Design and Use of the New Greater Manchester Land-use/transport Interaction Model (GMSPM2)

Design and Use of the New Greater Manchester Land-use/transport Interaction Model (GMSPM2)


A Dobson, E Richmond, D Simmonds, David Simmonds Consulting, UK; I Palmer, GMPTE, UK; N Benbow, MVA, UK


Describes key characteristics of GMSPM2, the methods used to constrain its reference case forecasts to previous assumptions, and discusses the use of the model to produce land-use and transport forecasts "pivoting" around those assumptions.


GMSPM2 is a large-scale land-use transport interaction (LUTI) model commissioned by the Greater Manchester Passenger Transport Executive from MVA Consultancy and David Simmonds Consultancy (DSC) in 2006. It replaces the original GMSPM, which was developed some ten years ago (and presented to ETC in 2000).

Compared to GMSPM, GMSPM2 has a more detailed zone system within Greater Manchester, much more extensive coverage of the areas around Greater Manchester, more sophisticated modelling of the economy and of interactions with surrounding areas, and a more elaborate treatment of the transport networks and of responses to transport changes. GMSPM2 has been implemented using MVA?s TRAM transport modelling package integrated with DSC?s DELTA land use/economic modelling package.

In this paper we present an overview of the land use model describing the way in which it provides inputs to the transport model. We then discuss the approach taken to constraining the model to groups of zones. This enhancement was commissioned to ensure that the Reference Case would match previous forecasts of the levels of population and employment within different parts of the Greater Manchester Area. The operation of the enhancements is carried out in the two stages. The first is the generation of a shadow utility function. This involved using the constraints on population and employment as targets. Secondly, with the constraints removed, applying the shadow utilities as attractors. This allowed the model to be run, in order to test different scenarios, with each of the new scenarios ?pivoting? around the constrained reference case.

This technique is then used to explore the impacts of varying fuel prices and their impact upon population and employment distributions across Greater Manchester.

Finally we conclude by reviewing the strengths and weaknesses of this approach of introducing constraints within the land use model.


Association for European Transport