Large-scale multi-agent transportation simulations

Abstract

In a multi-agent transportation simulation, each traveler is represented individually. Such a simulation consists of at least the following modules: (i) Activity generation. For each traveler in the simulation, a complete 24-hour day-plan is generated, with each major activity (sleep, eat, work, shop, drink beer), their times, and their locations. (ii) Modal and route choice. For each traveler in the simulation, the mode of transportation and the actual routes are computed. (iii) The Traffic simulation itself. In this module, the travelers are moved through the system, via the transportation modes they have chosen. (iv) Learning and feedback. In order to find solutions which are consistent between the modules, a relaxation technique is used. This technique has similarities to day-today human learning and can also be interpreted that way.-Besides, one needs input data, such data of the road network, or (synthetic) populations. In the future, further modules need to be added, such as for housing and land use, or for freight traffic. Using advanced computational methods, in particular parallel computing, it is now possible to do this for large metropolitan areas with 10 million inhabitants or more. We are currently working on such a simulation of all of Switzerland. Our focus is on a computationally efficient implementation of the agent-based representation, which means that we in fact represent each agent with an individual set of plans as explained above. We use a database to store the agent's strategies, then load them into the simulation modules as required, and feed back individual performance measures into the database. This approach allows that additional modules can be coupled easily, and without destroying computational performance. 1 Assuming 3 to 3.5 trips per person per day, this will result in about 20-25 million trips. This number includes pedestrian trips (like walking to lunch), trips by public transit, freight traffic, etc. The number of car trips on a typical weekday in Switzerland is currently about 5 million (see Vrtic (2001) for where the data comes from). The goal of our study is twofold: © Investigate the computational challenges and how they can be overcome. © Investigate what is necessary to make a simulation system realistic enough to be useful for such a scenario, and how difficult this is. This paper gives a report on the current status. Section 2 describes the simulation modules and how they were used for the purposes of this study. Section 3 describes the input data, i.e. the underlying network and the demand generation. Besides "normal" demand, we also describe one where 50 000 travelers travel from random starting points within Switzerland to the Ticino, which is the southern part of Switzerland. We use this second scenario as a plausibility test for routing and feedback. This is followed by Sect. 4, which describes some results and Sect. 5, which describes issues related to computational performance of the parallel micro-simulation. The paper ends with a discussion and a summary. 2 S...

Spatiotemporal Patterns of Urban Human Mobility

“…In general, these data sources lack the demographic information of individuals, the traces of their locations are not continuously collected and there is no detailed information about the specific choices of the individuals daily activities. In consequence, discrete choice models and transportation forecasting models [19,20,21] cannot be directly applied. In the light of this massive data availability, methods of analysis and modeling based in statistical physics are a suitable alternative.…”

Autocorrelations in the totally asymmetric simple exclusion process and Nagel-Schreckenberg model

“…While many results are known rigorously for the TASEP, our understanding of the NaSch model and its further generalizations typically rely on numerical simulation. This is particularly true of traffic network models, in which the NaSch model is often a component (see for example [8][9][10]). …”

Geosimulation, Automata, and Traffic Modeling

“…Various characteristics can be associated with the cells that form a queue, e.g., length, flow capacity, free flow velocity, free flow travel time, etc. (Cetin, et al 2001). In this way, automata cells and lattices can be used to build realistic traffic environments.…”

“…Other rules have been developed to simulate signal stopping behavior (Barrett, et al 1999) and traffic movement at junctions, with "gap acceptance" functions that determine how long vehicles must wait at a junction before they can proceed (Wylie, et al 1993). In models where collections of vehicles are simulated as traffic queues (Barrett, et al 1999, Cetin, et al 2001, Rickert, et al 1996, entrance and departure from vehicle queues can also be simulated, with vehicles leaving a queue freeing up space on a link, allowing another vehicle to join the queue.…”

“…In this sense, automata models resemble traditional queuing lane models. However, traditional queuing lane models are not truly multi-lane in their design (Cetin, et al 2001); they approximate multiple lanes by switching the positional order of vehicles to make it appear as if passing has occurred. In automata models, however, parallel lattices can be constructed adjacent to each other, facilitating the simulation of movement between lanes.…”