Search

This paper presents a stochastic model for spatially embedded social networks based on the ideas of spatial interaction models. Analysing empirical data, we find that the probability to accept a social contact at a certain distance follows a power law with exponent -1.6. With a simulation where the spatial distribution of vertices is defined by a synthetic population of Switzerland, we can reproduce the edge length distribution observed in the empirical data as well as some other typical properties of social networks.

Arbeitsberichte Verkehrs- und Raumplanung, 504

In a traffic network, capacities of parts of the network restrict the amount of transport units which can be handled by this network. For example a two lane highway can handle a larger amount of vehicles than an access road through the forest. The capacity of a given traffic network element is not fixed but influenced by parameters such as number of lanes, maximum speed, weather, view horizon and so on. These parameters also define the maximum capacity of intersections and roundabouts. Special shapes of intersections, particularly in urban regions, may further increase or decrease their capacity. This paper investigates how the capacity of such special intersections can be estimated with only the curbsides of the intersection as an input. It is also of interest if changes to the shape decrease the amount of space “wasted” for the traffic intersection while the capacity remains unchanged. In this case study one special intersection is examined: “Central” in downtown Zurich, Switzerland. The particularity of this intersection is that it partially behaves like a roundabout but also contains two uncontrolled intersections. Due to its central position in the city, the intersection is very busy with both individual cars and public transport vehicles. In the first part of this paper, a simulation model which is able to produce realistically behaving vehicles only by using information about the curb side locations of the intersection is described The simulation shows how vehicles produce and dissolve congestion and tailbacks, demonstrating that the topology of the road configuration is a major contributor to congestion. In the second part of the paper, the simulation changes the topology of the scenario based on the observed behavior of the vehicles. Using a feedback loop allows one to optimize the capacity of the intersection while its spatial extents are minimized.

Microsimulation tools are becoming increasingly important in traffic demand modeling. The major advantage over traditional assignment models lies in the fact that each traveler is simulated individually. This means, for example, that decision making processes can be included for each individual. The traffic demand is a result of different decisions made by each individual. Those decisions lead to plans which the individuals try to optimize. Because microsimulation includes decision making processes, some information about the individual's plans has to be given. On the other hand, traditional assignment tools like VISUM or EMME/2 use origin-destination matrices (OD-matrices) as inputs. Those matrices do not have any information about the chains of activities which define plans for individuals. This raises the question of reconstructing plans from OD-matrices, which is also discussed in this paper. The plans generated by this approach are simulated with MATSIM—a microsimulation model—and the outcome are compared to the results of a VISUM—a traditional assignment model—of the same OD-matrices. The comparison is carried out by a case study of the greater Zurich area of Switzerland.

TRANSIMS (TRansportation ANalysis and SIMulation System) is a microscopic simulation package for transportation planning, “microscopic” meaning that each traveler is resolved individually as a particle. This package is used with the goal of simulating 24-hours of all the traffic in Switzerland. This is used to “stress-test” TRANSIMS with respect to its capability to simulate large metropolitan areas with 10 million travelers or more. This paper gives an overview of the current status., Arbeitsberichte Verkehrs- und Raumplanung, 109

Technical Report / ETH Zurich, Department of Computer Science, 360

Arbeitsberichte Verkehrs- und Raumplanung, 49

This paper describes the parallel implementation of the TRANSIMS traffic micro-simulation. The parallelization method is domain decomposition, which means that each CPU of the parallel computer is responsible for a different geographical area of the simulated region. We describe how information between domains is exchanged, and how the transportation network graph is partitioned. An adaptive scheme is used to optimize load balancing. We then demonstrate how computing speeds of our parallel micro-simulations can be systematically predicted once the scenario and the computer architecture are known. This makes it possible, for example, to decide if a certain study is feasible with a certain computing budget, and how to invest that budget. The main ingredients of the prediction are knowledge about the parallel implementation of the micro-simulation, knowledge about the characteristics of the partitioning of the transportation network graph, and knowledge about the interaction of these quantities with the computer system. In particular, we investigate the differences between switched and non-switched topologies, and the effects of 10 Mbit, 100 Mbit, and Gbit Ethernet. As one example, we show that with a common technology – 100 Mbit switched Ethernet – one can run the 20 000-link EMME/2-network for Portland (Oregon) more than 20 times faster than real time on 16 coupled Pentium CPUs., Technical Report / ETH Zurich, Department of Computer Science, 344

CS technical report, 342

This thesis details the design of a modified MATSim agent-based simulation system that is driven by transit smart card data. A complete set of bus vehicle trajectories are reconstructed from the smart card transactions. The reconstructed trajectories are then used to estimate a number of regression models of the speed at which buses travel between stops. These models drive the simulation dynamics, along with a dwell time model which takes account of bus ridership, in a MATSim simulation where smart card transaction records have been transformed into an agent-based travel demand. Besides an exposition of the various elements of the data-driven system, as well as validation of the simulation output and a test case scenario, the thesis also covers a number of related topics, including a technique to improve simulation times for MATSim simulations in general, a method to distribute MATSim simulations across multiple computer nodes, a useful approach to perform parameter optimisation in MATSim using a Kriging meta-model, and ends with a technique to synthesise surrogate data that could be used to drive the simulation, while eliminating any associated privacy concerns.

The Multi-Agent Transport Simulation MATSim, ISBN:978-1-909188-77-8, ISBN:978-1-909188-75-4, ISBN:978-1-909188-78-5, ISBN:978-1-909188-76-1

The Multi-Agent Transport Simulation MATSim, ISBN:978-1-909188-77-8, ISBN:978-1-909188-75-4, ISBN:978-1-909188-78-5, ISBN:978-1-909188-76-1

The Multi-Agent Transport Simulation MATSim, ISBN:978-1-909188-76-1, ISBN:978-1-909188-78-5, ISBN:978-1-909188-75-4, ISBN:978-1-909188-77-8

The Multi-Agent Transport Simulation MATSim, ISBN:978-1-909188-77-8, ISBN:978-1-909188-75-4, ISBN:978-1-909188-78-5, ISBN:978-1-909188-76-1

The evolution of information technology brings an entirely new perspective to old issues of transportation and the problem of overloaded road traffic networks. At the forefront of progress in the field of information technology is the opportunity for the driver to acquire knowledge through media.The present study is aimed at investigating effects of spatial orientation in typical situations. To this end, it starts out from the following exemplary scenario: Traffic in the Zurich metropolitan area is congested. Vehicles often move at walking pace. Traffic demand leads to an average volume of 118 vehicles per kilometer. Every driver has planned his itinerary with the help of an off-the-shelf navigation device and sticks to his shortest route. In view of this situation, the question investigated in this study is: How much will the traffic situation improve if part of the drivers use real-time navigation information (such as may be available via smartphone)? The research to answer this question proceeds on the assumption that a driver behaves either in a “conventional” or in a “progressive” manner. The conventional drivers move along on the route they perceived as the shortest one when they planned it before starting on their trip. The progressive drivers are informed about the current traffic situation and head for their destination dynamically by choosing the currently most advantageous link at each traffic node on their trip. The decision processes of the informed drivers will be mapped in a simplified form and microscopically simulated using the MATSim software. A model postulated for the route choice describes the behavior of drivers guided by real-time navigation information, but not obstinately following it; their experience regarding the reliability of the traffic information also influences their route choice. The model analyzes how differing knowledge levels and modes of behavior of the drivers affect the state of the traffic system in the real-world setting of the Zurich metropolitan area. The results of the experiments testify to the existence of great differences in respect of the load on the road network, the mean daily travel times and the consequential properties of a trip up to the driver’s arrival time at his destination. – A key result is that all drivers benefit even when only part of them navigate by using current traffic information. Further results show in detail the time savings that each of the two classes of drivers achieves, and also how the entirety of drivers benefits from certain shares of informed drivers. Especially interesting for the analyst is the finding that the effect of descriptive and normative behavior in respect of route choice varies significantly. The scenario’s estimated mean saving potential of about 25 percent can be fully exploited if the informed drivers behave in a disciplined manner and follow the recommended links. When 30 percent of the drivers in the Zurich metropolitan area are guided by real- time navigation system information and comply exactly with it, the traffic density will be reduced from 118 vehicles to 56 vehicles per kilometer, and traffic speed will increase from four to 22 kilometers per hour. Starting from a share of 50 percent of informed drivers, traffic density will diminish to just above 30 vehicles per kilometer, and a driver will reach his destination at an average speed of little more than 50 kilometers per hour. The better distribution of the traffic may triple the distance of an informed driver, it is true; and yet it amounts to an 84 percent time saving for all drivers. – However, if more than 70 percent of the drivers go by real-time navigation system information, the traffic situation will again deteriorate to as many as 43 vehicles per kilometer moving at a speed of 34 kilometers per hour. This (probably unexpected) deterioration of the traffic situation at a high share of drivers being guided by real-time navigation system information asks for more research. Further analyses are required. Most likely they will show that to prevent this unwanted effect, the quality of the information must be improved. The hypothesis that suggests itself is that navigation system guidance must be based on marginal cost, which in turn requires that the traffic densities and the time-flow- capacity curves of the links