Micro-simulation Tools

This section describes the current status of the main micro-simulation tools that have been used to model UK road networks.

AIMSUN (Advanced Interactive Microscopic Simulator for Urban and Non-urban Networks) is a software tool capable of reproducing real traffic conditions in an urban network which may contain both expressways and arterial routes. It is based on a microscopic simulation approach. The behaviour of every single vehicle in the network is continuously modelled throughout the simulation time period, according to several driver behaviour models (car following, lane changing, gap acceptance). AIMSUN2 is a combined discrete-continuous simulator: there are some elements of the transportation system (vehicles, detectors) whose state changes continuously over the simulated time period, while there are other elements (traffic lights, entrance points) whose state changes discretely at specific points during the simulation time. It provides very detailed modelling of the traffic network: it distinguishes between different types of vehicles and drivers; it can deal with a wide range of network geometries; it can also model incidents, conflicting manoeuvres, etc.

AIMSUN2 needs three types of input data: the network description, the traffic signal control plans and the traffic conditions. The network description contains information about the geometry of the network, turning movements, layout of links (or sections) and junctions and location of detectors. The traffic control plans define the signal stages and their duration for signal controlled junctions, the priority definition for unsignalized junctions and any required ramp-metering information. The essential inputs for the simulator are the traffic flows for the input links, the turning proportions at junctions and the initial state of the network.

Recently, as part of a DGXVII funded project, AIMSUN2 has been linked to the UK SCOOT UTC system. AIMSUN2 passes details of the vehicle flows in the network to SCOOT. SCOOT uses this flow data to calculate signal timings which are designed to minimise the amount of delay experienced by all the vehicles in the network. It passes these signal timings back to AIMSUN2 where they are implemented.

The outputs provided by AIMSUN2 include a continuously animated graphical representation of the traffic network, a printout of statistical data (flows, speeds, journey times, delays, stops, fuel consumption, pollution emissions), and data gathered by the simulated detectors (counts, occupancy, speeds, queue lengths).

AIMSUN2 is integrated into the GETRAM simulation environment (Generic Environment for Traffic Analysis and Modelling), which consists of a traffic network graphical editor, a network data base, static assignment models, temporal simulation models and a module for storing and presenting results.

A parallel version of AIMSUN2 has recently been completed (Barcelo, 1996), so that computationally expensive problems can be tackled. It uses a Sun SparcStation 1000 with 8 processors under Solaris 2.4 via Solaris threads. It was developed in the framework of the ESPRIT Programme, High Performance Computing Initiative, as a subproject of Project PACOS, PCI project EP-9602.

DRACULA (Dynamic Route Assignment Combining User Learning and Micro-simulation) is a microscopic traffic network modelling suite, conceived and developed at the Institute for Transport Studies, University of Leeds over a five year period. It is part of the SATURN suite of programs developed at ITS and is exploited with WS Atkins consultants. The development, testing and validation of the model have been primarily funded by a large grant from the UK Engineering and Physical Sciences Research Council, although some early applications of the model were possible under funding from the EC's DRIVE II Telematics programme. Applications of this model in progress, or to commence shortly, include the study of congestion based road pricing, real time traffic signal control, dynamic route guidance, segregated busway design, emergency evacuation procedures (e.g. Following chemical explosions, floods), and strategic (inter-urban) modelling. Presently DRACULA is able to model the effect of policy, demand and network changes on route and departure time choice, but there are plans to extend this range of choices to cover "higher level" choices, concerning the mode of travel and residential/work location.

We refer to DRACULA as a "supermodel" because it incorporates a range of possible assumptions and levels of detail, which may be selected by the transport planner depending on the objectives of the study. For example, driver choices (e.g. of route) may be modelled at the level of the individual drivers or at an aggregate level; one second increment discrete micro-simulation may be used to move drivers along their chosen routes, or macroscopic traffic models may be used; route choice may be assumed to be the only choice open to drivers (or even be fixed), or we may model departure time choice, en route diversion in response to unexpected conditions, or the details of lane choice switching to avoid blocked or heavily queued lanes. A selection of a particular combination of supermodel parameters gives rise to a particular model within the DRACULA suite. (Of course, within such a model there will be model parameters which need to be calibrated to each particular network).

DRACULA differs from "traditional" equilibrium approaches in that it explicitly models the day-to-day dynamic evolution of driver behaviour and traffic conditions, as a discrete time stochastic process. At its most detailed and comprehensive level, DRACULA has the following structure:

  1. Initialisation. For each potential traveller in the network, assume initial perceived travel costs for each link in the network. Set day counter k=0.
  2. OD demand. Increment day counter: k=k+1. Generate the set of travellers who will actually make a car journey on day k.
  3. Travel choice. Each individual travelling on day k selects a departure time and route based on their currently perceived travel costs.
  4. Supply variability. Simulate day-to-day variability in characteristics of the traffic (supply) model, to represent rain/snow, accidents, parked vehicles, breakdowns, etc.
  5. Traffic model. Load the travel choices in step 3 onto the network using a one-second increment micro-simulation model, recording individual travel experiences. During this stage, en route diversion from the originally selected route may occur.
  6. Learning. Via some kind of learning mechanism, each individual forms an updated perceived (day-averaged) travel cost for each link/turn and arrival time interval. Return to step 2 to simulate the next day.
This stochastic process approach possesses a sound theoretical basis, and indeed has a number of advantages over its equilibrium counterpart. In rough terms, this theory establishes that under mild conditions, such a model will settle down, after an initial transient period, to a characteristic level of variability - this representing the real day-to-day variability in road conditions that we all know exists. From a practical viewpoint, the separation of traveller behaviour and traffic flow/congestion in the day-to-day approach is the key to the enormous flexibility and range of assumptions that DRACULA may incorporate, being highly suitable for further development.

In terms of modelling special events such as accidents, breakdowns or weather conditions having a severe effect on road capacity, DRACULA is ideal, being able to model how drivers respond in terms of en route diversions (when seeing queues or receiving radio information, for example), taking account of how they weight their typical experience (stored in their personal history file) compared to extreme conditions. On the traffic flow side, second-by-second micro-simulation using lane changing and gap acceptance models is the only feasible technique in existence for modelling severe queue spillback (as opposed to the vertical queue assumption of typical macroscopic and mesoscopic approaches), the effect on the details of driver behaviour (e.g. weaving through gaps in stationary cross-traffic at a junction, lane changing in response to accidents), and the dynamic propagation of congestion backwards in the system (often referred to as the shock wave phenomenon).

Paramics (Quadstone) (PARAllel MICroscopic Simulation) is a suite of software tools for microscopic traffic simulation where individual vehicles are modelled. The original software was developed from a program called MIXSIM which was funded through the DRIVE project 87-88. The program was developed by SIAS and EPCC and was further developed when Quadstone/SIAS bought out EPCC. In 1997 SIAS and Quadstone dissolved their agreements and have since marketed their own versions of the program.

The Quadstone version of Paramics can only be used in the UK for applications which involve the real time control of traffic. Following the split from SIAS, Quadstone have further developed the package and are now up to version 3.0. Quadstone versions of Paramics are being used in Singapore, Japan, Argentina, US and Malaysia. The package includes the following different modules:

The package can handle up to 4 million links, 1 million nodes and 32000 zones and 1 million control points (eg stop lines). It is Windows based and in reality there is no real limit to what can be modelled within the software, just a dependence on the hardware. The program works using a default 0.5 second timestep. There is a random release of vehicles onto the network. Results are repeatable if the same random seed is set. The random seed is used to determine the aggressiveness, awareness and variability of headways.

Paramics (SIAS) Development of the SIAS version of the program has largely been on an on-project basis. The Paramics software is portable and scaleable, allowing a unified approach to traffic modelling across the whole spectrum of network sizes, from single junctions up to national networks. Paramics claims to simulate the traffic impact of signals, ramp meters, loop detectors linked to variable speed signs, VMS and CMS signing strategies, in-vehicle network state display devices, and in-vehicle messages advising of network problems and re-routing suggestions. Vehicle re-routing in the face of ITS is controlled through a user-definable behavioural rule language for maximum flexibility and adaptability.

The Paramics software continues to undergo further development, driven by contract work and the continued incorporation of new technology in real-world transport systems. Currently development is underway in the following areas: detailed modelling of noise and exhaust pollution; multi-modal transportation simulation; traffic state determination from on-line vehicle counts; and provision of predictive traffic information for in-vehicle services.

SISTM (SImulation of Strategies for Traffic on Motorways) has been designed to study motorway traffic in congested conditions with the aim of developing and evaluating different strategies for reducing congestion. The program is not sold commercially and but is available from TRL or the Highways Agency. Development started in 1988. The current version of the program is version 5.

SISTM can assess

It has been developed for the UK Highways Agency but is available to anyone requiring the modelling of motorways.

It is a microscopic motorway simulation with a car following algorithm that uses a modified Gipps' equation. Driver behaviour is described by 2 parameters; aggressiveness and awareness, and these are used to produce distributions of desired speed and indirectly desired headway. The time increment used is 5/8th second. Lane changing is controlled through a lane changing stimulus with the user specifying the desire to change lanes. When making a lane changing manoeuvre, a driver is allowed to accept an "unsafe" headway temporarily. This is to allow smooth merging to take place when a driver has to move into a particular lane.

Useful technical features:

VISSIM (German for Traffic in Towns - Simulation) models transit and traffic flow in urban areas as well as interurban motorways on a microscopic level. It is a commercial product with continuos add-ons provided by research institutions.

Results of VISSIM are used to define optimal vehicle actuated signal control strategies, test various layouts and lane allocations of complex intersections, test the location of bus bays, test the feasibility of complex transit stops, test the feasibility of toll plazas, find appropriate lane allocations of weaving sections on motorways etc. VISSIM is coupled with micro-scale decentralised controllers of various signal control manufacturers to test their control strategies in detail before they are implemented. VISSIM is a multipurpose simulator aimed for technical staff at cities responsible for signal control, transit operators, city planners and researchers to evaluate the influence of new control and vehicle technologies.

The traffic flow model of VISSIM is a discrete, stochastic, time step based microscopic model, with driver-vehicle-units as single entities. The model contains a psycho-physical car following model for longitudinal vehicle movement and a rule-based algorithm for lateral movements (lane changing). The model is based on the continuous work of Wiedemann at the University of Karlsruhe and further calibrated and validated by PTV. Vehicles follow each other in an oscillating process. As a faster vehicle approaches a slower vehicle on a single lane it has to decelerate. The action point of conscious reaction depends on the speed difference, distance and driver dependent behaviour. On multi-lane links vehicles check whether they improve their speed by changing lanes. If so, they seek acceptable gaps on neighbouring lanes. Car following and lane changing together form the traffic flow model, being the kernel of VISSIM.

Useful technical features


Barcelo (1996) The parallelization of AIMSUN2 Microscopic Traffic Simulator for ITS applications, presented at the 3rd World Conference on Intelligent Transport Systems", held in Orlando on October 14-18, 1996.

Back to the Micro-simulation Home Page.