ALLOWING FOR VARIABLE DEMAND IN HIGHWAY SCHEME ASSESSMENT

Following the 1994 SACTRA Report and the Government's response, the assessment of highway schemes has had to make allowance for "induced travel". However, not only does the concept of induced travel itself raise some problems of definition, it also presents potentially severe technical difficulties. The project from which the material in this paper is drawn aims to advance the treatment of induced travel, or - more generally - variable demand, both at a practical and at a theoretical level. There are three more or less independent components of the process: (1) defining the impact of changes in (generalised) cost on demand; (2) defining the impact of external factors (changes in population, employment, car ownership) on demand; and (3) estimating a new equilibrium following a change in supply (e.g. new highway scheme). Each of these components presents substantial theoretical and practical difficulties. Whatever approach is taken to the modelling of demand, calculating a stable equilibrium is essential if economic appraisal is to be credible. There is a major gap here between theory and practice: the use of "equilibrium"-based congested assignment with fixed demand is becoming more widespread, but there is rather little experience of combining such assignment with more than the simplest form of variable demand functions (generically referred to as "simple elasticity models"). The present research shows that there are no theoretical obstacles to extending the demand functions to any form of (hierarchical) logit model. This opens the way to a much more appealing treatment of demand responses, including destination choice, mode choice and trip frequency. It also allows for improved algorithms for computing the equilibrium between demand and supply. The paper describes the completed theoretical work, and the programme of tests which are being carried out to investigate the practicality of the approach, including questions of computational performance, compared with more heuristic methods. For the covering abstract see ITRD E105584.

Language

  • English

Media Info

  • Features: References;
  • Pagination: p. 197-208
  • Serial:
    • Volume: P434

Subject/Index Terms

Filing Info

  • Accession Number: 00796497
  • Record Type: Publication
  • Source Agency: Transport Research Laboratory
  • ISBN: 0-86050-325-9
  • Files: ITRD
  • Created Date: Aug 2 2000 12:00AM