Based on 30 years of transport modelling, my aim here is to provide information which will be useful to others in developing transportation models, and this guidance focuses on strategic multimodal city transport models (often referred to as four step models). For the theory, you can refer to the many conference papers, reports and excellent books written by the leading academics and practitioners. But there is a gap in this material, and that is a set of 'how to do it' templates.
Topics covered in the guidance or checklists include: household travel surveys, rail passenger and car intercept OD surveys, travel survey data processing, road and public transport networks and assignment, land use planning data, the zone system, the estimation or calibration of car ownership, trip end (trip production and attraction), distribution and mode choice models. Commercial vehicle models, peak-spreading, and model testing and validation are also addressed.
I have written a text on transportation modelling which appears in the 2004 edition of "Traffic Engineering and Management", published by Monash University, Melbourne. This attempts to describe the wide range of mathematical models being used in transport planning. The page budget was too small to include this layman's introduction to transport modelling.
Prior to carrying out the technical model development tasks, there are preliminary tasks concerned with designing the transport model specifications, choosing a travel, person and household segmentation, approaches to data collection and travel surveys and design of the model estimation process.
While much of this is about city aggregate multimodal (four step) transport models, much of the transport modelling guidance and checklists is applicable in a wider range of transportation modelling contexts. The purpose of the guidance is to identify all of the transport model development tasks which have to be done on any transport model project. For reference, see the site map.
Some appreciation of the accuracy of travel demand estimates (eg trip ends, trip matrices) derived from the sample surveys used in transport models is necessary if sensible decisions are to be made about the design of surveys, the accuracy of model estimation and the accuracy of the travel demand outputs of transport models.
All too often, the reports on transport model development and on their application to transport project forecasting make inadequate allowance for the associated uncertainties, one significant component being related to the implications of sample surveys. This is discussed in simple terms here and a spreadsheet implementing these thoughts is here.
Transport model forecasts of 20-50 years into the future are self-evidently subject to uncertainties and it is common to analyse these uncertainties. But the task of analysing the uncertainty in transport models is very demanding.
In many circumstances pivot-point, marginal and incremental transport models offer considerable advantages.
Copies of selected conference papers on transport modelling.
Some links to important transport modelling guidance and conference paper archives.
My transport modelling experience is summarised here.
Much of the information made available here was derived from a two and half year project to develop the Wellington Transport Strategy Model. The New Zealand client, Greater Wellington Regional Council, has not only given me permission to publish this material but has actively encouraged it, for which I am exceedingly grateful. The work was done jointly by Sinclair Knight Merz and Beca Carter Hollings and Ferner, under Daniel Brown's management.
Additional data is based on the three year project to update the Auckland Regional Transport Model, and Auckland Regional Council has given permission for me to use the material developed on that project. The project was done jointly by Sinclair Knight Merz and Beca Carter, and technically managed by David Young.
You are free to use and quote from this material but, if you do, please acknowledge this website.
The website was originally launched in 2003/4 and most of the documents were written at that time. Since then I have added more material and will continue to do this during 2014. See the update history.
This page has been accessed times since July 2009.