Highlights ► The aim of the research is to specify and estimate a parking demand model. ► Two models will be estimated: a global regression and a local regression. ► The global model will be based on the operational data of the parking zones. ► The detection of spatial correlation suggested that a local model should be estimated. ► The local model demonstrated it had better fit and good predictive capacities.
Abstract Multiple linear regression (MLR) and geographically weighted regression (GWR) models are used for estimating parking demand in areas with paid short stay parking systems. These models have been applied to the city of Santander (Cantabria, Spain) to check their goodness of fit and their predictive ability. The results show the main advantages and disadvantages of using GWR models. The technique proved to be useful in this case study because it offered a better fit and made better predictions in a scenario showing a certain degree of spatial heterogeneity unexplained by any of the variables introduced into the global model. However, the GWR model also presented situations of local correlation although this was considered moderate given the results provided by the variance inflation factors and the local condition indexes.
Modelling demand in restricted parking zones
Transportation Research Part A: Policy and Practice ; 45 , 6 ; 485-498
2011-03-29
14 pages
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
Modelling demand in restricted parking zones
Online Contents | 2011
|Modelling demand under parking and cordon pricing policy
Elsevier | 2012
|Modelling demand under parking and cordon pricing policy
Online Contents | 2013
|Europäisches Patentamt | 2020
|