Thesis (M.Ing.)--Stellenbosch University, 2001. / ENGLISH ABSTRACT: A road safety remedial programme has as an objective the improvement of road
transportation safety by applying road safety engineering remedial measures to
hazardous road network elements in a manner that will be economically
efficient.
Since accident data is the primary manifestation of poor safety levels it must be
analysed in manner that will support the overall objective of economic efficiency.
Three steps in the process of implementing a road safety remedial programme,
that rely on the systematic analysis of accident data, are the identification of
hazardous locations, the ranking of hazardous locations and the evaluation of
remedial measure effectiveness.
The efficiency of a road safety remedial programme can be enhanced by using
appropriate methodologies to measure safety, identify and rank hazardous
locations and to determine the effectiveness of road safety remedial measures.
There are a number of methodologies available to perform these tasks, although
some perform much better than other. Methodologies based on the Empirical
Bayesian approach generally provide better results than the Conventional
methods. Bayesian methodologies are not often used in South Africa. To do so
would require the additional training of students and engineering professionals
as well as more research by tertiary and other research institutions.
The efficiency of a road safety remedial programme can be compromised by
using poor quality accident data. In South Africa the quality of accident data is
generally poor and should more attention be given to the proper management
and control of accident data.
This thesis will report on, investigate and evaluate Bayesian and Conventional
accident data analysis methodologies. / AFRIKAANSE OPSOMMING: Die doel van 'n padveiligheidsverbeteringsprogram is om op die mees koste
effektiewe manier die veiligheid van onveilige padnetwerkelemente te verbeter
deur die toepassing van ingenieursmaatreëls.
Aangesien padveiligheid direk verband hou met verkeersongelukke vereis die
koste effektiewe implementering van 'n padveiligheidsverbeteringsprogram die
doelgerigte en korrekte ontleding van ongeluksdata.
Om 'n padveiligheidsverbeteringsprogram te implementeer word die ontleding
van ongeluksdata verlang vir die identifisering en priortisering van gevaarkolle,
sowel as om die effektiwiteit van verbeteringsmaatreëls te bepaal.
Die koste effektiwiteit van 'n padveiligheidsverbeteringsprogram kan verbeter
word deur die regte metodes te kies om padveiligheid te meet, gevaarkolle te
identifiseer en te prioritiseer en om die effektiwiteit van verbeteringsmaatreëls te
bepaal. Daar is verskeie metodes om hierdie ontledings te doen, alhoewel
sommige van die metodes beter is as ander. Die 'Bayesian' metodes lewer oor
die algemeen beter resultate as die gewone konvensionele metodes. 'Bayesian'
metodes word nie. in Suid Afrika toegepas nie. Om dit te doen sal addisionele
opleiding van studente en ingenieurs vereis, sowel as addisionele navorsing
deur universiteite en ander navorsing instansies.
Die gebruik van swak kwaliteit ongeluksdata kan die integriteit van 'n
padveiligheidsverbeteringsprogram benadeel. Die kwaliteit van ongeluksdata in
Suid Afrika is oor die algemeen swak en behoort meer aandag gegee te word
aan die bestuur en kontrole van ongeluksdata.
Die doel van hierdie tesis is om verslag te doen oor 'Bayesian' en konvensionele
metodes wat gebruik kan word om ongeluksdata te ontleed, dit te ondersoek en
te evalueer.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:sun/oai:scholar.sun.ac.za:10019.1/52483 |
| Date | 03 1900 |
| Creators | Mollet, C. J. |
| Contributors | Bester, C. J., Stellenbosch University. Faculty of Engineering. Dept. of Civil Engineering. |
| Publisher | Stellenbosch : Stellenbosch University |
| Source Sets | South African National ETD Portal |
| Language | en_ZA |
| Detected Language | English |
| Type | Thesis |
| Format | 1 v. (various pagings) : ill. |
| Rights | Stellenbosch University |
Page generated in 0.0019 seconds