers change their route when faced with unexpected traffic situation. To this end, traffic data from days with serious incidents are analysed in this contribution. The flows retrieved from loop detectors on the routes past the incident and on alternative routes are compared with the same values on days without an incident. It is found that for major accidents up to 50% of the travellers deviate from their normal route if the traffic situation is different. Furthermore, more travellers take an alternative route if the delay on the original route is caused by an accident than if they are faced with the same delay on the original route without an incident. These findings are for instance important for providing route information or suggestions on alternative routes or for finding vulnerable links.

This paper presents a continuum approach to model the dynamics of cooperative traffic flow. The cooperation is defined in our model in a way that the equipped vehicle can issue and receive a warning massage when there is downstream congestion. Upon receiving the warning massage, the (up-stream) equipped vehicle will adapt the current desired speed to the speed at the congested area in order to avoid sharp deceleration when approaching the congestion. To model the dynamics of such cooperative systems, a multi-class gas-kinetic theory is extended to capture the adaptation of the desired speed of the equipped vehicle to the speed at the downstream congested traffic. Numerical simulations are carried out to show the influence of the penetration rate of the equipped vehicles on traffic flow stability and capacity in a freeway.

The steadily increasing numbers and lengths of traffic jams on freeways have led to the application of Dynamic Traffic Management (DTM) measures all over the world. Ramp metering control has proven to be one of the most efficient means to reduce freeway congestion. Currently, it is expected that integrated and coordinated application of DTM measures will further improve its impact. This paper studies a coordinated ramp metering control algorithm called HERO/RWS. This algorithm has been developed for the current Dutch ramp metering systems and it will be applied on the Amsterdam A10 freeway network in the near future. The aim of this algorithm is to postpone congestion on freeways by effectively using ramp storage space from upstream on-ramps. VISSIM-based microscopic simulation results show that the HERO/RWS coordinated control outperforms non-coordinated ramp metering control. Parameter settings have been optimized for the specific A10-west network through a robustness study. In addition, the concept of coordination between ramp meter and upstream intersection traffic controllers is developed. The feasibility of this idea has been proven by a simulation study.

Roads in the Netherlands are often heavily congested. Real-time travel time information can be a valuable instrument to reduce the impact of increasing traffic demand on travel time with advantages for traffic participants as well as for the traffic network managers. For urban roads travel time estimation is a more complex problem than for freeways. In order to provide online, real-time and accurate travel times, in this paper a new estimation approach is developed based on measurements supplied by the (vehicle-dependent) traffic controllers. These measurements include the occupancy from the corresponding long loop, the flow from the short loop, and the green percentages. Based on these measurements the location and length of the queues is estimated. Moreover, the approach transforms the data in a way that the underlying reason(s) of congestion becomes apparent (in terms of bottleneck location), which can be very useful for, e.g., advanced traffic information systems or decision support systems. Compared with the approach based on license plate recognition cameras, which is used in the Netherlands in many locations, this approach is expected to be more accurate in heavily congested situations, since for license plate recognition vehicles have to pass the end of the route before the traveled time can be measured. The approach is tested with real traffic data for an urban route in the Netherlands. The test case presented in the paper shows that the use of the additional data from traffic controllers (occupancy, green percentage, waiting time of first vehicle in queue) is beneficial.

Marginal costs in traffic networks are the extra costs incurred to the system as the result of extra traffic. Marginal costs are required frequently e.g. when considering system optimal traffic assignment or tolling problems. When explicitly considering spillback in a traffic flow model, one can use a numerical derivative or resort to heuristics to calculate the marginal costs. Numerical derivatives are computationally demanding, restricting its use to simple networks. Heuristic approaches in most cases approximate the marginal costs by only considering the extra costs on the links which are traveled by the extra traffic, excluding the possibly external costs incurred on other links due to spillback. This paper proposes a novel way to estimate the true marginal costs of traffic in a dynamic discrete LWR model which correctly deals with congestion onset, spillback and dissolution. The proposed methodology tracks virtual changes in density through the network by means of particles which travel along with the characteristics of traffic. By using density based cost functions, the virtual changes in density can be directly related to the marginal costs. The computational efficiency of the methodology stems from the fact that only local conditions are considered when propagating the virtual change in density. The paper discusses the methodology and necessary model extensions, provides a numerical validation experiment illustrating the exact detail of the solution by comparison to a numerical derivative and discusses some generalizations.

Recent research has revealed that there exist large inter-driver differences in car-following behavior such that different car-following models may apply to different drivers. This study applies Bayesian techniques to the calibration of car-following models, where prior distributions on each model parameter are converted to posterior distributions. The priors and posteriors are then used to calculate the so-called ‘evidence’, which can be used to quantitatively assess how well different models explain one driver's car-following behavior. When considered over multiple drivers, the evidence represents probabilities for different models as a whole. These model probabilities can be used in a micro simulation, where for each driver first a model is drawn according to these probabilities, after which parameters are drawn from the posterior distribution for each parameter of that model that were obtained when calibrating the model. In a test case on actual data the Bayesian evidence indeed reveals inter-driver differences and it is shown how these differences can quantitatively be assessed.

This paper presents the results of two calibration efforts and improvements of the Nomad microscopic walker model. Each calibration consisted in comparing the outcome of 19 sets of model parameters with results from laboratory experiments. Three different flows were used in the calibrations: bidirectional, unidirectional and a narrow bottleneck. For the first two types of flow the macroscopic speed-density relation is used for comparison. For the narrow-bottleneck the capacity for the bottleneck is assessed. Additionally, quantitative measures of self-organizing lane formation for the bidirectional flow are included in the calibration. In the first calibration effort only 4 parameter sets did not present gridlocks in the bidirectional flow and their results are presented. These four sets over estimated the capacity of the bottleneck and the efficiency of the unidirectional flows. These results are discussed and it is shown that a modification in the model is necessary. A modified model is presented and submitted to a new round of calibration. One parameter set showed significant improvement for the speed-density relations and the capacity of the bottleneck. The experience gained in this calibration effort indicates that pedestrian microscopic models can be calibrated over several types of flows simultaneously. However, this research shows that care must be taken when using only macroscopic flow properties for calibration. They may not guarantee the generality of the model due to the complexity and variety of possible flows.

A method is described to develop a set of optimal instructions to evacuate by car the population of a region threatened by a hazard. By giving these instructions to the evacuees, traffic conditions and therefore the evacuation efficiency can be optimized. The instructions, containing a departure time, a destination, and a route, are created using an optimization method based on ant colony optimization. Iteratively is searched for an approximation of the optimal evacuation instructions. The usefulness of the optimization method compared to other optimization methods is the simultaneous optimization of the departure time, destination, and route instructions instead of the optimization of only one or two of these variables for a dynamic instead of static evacuation problem. In a case study, the functioning of the method is illustrated. The relative high fitness in the case study of the set of instructions following from the optimization method compared with the fitness of a set of instructions set up by straightforward rules (like evacuating to the nearest destination) shows also the usefulness of applying an optimization method to create a set of evacuation instructions.

In this paper we investigate a real-time traffic surveillance system based on a multi class first-order traffic flow model called F astlane . We demonstrate a dual extended Kalman Filtering approach in which the model state and parameters can be estimated simultaneously from real-time data. The innovation is that although F astlane maintains the dynamics of multiple vehicular classes (e.g. trucks, buses, cars), only the total mixed-class density is corrected by the filter, which is ‘translated’ into multi-class state corrections by means of state-dependent person car equivalents and class flow shares. Results on real data from a densely used freeway show that the DEKF procedure is able to reproduce accurate speeds and flows and physically plausible parameters.

Reliable tracking of cars from aerial video imagery is one of the main ingredients of microscopic traffic monitoring. Current tracking methods however are not able yet to track all the vehicles in all frames of video imagery taken by e.g. a helicopter. Several problem scenarios can be distinguished, like situations with many similar cars in congested traffic areas, cars that appear in low contrast compared to the background and cars that are occluded in some frames by other cars or by traffic signs. In this paper an improved method is described that continuously tracks all vehicles from their appearance into the viewing area until their exit. Our algorithm starts by separately tracking individual car pixels and the complete car region as a whole using the gradient-based optical flow method. A scale space approach is used to initiate the optical flow method. The best result as obtained from the intermediate results is used in the following statistical decision making step. Finally, either the best results are accepted and by applying a rigid body assumption, one displacement result is adapted for the car as a whole, or the best results are rejected, because even the best results fail a quality criterion. Continuation of these steps for all frames constitutes the final tracking result. This method solves most of the sketched problem scenarios as is illustrated by applying it on suited helicopter video imagery.

Accuracy and robustness with respect to missing or corrupt input data are two key characteristics for any travel time prediction model that is to be applied in a real-time environment (e.g. for display on variable message signs on freeways). This article proposes a freeway travel time prediction framework that exhibits both qualities. The framework exploits a recurrent neural network topology, the so-called state-space neural network (SSNN), with preprocessing strategies based on imputation. Although the SSNN model is a neural network, its design (in terms of input- and model selection) is not “black box” nor location-specific. Instead, it is based on the lay-out of the freeway stretch of interest. In this sense, the SSNN model combines the generality of neural network approaches, with traffic related (“white-box”) design. Robustness to missing data is tackled by means of simple imputation (data replacement) schemes, such as exponential forecasts and spatial interpolation. Although there are clear theoretical shortcomings to “simple” imputation schemes to remedy input failure, our results indicate that their use is justified in this particular application. The SSNN model appears to be robust to the “damage” done by these imputation schemes. This is true for both incidental (random) and structural input failure. We demonstrate that the SSNN travel time prediction framework yields good accurate and robust travel time predictions on both synthetic and real data.

We present a new approach to estimating the distribution of free speeds based on the method of censored observations. The original distribution-free method of Kaplan–Meier is generalized to include partially censored data, i.e. observations that are censored with a certain probability. This is achieved using a composite time headway distribution model that is estimated as well. Using synthetic data, the method is validated. The method is applied using real life data collected at different two-lane rural roads in the Netherlands to establish free speed distributions and the differences between different vehicle-types.