Search

Traditionally, traffic is studied at the level of individual vehicles (microscopic), or the level of road sections (macroscopic). Recently, a higher level of study has been introduced, the level of an area. In homogeneous traffic conditions, there is a relationship between the number of vehicles in an area (accumulation) and the average flow in that area (production). This paper studies this relationship under inhomogeneity. Using simulation on a grid network it is shown that the performance of a network is a smooth function of the average network density (accumulation) and the spatial variation of the density. Hence, we introduce a two-variable macroscopic fundamental diagram (TV-MFD) which also holds in a inhomogeneous transition state, for instance in queue build up. The TV-MFD can be used for network control, for instance at ramp metering installations or at a more aggregate level perimeter control.

This paper describes the results of calibration of the Nomad pedestrian simulation model using empirical data from laboratory experiments. The results of the calibration show that complex walker models with large amounts of parameters can indeed be calibrated. The estimated average parameter values are shown and discussed, as well as their significance and correlation. Furthermore, pedestrian behaviors are investigated using the estimated parameters values in various conditions, among which flow configuration, pedestrian heterogeneity and traffic conditions.

Emergency doors may be bottlenecks in the evacuation of a building. To assess and improve ways in which bottlenecks are used during evacuation conditions, knowledge regarding the microscopic and macroscopic phenomena at bottlenecks is required. Using data from laboratory experiments, parameters have been estimated using the microscopic simulation tool Nomad. The conclusion is that the pedestrian behavior shown upstream of bottlenecks in evacuation conditions is different from normal walking behavior. In the latter, especially in higher densities, pedestrians look for gaps which they can use to overtake other pedestrians when they have a higher free speed. In the experiments described here, pedestrians seem to have determined a path leading towards the door (the bottleneck), along which they only progress slowly, but where they stick to rather consistently. Overall, we could conclude that both types of behavior cannot be covered with a similar model type, but a multiregime model seems to be more appropriate, in which situations with low and high densities can be distinguished and dedicated models can be applied in the specific situations (regimes).

Large scale routing strategies require many data, and high computational power if the basic information unit is small. This paper checks the effects of a routing strategy based on aggregated information, the average speed, of a subnetwork. In a grid-network this routing is compared with shortest-distance routing, and data-demanding shortest-time routing. Contrary to the fixed routing, the proposed average-speed routing algorithm can avoid congestion quite long time, and it still needs very few data. Furthermore, it is less sensitive to fluctuations than the control based on speeds on all links. This is a good starting point for further control based on the macroscopic fundamental diagram.

Fusing traffic data from a variety of traffic sensors into a coherent, consistent, and reliable picture of the prevailing traffic conditions (e.g. densities, speeds, flows) is a critical and challenging task in any off- or online traffic management or information system which use these data. Recursive Kalman filter-based approaches provide an intuitive and powerful solution for traffic state estimation and data fusion, however, in case the data cannot be straightforwardly aligned over space and time, the equations become unwieldy and computationally expensive. This chapter discusses three alternative data fusion approaches which solve this alignment problem and are tailored to fuse such semantically different traffic sensor data. The so-called PISCIT and FlowResTD methods both fuse spatial data (individual travel times and low-resolution floating car data, respectively) with a prior speed map obtained from either raw data or another estimation method. Both PISCIT and FlowResTD are robust to structural bias in those a priori speeds, which is critically important due to the fact that many real-world local sensors use (arithmetic) time averaging, which induces a significant bias. The extended and generalized Treiber–Helbing filter (EGTF) in turn is able to fuse multiple data sources, as long as for each of these it is possible to estimate under which traffic regime (congested, free flowing) the data were collected. The algorithms are designed such that they can be used in a cascaded setting, each fusing an increasingly accurate posterior speed map with new data, which in the end could be used as input for a model-based/Kalman filter approach for traffic state estimation and prediction.

Microscopic pedestrian models describe individual pedestrian behavior and the interaction of pedestrians with other pedestrians and obstacles. Continuous time models generally calculate the acceleration of pedestrians due to repulsive or attractive interactions using for instance distances to other pedestrians and obstacles within a two-dimensional influence area. Two problems that usually arise with these types of models when simulating very large crowds are extreme accelerations that occur due to very short distances to other pedestrians and obstacles and large computational times to assess and to calculate the accelerations for individual pedestrians. This paper presents a hybrid pedestrian management algorithm that combines a traditional optimized time-based simulation and an event-driven simulation. This way, the task of assessing the surroundings and the task of dealing with interactions on very short distances are each treated in an optimized way leading to more reliable accelerations in high densities as well as shorter calculation times.

In the last few decades, a number of driving models aiming at modeling the longitudinal and lateral driving tasks have been put forward based on the analogy with self-driven many particle systems. Examples of such models are the social-forces type models for car-following behavior [1], the IDM (Intelligent Driver Model) and its modifications [2], and the MOBIL model [3] describing lane-changing behavior. Although these models can describe many phenomena in motorway traffic flow operations, a clear behavioral foundation has however been lacking so far.

Modeling breakdown probabilities or phase transition probabilities is an important issue when assessing and predicting the reliability of traffic flow operations. Looking at empirical spatio-temporal patterns, these probabilities clearly are not only a function of the local prevailing traffic conditions (density, speed), but also of time and space. For instance, the probability that start-stop wave occurs generally increases when moving upstream away from the bottleneck location.

Microscopic simulation models predict different forms of self-organization in pedestrian flows, such as the dynamic formation of lanes in bi-directional pedestrian flows. The experimental research presented in this paper provides more insight into these dynamic phenomena as well as exposing other forms of self-organization, i.e. in case of over-saturated bottlenecks or crossing pedestrian flows. The resulting structures resemble states occurring in granular matter and solids, including their imperfections (so-called vacancies). Groups of pedestrians that are homogeneous in terms of desired walking speeds and direction appear to form structures consisting of overlapping layers. This basic pattern forms the basis of other more complex patterns emerging in multi-directional pedestrian flow: in a bi-directional pedestrian flow, dynamic lanes are formed which can be described by the layer structure. Diagonal patterns can be identified in crossing pedestrian flows. This paper both describes these structures and the conditions under which they emerge, as well as the implications for theory and modeling of pedestrian flows.

Free speeds are defined as the speeds pedestrians like to walk with when they are not hindered by other pedestrians.

This paper discusses the validity of first-order traffic flow theory to describe two-dimensional pedestrian flow operations in case of an oversaturated bottleneck upstream of which a large high-density region has formed. Pedestrians passing the same cross-section inside of the congested region appear to encounter different flow conditions. In the lateral center, high densities and low speeds are observed. However, on the boundary of the congested region, pedestrians may walk in nearly free flow conditions. Visualising pedestrian flow data in the flow-density plane results in a large scatter of points having similar flows (bottleneck capacity), but different densities. Observations on congestion of pedestrian traffic over the total width of the cross-section are found to belong to a set of different fundamental diagrams instead of a single one. This has consequences for the estimation of the fundamental diagram describing pedestrian traffic.

Insights into pedestrian behavior, and tools to predict this behavior, are essential in the planning and design of public pedestrian facilities, such as transfer stations, shopping malls, and airports. Both macroscopic features of pedestrian flows and microscopic walking behavior underlying these features are important.

This contribution proposes a new approach to estimate model parameters of microscopic pedestrian models using individual pedestrian trajectory data. To this end, a generic approach is proposed that enables parameter identification for microscopic models in general and in particular for walker models. The application results provide new insight into the behavior of individual pedestrians, inter-pedestrian differences, as well as the resulting pedestrian flow characteristics. By comparing different models of increasing complexity, it is investigated which of the model amendments are significant from a statistical point of view and which are not. It is shown that besides anisotropy, finite reaction times play an important role in correctly describing microscopic walking behavior. The implications of these findings in the microscopic description of pedestrians flows are considered by studying the predicted flow operations at a narrow bottleneck. It turns out that the finite reaction times have a significant effect on the pedestrian flow operations.

Using vehicle trajectories for a motorway collected from a helicopter, and a newly developed approach to parameter identification, this paper provides new empirical evidence of multi-anticipative car-following by estimating the driver-specific parameters of the different multi-anticipative car-following models. We investigate the nature of the multi-leader stimuli, providing insight into the number of vehicles ahead to which drivers react and the kind of stimuli drivers respond to. The paper also shows that there is large inter-driver variability in multi-leader driving behavior. The implications of our research findings for microscopic modeling are discussed as well.

This article presents a study of the performance decrease caused by blocking one link. We predict which types of links are important for the well functioning of the network as a whole. Traffic is simulated on a regional network in which the links are blocked sequentially. Each of these outcomes is compared to the outcome of the fully operational network. We argue that traffic dynamics including spillback are important in the reduction of performance. We assessed the performance decrease for two situations, one in which people stick to their everyday routes and one in which they will adapt their routes to the new situation. A blocking of motorways turns out to influence the traffic flow most. In the situation that people stick to their everyday routes, urban roads close to the motorway are equally important: spillback effects causes delays also for through traffic on the motorway.

We present a macroscopic traffic flow model that explicitly builds on continuous individual driving behaviour. Not only do we start from classical car-following rules (like the kind that are used in microscopic simulators), the model also explicitly accounts for the finite reaction times of drivers, anticipation behaviour, anisotropy in driver responses and the finite space requirement of drivers in the stream. Moreover we allow variations in driver psychology, which lets drivers adopt different ‘driving styles’ dependent on traffic conditions, like the presence of a merging zone.

Empirical Traffic Data.- Lane-Change Maneuvers Consuming Freeway Capacity.- Empirical Description of Car-Following.- Optimization Potential of a Highway Network: An Empirical Study.- Observation, Theory and Experiment for Freeway Traffic as Physics of Many-Body System.- Fluctuation in Expressway Traffic Flow.- Calibration and Validation of Microscopic Traffic Flow Models.- Real Highway Traffic Simulations Based on a Cellular Automata Model.- Traffic Data Collection and Study with the BirdEye System.- Traffic Models and Theory.- Probabilistic Traffic Flow Breakdown in Stochastic Car Following Models.- A Wave-Based Resolution Scheme for the Hydrodynamic LWR Traffic Flow Model.- Various Scales for Traffic Flow Representation: Some Reflections.- Comparison of Congested Pattern Features at Different Freeway Bottlenecks.- Congestion Due to Merging Roads: Predictions of Three-Phase Traffic Theory.- Production, Supply, and Traffic Systems: A Unified Description.- Capacity Funnel Explained Using the Human-Kinetic Traffic Flow Model.- A Comparison of a Cellular Automaton and a Macroscopic Model.- Stochastic Description of Traffic Breakdown: Langevin Approach.- Towards Noised-Induced Phase Transitions in Systems of Elements with Motivated Behavior.- Using the Road Traffic Simulation "SUMO" for Educational Purposes.- Approximating Traffic Flow by a Schrodinger Equation - Introduction of Non-Reflecting Boundary Conditions.- Modeling and Managing Traffic Flow Through Hyperbolic Conservation Laws.- Conservation Versus Balance Laws in Traffic Flow.- Introducing the Effects of Slow Vehicles in a LWR Two-Flow Traffic Model.- Bifurcation Analysis of Meta-Stability and Waves of the OV Model.- Pinch Effect in a Cellular Automaton (CA) Model for Traffic Flow.- Traffic Networks and Driver's Behaviour.- Intersection Modeling, Application to Macroscopic Network Traffic Flow Models and Traffic Management.- Granular Transport Policy Is Cooperation Better than Optimization?.- Real-Time Motorway Network Traffic Surveillance Tool RENAISSANCE.- Minority Game - Experiments and Simulations.- Experimental Investigation of a Two Route Scenario with Construction Areas.- Phase Diagrams of an Internet Model with Multi-Allocation of Sites.- Adaptive Traffic Light Control in the ChSch Model for City Traffic.- Vehicle Flow and Phase Transitions in Traffic Networks.- Pedestrian Dynamics.- Anomalous Fundamental Diagrams in Traffic on Ant Trails.- Models for Crowd Movement and Egress Simulation.- Self-Organization in Pedestrian Flow.- A Comparison of Video and Infrared Based Tracking of Pedestrian Movements.- Statistical Analysis of the Panic in the Pedestrian Flow.- Group Formation of Organisms in 2-Dimensional OV Model.- Simulations of Evacuation by an Extended Floor Field CA Model.- Pedestrian Pulse Dispersion in an Underground Station.- Population and Distance Criteria for Pedestrian Decisions.- Simulation of the Evacuation of a Football Stadium Using the CA Model PedGo.- Granular.- Statistical Properties of Dense Granular Matter.- Small Scale Modelling of Large Granular Systems in a Centrifuge.- Weakly Frictional Granular Gases.- The Physics of Overcharging.- Computer Simulations of Magnetic Grains.- Simulation of Particle Transport and Structure Formation by Deposition.- Experiment and Simulation of Electrospray Particle-Flows for Controlled Release of Drugs.- The Branly Effect and Contacting Grains in a Packing.- Universal Wide Shear Zones in Granular Bulk Flow.- Numerical Stress Response Functions of Static Granular Layers.- Granular Segregation in Hydraulic Ripples.- Grain Motion Under Air Flow.- Bubble and Granular Flows : Differences and Similarities.- The Global Property of the Dilute-to-Dense Transition of Granular Flows in a 2D Channel.- From Granular Flux Model to Traffic Flow Description.- Simple Models for Compartmentalized Sand.- Several Numerical Approaches of Granular Flows Applied to Inclined Plane Studies.- A Modified Tetris Model Including the Effect of Friction.- An Aspect of Granulence in View of Multifractal Analysis.

Gaining insights into pedestrian flow operations and assessment tools for pedestrian walking speeds and comfort is important for e.g. planning and geometric design of infrastructural facilities, as well as for management of pedestrian flows under regular and safety-critical circumstances. Pedestrian flow operations are complex, and vehicular flow simulation modeling approaches are generally not applicable to pedestrian flow modeling.

The highly non-linear characteristics of the freeway travel time prediction problem require a modeling approach that is capable of dealing with complex non-linear spatio-temporal relationships between the observable traffic quantities. Based on a state-space formulation of the travel time prediction problem, we derived a recurrent state-space neural network (SSNN) topology. The SSNN model is capable of accurately predicting experienced travel times - outperforming current practice by far - producing approximately zero mean normally distributed residuals, generally not outside a range of 10% of the real expected travel times. Furthermore, analyses of the internal states and the weight configurations revealed that the SSNN developed an internal models closely related to the underlying traffic processes. This allowed us to rationally eliminate the insignificant parameters, resulting in a Reduced SSNN topology, with just 63 adjustable weights, yielding a 72% reduction in model-size, without loss of predictive performance.