1. Topic

  Strategic multi-modal transport models

2. Introduction

   

The urban transportation system is very complex, and its performance depends on decisions made on many levels of society whose goals and purposes may be in conflict with each other. The process of evaluating, designing and managing such a system can therefore not be carried out without the aid of properly formulated models. In fact, over the past three decades, there has been growing interest in the ability of computer applications and simulation models for all aspects of transport operations, planning and management.

The choice of the Computer Simulation Systems (CSS) to be used in transport studies is governed by the objectives of the analysis as well as the available resources. A common classification method for CSS is based on the detail level with which the incorporated models suite intends to simulate the components of the transport system. According to this, CSS can be conveniently classified into four categories that reconcile the differences between alternative modelling concepts and theories, as well as between different levels of investigation in traffic and transport studies. Starting at the most detailed (micro-) level, we have:

· Operational micro-simulation models that consider the characteristics of each individual vehicle and its interactions with other vehicles in the traffic stream;

· Tactical network models that are suitable for dynamic traffic effects analysis which are critical in network simulation during medium to congested flow conditions;

· Strategic Multi-modal transport models that are best suited to the urban-scale analysis of travel demand and transportation network performance;

· Land-use/transport interaction models that synthesise the dynamic interaction between transport provision and land-use activities.

3. Discussion

   

Strategic Multimodal transport model packages are used to predict the number of trips within an urban area by type (work, non-work, etc.), time of the day (peak-period, daily, etc.) and zonal origin destination (O-D) pair, the mode of travel used to make these trips and the routes taken through the transportation network by these trips. The final output is a predicted set of modal flows on link in the network and the related level of service. The incorporated models are used in a sequence of steps corresponding to the sequential decision process in which people decide to make a trip (generation), decide where to go (distribution), decide what mode to take (modal split) and decide what route to use (assignment).

These models rely on the assumption of a steady state within the modelled period and treat vehicular traffic as aggregate flows. Therefore, as compared to the tactical network models, they provide a less detailed treatment of the time dynamics of congestion. However, they possess the ability to represent the range of travel choices and traffic interactions in space on a realistic network, and they provide coverage of much larger geographical areas within practical constraints of computational capacity compared to detailed simulation models. Variability of road travel demand and of road network travel conditions may be catered for to some extent by the subdivision of demand into a series of time slices to be applied within a sequential steady state model runs, with the initial travel conditions for any time slice being based on information from the previous period.

In an urban context, strategic models generally cover a whole city or town. They generally include capacity restraint procedures whereby the speed on each road link will vary in relation to the volume of traffic which it carries. Strategic models are used to evaluate schemes and policies which are expected to have large scale effects over a considerable geographical area. They tend to be used for long term forecasting, perhaps as far as 20 or 30 years ahead.

Strategic transport models packages incorporate four main sub-models: 1) a trip generation sub-model which estimates the number of trips that originate or terminate in each zone using land-use and socio-economic data; 2) a trip distribution sub-model which estimates where the trips from a particular origin zone are going (i.e. which destination zone); 3) a modal split sub-model which estimates the proportion of trips between each zone pair that are made by each mode of transport 4) an assignment sub-model which allocates trips to particular routes through the transportation system.


4. Recommendation / Conclusion

   

Strategic Multimodal transport models packages can be used for a variety of purposes: 1) to forecast the overall demand for travel at some specified date in the future, given predicted changes in factors external to the transport system, such as population, employment and household income; 2) to allocate forecast demands for travel to the various modes of transport and, within each mode, to individual roads and public transport services; 3) to calculate the levels of service offered by each mode; 4) to provide information on vehicle and passenger flows and travel costs necessary for operational, environmental, economic and financial appraisals.

On the basis of the purpose above, the strategic models may be synthetically aggregated in the following main functional categories:

· Data processing;

· Network assignment;

· Project management;

· Site analysis.

Typical application phases for these models are the investigation on transport demand and the operation of multimodal transport systems within urban or regional transportation planning and feasibility studies of new infrastructure, network upgrading, global operational measures application.

Typical outputs are graphic and numerical network and flows representation, flows assignment, isochronic lines calculation and mapping, emission calculation.

The disadvantage of strategic models for emissions modelling is that there is often a need to compromise in term of spatial details. Zones are generally quite large and networks relatively coarse. Also, vehicle emissions output can normally only be calculated from average link speeds, whilst actual vehicle emissions depend on actual driving patterns (e.g. acceleration/deceleration, stops/starts). As a consequence they are not suited to evaluating the effects of traffic management measures and small scale schemes, and all the measures triggering changes in vehicles speed cycle.


5. Examples / Further Reading

   

Examples of the use of strategic transport models are given here by:


6. Additional Documents / Web Links

   

Two transport models (EMME2 and Saturn) are currently being applied in Bristol primarily to assess the road user charging and the light rapid transit. EMME2 software is applied to assess modal split effects whereas Saturn is applied to simulate traffic on the road network.

An extended list of strategic multi-modal transport models within this category includes:

  • Advanced General Network Editor
  • ASSIGN
  • Better Decision
  • CANDLINK
  • CONTRAM
  • Decision Support System
  • emme/2
  • Highway Design and Maintenance
  • Hyperplan
  • Intersection Analysis Spreadsheets
  • L-TASS
  • MicroBENCOST
  • MICROTRIMS
  • Municipal Equipment Management System
  • MVMACH
  • Planning and Project Development Spreadsheets
  • Project Analysis Package
  • Program Development and Management System
  • QRS
  • Roadway/Intersection Air Quality
  • SATURN
  • SITE
  • SPARKS
  • Survey
  • TEAPAC
  • TMODEL
  • Traffic Interpolater & Extrapolater Software
  • TRAFFIX
  • TrafikPlan
  • TRANSCAD
  • Transportation Network Analysis System
  • Transportation Planning Conference Proceedings
  • TRIP GENERATION
  • TRIPS
  • TSM
  • URPDB
  • VISUM
  • WinTASS
  • ZDATA

Details on these models can be searched on the Internet.

Other related references:

· University of Florida, Transportation Research Center, McTrans. http://www-mctrans.ce.ufl.edu/

· PTV AG. http://www.ptv.de/

Last Updated


 

25th January 2005

Back