Minnerva has supported SDG in modelling public transport for schemes to introduce rapid transit type systems into the cities of Leeds and Dublin.

The work for Dublin required particular attention to making the best of the available data that came from varied sources and which, inevitably, displayed various inconsistencies and limitations.

Much of the work was concerned with estimating trip matrices through combining information from bus surveys, bus electronic-ticketing machine data, and from rail-based PT matrices.

The project was also concerned to produce a combined public transport matrix, rather than the separate bus and rail matrices that had been used hitherto, but in order to model the interaction between the modes it was necessary to analyse and enhance aspects of the public transport network modelling.

An important requirement of the public transport demand modelling was to segment the demand on the basis of valuations of time by travellers. This was because Stated Preference surveys had shown some travellers had unusually low valuations of time. Once the demand had been segmented, it was necessary to adapt the modelling to cater for multi-class public transport assignment.

The work included using TRIPS-based matrix estimation methods to integrate the data sources, as well as developing GIS-based functionality to segment trip matrices according to different valuations of time associated with different population characteristics.