A metamodelling based approach for estimating travel distances for distributed public services, with particular reference to formula funding applications

Adam Hindle (2005), PhD Thesis, University of Lancaster.

This research study has been concerned with developing effective methods for estimating travel effects for distributed public services in areas having differing geographies and differing geodemographic properties. The main intended applications context has been situations where strategic decisions are needed concerning the allocation of resources and funding to equitably compensate areas having differing geographies for their expected travel-related costs.

Specifically, the aims of this research have been to develop suitable models which can provide travel distance estimates (and related travel time and travel costs) associated with providing distributed services within different geographical regions, subject to population distributions and expected demand for services. To achieve these aims a number of models have been developed using a metamodelling approach based on a large number of simulations of simplified, synthetic geographies, designed to investigate how the spatial characteristics of an environment affect travel distances, based on Travelling Salesman-type routes.

The research also demonstrates how the models can be applied to District Nursing services for a sample of regions in the UK. The approach provides an effective method to translate census and GIS information, via the suite of metamodels, into the type of distance-based outputs which are valued by decision makers at the strategic level interested in understanding the effects of rurality on service costs.

Funded by ESRC.


Hindle AJ and Worthington D, 2004, ‘Models to estimate average route lengths in different geographical environments’,Journal of Operational Research Society, 2004, Vol 55, pp662-666.