VISSIM: 3rd UK USER Group Meeting

(Imperial College London - April 16th 2003)

more info on: http://www.ptv-newcastle.co.uk/UG3/program.htm

 

Summary of presentations (the programme can be accessed on:

http://www.ptv-newcastle.co.uk/UG3/UGM3_program_outline_0303.htm):

 

Optimising Traffic Signals in VISSIM. A Comparison with LINSIG and TRANSYT Models

By Sonal Ahuja (Mott MacDonald)

 

Pedestrian Simulation with VISSIM

By Paul Clifford (Babtie)

 

Use of VISSIM for Strategic Highway Modelling

By Dave Keenan and Stuart McNaughton (Faber Maunsell)

 

Use of external data sources to develop VISUM models: Case study PRISM (Policy Responsive Integrated Strategic Model for the West Midlands)

By Serbjeet Singh Kohli

 

Measurement of saturation flow and degree of saturation across signal controlled stoplines in VISSIM

By Ioannidis Ioannis (Transport for London)

 

A Traffic Model for Dalkeith

By Chris Pownall (MVA)

 

Application of Fuzzy Logic to Traffic Signal Control under Mixed Traffic Conditions in Developing Countries (SMEED Price winner UTSG conf. 2003)

By Budi Yulianto (TORG, School of Civil Engineering and Geosciences, University of Newcastle)

 

Optimising Traffic Signals in VISSIM. A Comparison with LINSIG and TRANSYT Models

By Sonal Ahuja (Mott MacDonald)

Abstract

Traffic signals are essential traffic management tools in urban conditions.  The design of signals has been an evolving science, driven by the need to make best use of available road space, and changing policy objectives, moving away from capacity maximisation or delay minimisation for private vehicles, towards priority operations for public transport, recognising the passenger numbers transported. 

 

Traditional design tools to calculate signal timings in the UK are TRANSYT and LINSIG.  These were initially used in an application for congested corridors in Wolverhampton, aimed at designing demand responsive traffic signals that allow for extra priority for public transport. LINSIG and TRANSYT failed to represent the congested corridor properly:

§    inaccurate representation of blocking back, the vertical queuing assumptions underestimating the interactions between closely spaced junctions

§    inability to model dynamic changes in demand, particularly pedestrian demand

§    poor validation, i.e. fit with observed conditions

 

As an alternative design tool a VISSIM microsimulation model was set up to assess and design the appropriate signal timings.  This involved development of a traffic signal optimiser that minimises some measure of delay for road traffic whilst allocating priority to public transport.

 

Comparative results show that:

§    traditional, average flow rate based tools such as TRANSYT and LINSIG have severe limitations in evaluating congested corridors

§    microsimulation, because of its dynamic and disaggregate nature, can produce a much better fit to observed conditions

§    the VAP vehicle actuated programming capability of VISSIM enables us to develop a signal optimiser that will both calculate an optimum signal design and show its operation in reality.

Pedestrian Simulation with VISSIM

By Paul Clifford (Babtie)

Abstract

This presentation will set out current experience with the simulation of pedestrian movements and behaviour with VISSIM and explore the potential to assist interchange design, both on the highway and for public transport systems. The integration of highway and pedestrian networks and the analysis criteria for performance review will be discussed.

Use of VISSIM for Strategic Highway Modelling

By Dave Keenan and Stuart McNaughton (Faber Maunsell)

Abstract:

The presentation will introduce the use of VISSIM for large-scale and strategic modelling purposes. 

Two examples will be discussed:

 

1.         Objective One Study, South Yorkshire

This model covers the motorway network in South Yorkshire and is used primarily as a development control planning tool.  A link is provided to a MapInfo GIS database with the aim of assessing the impact of proposed developments locally and on a wider, strategic scale.  The model has also been used to extract Average Speed v’s Location plots and Level of Service profiles, which has enabled closer analysis of individual junctions and examination of wider network performance.

 

2.         M1, J.40 Ramp Metering Study

M1 in the vicinity of J.39 – 42 (just south of the M62) suffers from severe congestion in the weekday peak periods due to large volumes of traffic attempting to merge with the mainline flow. Under work commissioned by the Highways Agency FaberMaunsell constructed a three-junction network of the M1 motorway in the area, to examine the potential benefits available from ramp metering of the merge flows. Journey time measurements and the concept of Level of Service categories were used extensively to monitor changes in mainline performance, and compare the benefits available from different methods of metering at three-year intervals through to 2011.

 

Return on top

Use of external data sources to develop VISUM models: Case study PRISM (Policy Responsive Integrated Strategic Model for the West Midlands)

By Serbjeet Singh Kohli

Abstract:

The presentation explains the use of various existing data sources to develop large VISUM networks. Developing large networks in VISUM, or as a matter of fact in any other modelling package, is a time consuming exercise. VISUM provides the flexibility of converting external data sources to VISUM networks. VISUM offers a unique feature, through the Shapefile converter and other interfaces, to create the basic network consisting of links and nodes from GIS information such as landline data, digitised OS maps, Featureline database etc. Further information about the link types/lengths can also be imported to VISUM in the conversion process. Hence landline data and digitised OS maps become the basic data source for highway network building. Public transportation networks can be converted from VIPS networks to VISUM networks as VIPS has a direct format interface with VISUM. Information about public transportation lines, routes, timetables etc. can be imported directly from VIPS into the VISUM network format. Area level zoning information such as census statistics, land use information and socio-economic data can also be imported from MAPINFO/GIS database as a separate file. This file can be appended to the link and node networks by reading the network file additionally to the VISUM highway network. VISUM can read EMME/2 files therefore information from the existing EMME/2 models can be imported into VISUM networks. VISUM has an interface with MS ACCESS, which works as a key in the integration of different networks.

 

Mott MacDonald is currently in the process of building a comprehensive highway and public transportation network for the West Midlands strategic transport model PRISM (Policy Responsive Integrated Strategic Model). The model covers the entire country in 898 zones and has 13000 nodes and 20000 links. Mott MacDonald has developed the highway network and the zoning system for PRISM using the Featureline database through the Shapefile converter. MAPINFO representation of the existing EMME2 strategic model has been used to obtain link characteristics such as free flow speeds, capacities and number of lanes for the VISUM network. Flow-delay relationships from a previously existing strategic model developed in Q-VIEW have been used to define the speed flow functions for various link types in the VISUM network.

 

The public transportation network has been imported from the bus network from the existing VIPS model. Since the bus network uses the road network, for development of PRISM network the bus network had to be merged with the highway network. The facility of exporting VISUM networks to other databases like MS ACCESS was used to achieve this. In addition VISUM has the facility for matching bus routes on the highway network if the bus stops are merged with the highway network.

This facility was used to generate bus routes in the highway network. For the rail network, data from existing CENTRO model for West Midlands in VIPS, and EMME/2 data from PLANET have been merged using the MS ACCESS interface of VISUM.

 

In summary VISUM provides a unique set of interface features that help the development of a network faster. Our experience with the development of a mammoth network for PRISM in a relatively short time has been more successful than conventional methods.

Measurement of saturation flow and degree of saturation across signal controlled stoplines in VISSIM

By Ioannidis Ioannis (Transport for London)

Abstract:

This is a presentation of a subroutine that is based on the SCATS methodology of measuring the degree of saturation and consequently the saturation flow across a signal controlled stopline. The technique relies on the presence of a detector per lane just downstream of every signal stopline.

    The subroutine can be adjusted quite easily to monitor as many stoplines and number of detectors as the user requires for every specific junction. A separate copy of it has to be attached to the VAP file that contains the control logic of every signal controlled junction. All the information needed about detector number and signal group (phase) numbers are contained in the VAP file that controls the junction and the saturation flow subroutine, receives the information necessary from there.

 

Methodology

  The degree of saturation according to this technique is the ratio of the time vehicles are occupying a detector plus the total of time headways when vehicles follow each other under saturated conditions over the time the traffic lights are on green or leaving amber.

  The time when vehicles are occupying a detector, the number of vehicles as well as the time when traffic lights are green or amber is measured accurately during every single cycle.

  The saturated time headway however for each lane is not a variable that can be measured accurately. A value for this headway has to be assumed for each lane which is used at the beginning of the simulation period. At the end of each of the first sets of green the degree of saturation is calculated using the value of saturated time headway assumed initially. If the degree of saturation as a result of this value is above 1.0 then this means that vehicles were travelling closer to each other during the green period just finished than the time headway used for the calculation. As a result a new value of the time headway is now worked out that will be used for the calculation of the degree of saturation at the end of the green in the next cycle. This is done by dividing the part of the green plus amber time when the detectors were not occupied by vehicles by the total number of vehicles that went over the detector in the green and amber period just finished.

   This process is repeated until the value of the time headway in each lane stops being decreased and stabilises around a value that represents the actual saturated time headway for the way the geometric characteristics of that lane have been modelled in VISSIM at that point and the vehicle characteristics of the traffic composition using this lane.

    From trials using this methodology and this subroutine in VISSIM models it was found that this process takes about 15 minutes of simulated time and then the saturated time headway reaches a stable value. Beyond this point the degree of saturation for each lane or across a stopline can be considered an accurate measure of what is the degree of saturation across each stopline of the junctions that are monitored by this subroutine.

    Having determined the degree of saturation the saturation flow for each stopline is calculated from the formula:

 

 

                                        Traffic demand            v * C

 Degree of saturation = ------------------------ = --------------

                                         Traffic capacity          S * g’

 

 

       Where            v = traffic demand  (veh/hour)

                              C = cycle length    (seconds)

                              S = saturation flow  (veh/hour)

                              g’ = eff.green         (seconds)

A Traffic Model for Dalkeith

By Chris Pownall (MVA)

Abstract:

A traffic model of the centre of Dalkeith has been created on behalf of Midlothian Council, to predict the impact on traffic flow and parking requirements of the redevlopment of the town centre areas. The model includes all key links in the town cnetre area and is capable of representing the performance of streets and junctions under current week-day peak traffic levels.

The model has been used to test a number of proposed traffic management schemes. Most of these involve the diversion of A68 trunk route which currently runs through the town centre onto other routes, in order to reduce traffic levels in the principal shopping areas.

 The model will make use of VISSIM's 3D graphics capabilities, to allow stakeholders to see the combined impact of the redevelopment proposals and the associated traffic management measures on the town.

 

Return on top

Application of Fuzzy Logic to Traffic Signal Control under Mixed Traffic Conditions in Developing Countries (SMEED Price winner UTSG conf. 2003)

By Budi Yulianto (TORG, School of Civil Engineering and Geosciences, University of Newcastle)

Abstract:

Traffic signal control is commonly used at road intersections to minimise travel time and delay. In time-varying traffic conditions, a fixed time controller becomes inflexible and not efficient. Using fuzzy logic, traffic signal controllers can be made more adaptive to traffic demands. However, most existing applications of fuzzy logic traffic signal controls are based on non-mixed traffic conditions, where they consider the passenger car only and neglect small-size vehicles such as motorcycles.

This paper describes the design and evaluation of an adaptive traffic signal controller based on fuzzy logic for an isolated four-way intersection with specific reference to mixed traffic (including high proportion of motorcycles). The controller is designed to be responsive to real-time traffic demands. The fuzzy controller uses video image processing to capture traffic data such as average occupancy rate (%) and maximum queue length (in metres) from each approach of the intersection. The proposed Fuzzy Logic Signal Controller (FLSC) uses maximum queue lengths and average occupancy rates collected during the previous cycle in order to estimate the number of seconds of green time required by each set of signal groups (stage) during the next cycle.

The effectiveness of the proposed fuzzy logic signal controller was examined and analysed by the simulation program VISSIM. The performance of this controller is compared to an optimised fixed time controller for different traffic conditions on a simulated an isolated four-way intersection. Simulation results showed better performance of the fuzzy logic signal controller when compared to an optimised fixed time controller in terms of the average travel time and the average delay of vehicles, especially under time-varying traffic conditions.

Keyword: Traffic signal control; Fuzzy logic; Mixed traffic; Video image processing; Isolated intersection; Simulation, VISSIM.

 

 

Return to UGM3

Return to Homepage