Principles of transport analysis and forecasting.

Author(s)
Bonsall, P.W.
Year
Abstract

Models are representations of reality which can be used to explore the results of specific strategies or policies; they are deliberately simplified to keep them manageable. They can provide estimates of likely policy outcomes more quickly and at less cost than would be possible through implementation and monitoring. They have various applications. The desirable properties of a good model include accuracy and precision. 'Specification' is the process of deciding what 'rules' to include in a model. 'Calibration' is the process of quantifying a rule so that it can reproduce observed behaviour. 'Validation' is the process of checking a calibrated rule's more general validity. Most widely used transport models include the concepts of maximum utility, generalised cost, equilibrium and aggregation; they also ignore 'irrelevant' dimensions of response. When selecting a model from the wide range of available possibilities, a researcher should specify the relevant requirements and set constraints of time and money. Classes of model available to transport analysts include: (1) simple formulae; (2) time series models; (3) forecasting without a trend; (4) regression analysis; (5) matrix estimation models; (6) elasticity models; (7) modelling choices between alternatives; and (8) simulation models. For the covering abstract, see IRRD 892228.

Request publication

14 + 1 =
Solve this simple math problem and enter the result. E.g. for 1+3, enter 4.

Publication

Library number
C 40758 (In: C 40753) /10 /72 / IRRD 892233
Source

In: Transport planning and traffic engineering, edited by C.A. O'Flaherty, London, Arnold, 2003, ISBN 0-340-66279-4, 4th edition, p. 103-131, 12 ref.

Our collection

This publication is one of our other publications, and part of our extensive collection of road safety literature, that also includes the SWOV publications.