The use of response surfaces in specifying transport strategies

Integrated transport strategies can involve a wide range of measures, each varying in the intensity with which it is applied. Objective functions can be specified to reflect public policy objectives, and their value will change as changes are made in the measures included and their intensity, generating a response surface. This process has been used in earlier research to identify the optimum combination of measures for a given objective function. In this paper the data generated is used to explore the shape of the response surface around the optimum as individual policy measures are varied in intensity. Results are presented for three cities, using different models. They indicate particular sensitivity to changes in pricing and service frequency, and greater sensitivity for economic objective functions than those which include environmental attributes. Performance is particularly sensitive when public transport is deregulated. Changes in objective function also lead to marked changes in the optimum values for pricing measures. Road pricing performance is very sensitive to the definition of the charging method. Response surfaces have been shown to be particularly useful in indicating sensitivity to a given policy instrument; in determining the loss of benefit if a sub-optimal level is required; and in assessing robustness of strategies against changes in objectives.