Descriptive models of parking to complement transport planning studies

Abstract

The rise in the importance of traffic restraint as an instrument of transport policy has focused interest on the policies of provision and pricing of parking spaces. The network models found appropriate to assess parking policy as an instrument of transport planning and restraint are formulated in terms of flows and averages over time. The relationships between such averaged occupancy and price level simplications of parking policies and the actual spaces and charging policies in effect are very poorly defined. Two models of the occupancy and revenue collection from parking spaces are developed, both of which provide links between the use of parking areas by users with various lengths of stay and the averaged demand and charging levels required for transport network flow models which have no provision for handling such interacting time dependent descriptions of parking behaviour. One is a Markov model of the behaviour of a representative parking space over time, and the other is a direct probabilistic dynamic model of a parking area with specified capacity: the latter model is extended to include price elastic demand, and to platoon arrival patterns, which is used to reduce the cost of running such models at the price of some loss of detail and precision. Examples based on observed data are given for each of these variations.