1887

oa Civil Engineer in South Africa - A new approach to car ownership forecasting in South Africa

 

Abstract

The paper covers a review of car ownership forecasting in the South African context, the introduction of the power growth forecasting model, a South African car ownership trend analysis and a national and regional car ownership forecast.


Forecasting applications are based upon extrapolatory or causal techniques, with the latter offering advantages in respect of aiding the policy-making process. South African applications are characterized by the extrapolatory or trend methodology, being based primarily upon simple extrapolatory methods and the use of simple logistic functions.
The power growth model offers particular advantages when compared with the application of the logistic model in car ownership forecasting. In particular the apparent / true saturation level issue is resolved by the power growth curve accounting for curve convexity, the saturation level is determined by model calibration of time-series data rather than by exogenous considerations, and model flexibility is obtained through manipulation of up to three constants.
Data availability pertaining to car ownership analysis in the South African context is summarized and the various problems associated with this such as availability, consistency, duplications, ownership distributions and categorizations are highlighted. The analytical terms 'total car population', 'total car population density', 'household car population' and 'household car population density' are introduced. Car ownership trends based on these analytical terms in respect of South Africa and the PWV region are investigated in the light of their forecasting implications.

Die verhandeling dek 'n oorsig van motoreienaarskapvoorspellings in die Suid-Afrikaanse konteks, die bekendstelling van die model vir mag-groeivoorspelling, 'n Suid-Afrikaanse motoreienaarskaptendensontleding en 'n nasionale en streeksmotoreienaarskapvoorspelling.


Die aanwending van voorspellings is gebaseer op ekstrapolerende of oorsaakliketegnieke. Laasgenoemde bied voordele wat bydra tot die beleidmakingsproses. Die Suid-Afrikaanse aanwending word daardeur gekenmerk dat die ekstrapolerende of tendensmetodologie hoofsaaklik gebaseer word op eenvoudige ekstrapolerende metodes en die gebruik van eenvoudige logistiese funksies.
Die mag-groeimodel bied besondere voordele wanneer dit vergelyk word met die aanwending van die logistiese model in motoreienaarskapvoorspelling. In besonder word die kwessie van die skynbare werklike versadigingsvlak opgelos deurdat die mag-groeikurwe kurwekonveksiteit verantwoord, die versadigingsvlak bepaal word deur modelkalibrering van tydreeksdata eerder as deur eksogene oorwegings, en modelbuigbaarheid verkry word deur manipulasie van tot drie konstantes.
Die beskikbaarheid van data met betrekking tot motoreienaarskapontleding in Suid-Afrikaanse verband word opgesom en verskeie verwante probleme soos beskikbaarheid, konsekwentheid, dupliserings, eienaarskapverspreiding en kategoriserings word uitgelig. Die ontledende terme 'totale motorbevolking', 'totale motorbevolkingsdigtheid', 'huishoudelike motorbevolking' en 'huishoudelike motorbevolkingsdigtheid' word bekendgestel. Motoreienaarskaptendense, gebaseer op hierdie ontledende terme en rakende Suid-Afrika en die PWV-gebied wordondersoek in die lig van hul voorspellingsimplikasies.

Loading

Article metrics loading...

/content/civeng1/23/12/EJC24838
1981-12-01
2016-12-04
This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error