Modelling Land Use and Transport Interaction in London: Adopting the Right Data Sources and Methodology
M Swiderski, SKM Colin Buchanan, UK
This abstract describes the methodologies adopted for four separate land-use and transport interaction modelling studies in London. It also draws out the policy implications of the studies and makes recommendations based on lessons learned.
When modelling the trips from a new development as part of a scenario to be compared against a reference case which does not contain the development, the main difficulty is often to be able to reconcile different data sources used in the modelling in order to represent those trips correctly. SKM in partnership with their clients, has recent experience of addressing this problem in the context of modelling the impact of major development areas in London on the transport network.
Most transport studies in London start with overall assumptions from the Mayor's Transport Strategy as implemented in the London Transportation Study (LTS) model as a baseline, or reference case forecast. Different methodologies for representing new trips at specific locations which are explored in this paper are as follows;
Trips from the new developments are only accounted for at the strategic level within the London LTS trip matrix, but are not allocated to the specific locations identified in the local land use review. New development trips added to the LTS matrix thus requires re-balancing of LTS trip levels in the rest of the matrix. This method was deployed in modelling the transport impacts of the Vauxhall Nine Elms and Battersea Opportunity Area (VNEB) and brought about a recommendation for extending the Northern Line from Kennington. In some scenarios, this method required trip levels being re-balanced in more than the Boroughs containing the immediate study area.
Trips from the new developments are accounted for at the Borough level within the London Transportation Study (LTS) trip matrix. A local data source, such as the East London-based Land Use Trip End Database (LUTE) is used to provide the distribution of new development trips within the Boroughs within the study area whilst holding to LTS Borough trip totals and assuming baseline LTS trip totals (to which the new development trips are added). This method was deployed in modelling the transport impacts of land-use developments in the A12 highway corridor and Olympic Legacy developments; in both cases it was recommended that the transport capacity assumed within the reference case scenario would be sufficient to cater for demand. In contrast to some scenarios in the VNEB study, this method led to the re-balancing of trip levels only in the Boroughs directly affected by the study.
Trips from all developments (both existing and future) are accounted for entirely by LUTE within the modelling process. LTS is used to provide the distribution of those trips between model zones. This method has been deployed in modelling the impacts of land use developments in East London on the Docklands Light Railway (DLR). The modelling has provided a case for major extensions of the DLR network. Here, in contrast to the other methods, there is no balancing of trips applied because it is considered that LUTE provides the most accurate forecast of total trip levels within the study area.
LTS is the main strategic-level source of trip data used for transport modelling projects in London. LUTE operates both as a database of development information and as a trip generation forecasting tool. The main benefit of LUTE is that it uses a "bottom-up" approach based on known development information, rather than the "top-down" approach adopted in LTS, whereby strategic forecasts are distributed in a less detailed manner between model zones. LUTE is being increasingly adopted for transport planning projects in East London.
Lessons learned from the modelling are;
A methodology should be adopted which is suited to the circumstances of the specific study being undertaken. There is no correct methodology which should be adopted across the board for all studies. For instance, in the case of some scenarios in the VNEB study, it was considered that any development taking place in the opportunity area would be restricted to London Plan totals, hence a wider balancing approach was appropriate.
It is preferable to use detailed development information such as that found in LUTE where that information provides a reliable and comprehensive representation of development. Where gaps exist, other data sources such as LTS should be used to verify and potentially adjust the detailed development forecasts. A comprehensive audit of the use of LUTE data and the potential use of other data sources for DLR modelling is taking place during 2011; it is intended that this paper will report on the findings of this audit if available.
Association for European Transport