<Minnerva>

The Dublin Transportation Office (DTO) have been responsible for modelling transport in the Greater Dublin area since 1996 and since that time they have developed an extensive multi-modal model for the region. Since then, the model has undergone several revisions as the area of coverage has expanded, as more data has come available and as the transport infrastructure has evolved. Each revision has provided the opportunity to review and develop in more detail components of the model’s structure. 2005 sees such a model update, and as part of this process a revised Trip End and Trip Distribution Model has being developed by Minnerva, and OmniTRANS is being used to deliver the models to the DTO.

The Trip Generation model is driven by a Car Ownership/Car Availability model and both the Generation and Attraction components of the model draw heavily on new 2002 Census data as well as a specially conducted Home to School survey. Overall, the new Trip End Model is structured by Person Type, Person Mobility and Area Type. Walk and cycle trips are included in the calculation of trip rates and this represents an extension to the current model structure.

Trip Distribution is based on a Destination Choice model.This is an extension of the well understood Gravity model structure with the exception that unexplained ‘noise’ in the model, usually dealt with by ‘K factors’ is treated in a more refined manner. This is done by introducing explanatory variable(s) other than generalised cost into the model to help understand why people choose one destination ‘(identical’ in cost terms) in preference to another. To calibrate this more complex model structure, calibration software developed by Dr George Skrobanski (Numeral Software) has been used; these have been developed in Ruby as OmniTRANS compatible classes.

The new models are implemented in OmniTRANS and delivered as a ‘user friendly’ package to the client so that they can develop and run the model themselves as and when new data is available for different future planning horizons.

DTO Head of Modelling Frank McCabe said "We felt that the way in which Minnerva approached this project, and the use of OmniTRANS, has been very innovative. We see that OmniTRANS will help us in visualising and understanding our data and look forward to using the model".


(Contact: Martin Bach)