The OTM Model and Its Application at the Metro City Ring Project in Copenhagen
G Vuk, Danish Transport Research Institute, DK; C Overgård Hansen, Technical University of Denmark, DK; J Fox, RAND Europe, UK
The Danish Parliament is currently discussing a ? 2 billion project to complete the Copenhagen Metro City Ring. Part of the technical support to the politicians is being provided by running the OTM model, which has recently been updated.
The Metro City Ring project is the 4th phase of the expanding metro network in the Danish capital, and is expected to become operational in 2015. The metro?s phase 1 was opened in 2002, phase 2 in 2003, and the 3rd phase will be opened in autumn 2007.
The operational traffic model for the capital, the OTM model, was updated over the last two years specifically for the purpose of the Metro City Ring project. The efforts were concentrated on the data part of the model where new 2004 base matrices were built (this component of the project was presented at ETC 2006 by Professor Otto A. Nielsen). In addition, the demand model components have been re-estimated, combing values-of-time (VOT) derived from a large Stated Preference survey (based on the newly completed data for the Danish national VOT project) with Revealed Preference data collected across the Greater Copenhagen Area.
The aim of the paper is twofold. Firstly, we depict the structure of the new OTM demand model and secondly, we present and discuss the demand forecasts produced by the model for the Metro City Ring project.
2. The model structure
The OTM model system predicts demand for transport across the Greater Copenhagen Area. Forecasts of demand are made separately for seven passenger purposes: home-based work, home-based business, home-based education, home-based shopping, home-based leisure (all home-based travel that does not fall under the first four categories), non-home-based business and non-home-based other (all non-home-based travel that is not for business). In application, the predictions from the home-based business and non-home-based business models are aggregated into a single business purpose prior to assignment.
The OTM consists of separate model components by purpose for travel frequency (generation) at the all-day level, and models of simultaneous mode and destination choice. The modes in the model are car driver, car passenger, public transport, bike and walk.
Three tree structures were tested to determine the relative sensitivities of the mode and destination choice decisions: mode and destination choice in parallel, mode choice under destination choice and destination choice under mode choice.
A key feature of the model is that the VOT are income dependent, which allows the demand models in the model system to reflect variations in cost sensitivity with income.
The model base travel matrices reflect conditions on an average weekday in 2004. Once the model has been estimated synthetic 2004 travel matrices are produced. A pivoting procedure is used then to combine information from the base matrices with the changes in synthetic demand between the base and forecast years.
Prior to assignment, the model predictions are split between seven model time-periods distinguished in the assignment by using fixed factors.
3. The Metro City Ring project
When completed, the Metro City Ring alignment will be 15 km long including 17 underground stations. The total travel time, including time spent at stations, will be 23 min. Approximately 180.000 people and 190.000 working/education places will be on a walk distance to one of the Metro City Ring stations. In 2015, when it is expected that the new metro infrastructure is finished, some 85% of all dwellings and working/education places in the city will be within a 600 m radius of the metro and S-train network.
In the peak periods, the Metro City Ring is forecast to carry 5.500 passengers per hour per direction. With the 100 sec headway in the peak periods, the maximum capacity is 10.800 passengers per hour per direction, i.e. approximately the double of what is needed.
The previous version of the OTM was applied for forecasting traffic flows in 2015 where the following main results were obtained:
? there will be 280.000 metro city ring trips in an average weekday,
? the biggest part of the 280.000 metro trips origin from bus,
? 25% of the 280.000 metro trips (70.000 trips) are newly generated trips, and
? 10% of the 280.000 metro trips origin from car traffic.
Model forecasts with the new OTM will be carried out during spring 2007 where different alignment possibilities will be tested. Prior to model forecasts we will validate the model by comparing the calculated and observed traffic flows in the base 2004 year.
Association for European Transport