Software Environments to Support Traffic and Transport Analysis Methodologies

Software Environments to Support Traffic and Transport Analysis Methodologies


J Barcelo, H Grzybowska, Universitat Polit├Ęcnica de Catalunya, ES; S Pardo, TSS-Transport Simulation Systems, ES



The current starting point of most of the advanced traffic and transport projects - either aimed at conducting the analysis of an already existing system, the design of a future system, the evaluation of the potential impacts of a future intervention, or the design and operation of an advanced system - combines the ideas, wishes and objectives of the user, usually a traffic authority, with the requirement to exploit as far as possible the available related information: traffic data, digital maps, etc. This raises practical problems especially in the case when the analyst conducting the study or developing the application is not the one responsible for collecting the required data. At most the analyst will be able to define the specifications for data collection and then hope that they will be followed. All this data usually comes from various sources. To be able to use it in an efficient way the analyst should be able to integrate all of it in an easy way into a single exploitation platform. The main functions of such a working platform should provide graphical support to: Importing digital maps from the most common GIS and CAD systems; exporting the models and their outputs for exploitation in the other systems, as for example, the results visualization in a GIS; translating the digital maps into skeletons of the simulation (if the quality of the data and the presence of semantic information in the origin source make it possible); importing traffic and transportation data for use in the model, i.e. an Origin-Destination matrix from an Excel file, or a transport planning model; capability to numerically handle the traffic and transportation data, as for example generating local Origin-Destination matrices from global matrices for local analysis, or adjusting and updating mobility matrices from flow counts; exporting the data resulting from the simulation experiments, i.e. to an ODBC database for its standard access by any other statistical analysis software tool.

These modelling and analysis requirements ask for a new working environment purposely conceived and designed to efficiently assist and support the analyst in all steps of the traffic and transportation studies, and at the same time expand the territory of traffic simulation: from transport and traffic analysis to support operational applications (i.e. traffic management). AIMSUN NG is a working environment conceived to support such methodological approach to transport analysis, which integrates modules purposely designed and developed to support the traffic and transportation analysis tools / algorithms to address the different types of traffic and transportation problems of interest. The main components for Transportation Analysis currently integrated in AIMSUN NG are:

Kernel: a common, extensible, object environment that can be used by many transport applications. The most important consequence of having an Extensible Object Model and a Data Base shared by all traffic and transport analysis models implemented as plug-ins in AIMSUN NG is that it allows a Multi-level Network representation, that is networks with micro and macro details, or edit at macro level (link-node edition) and refine to micro level just when and where required
Filters: access to/from most used data sources and formats. The Data Import Filters provide functions enabling the software environment to accept inputs from other models, or work directly on road digital maps to build the transport models. Examples of functions of these filters are those for: Importing CONTRAM, CUBE or EMME/2 Networks, providing CAD Support, and GIS Interface (creating Network geometry automatically when the additional necessary information is available - i.e. number of lanes per section, flow directions, turnings at intersections)
Traffic Tools: access to the necessary software tools to carry out a project from start to finish, to set up the working environment for an application, or to support advanced technologies and their application to modern traffic management systems conducting more sophisticated studies based on a dynamic view complementing the static estimates provided by the assignment tools, combining a traffic assignment model, with a microscopic traffic simulator as AIMSUN. This combination requires a consistency guaranty in the correspondence between the network representation at levels macro and micro. The proposed approach based on sharing the extensible model and the data base provides a solution to this problem. It also allows a more efficient solution to the problems raised by the manipulation of trip matrices, namely in the processes of time-slicing origin to destination matrices to generate inputs to microsimulation models in order to emulate the time variability of the traffic demand, or adjusting the origin to destination matrices from observed flow counts on subsets of network links. A solution which exploits the common data structures with bi-level optimization procedures is described. This architecture also allows to dynamically select subnetworks of large networks for detailed studies. Two key problems solved by this approach are the interactive graphic extraction of the subnetwork model and the interactive estimation of the local origin to destination matrix.
GUI: graphical user interface to set up a user-friendly environment for all the required tasks

ITS applications have also generated new ways of using microsimulation to design and evaluate traffic management strategies and support the decision making processes This has been specifically addressed in a the System Analysis Module, another component of the software environment, that includes the possibility of working in real time combined with ALMO, a structured and geographically referenced traffic data-tool that merges different data assets and who manages among other things traffic data, traffic messages, weather data or event calendars. By means of an intelligent filter it is possible to restore the data asset corresponding to a special traffic situation.


Association for European Transport