Abstract
Four problems are posed: how to develop demand models that can cope with more than two modes of transport, how to estimate demand for such a mode of transport, how to make existing models yielding more pertinent information, and to what extent to include such factors as reliability, safety, comfort and convenience in demand models. The discussions are grouped under the headings: results of recent analysis, operator requirements for demand analysis, general applicability of demand estimation procedures, value of time diversion models, and directions for future research.