High accident locations, Vol. I: the principal inter urban roads.
Hall, P.R. Hearne, R. & Holland, T.J.
Dublin, The National Institute for Physical Planning and Construction Research, 1970, 70 blz
|Samenvatting||The present report is part I of a two volume report. These two volumes comprise part of a comprehensive road safety project aimed at the elimination of accident black spots. The general and ultimate objectives of the project are described in detail in Volume I. These objectives may be summarized in three phases as follows: (1) The identification and designation of high accident locations; (2) The development of a system for forecasting reductions in accidents and accident severity attainable from various road improvement projects; and (3) The initiation and execution of cost effectiveness analyses of proposed improvement and modernization schemes. In volume I, a start was made on Phase (1) above. The distribution of accidents during 1968-69 on the top 4996 kilometers of the rural road network (the "principal inter-urban roads") was examined in detail. Reliable traffic flow data were available for these roads as a result of the systematic traffic counting program. Under this program, the roads were divided into traffic counting sections such that each section had reasonably uniform traffic flow throughout its length. The counting sections varied from 0.33 km to 25 km in length. In Volume I the rural counting sections were grouped into eighteen AADT groups. The average accident density per mile for these groups was worked for each of four accident classifications: (a) Fatal and personal injury accidents; (b) Non-intersection fatal and personal injury accidents; (c) All accidents; and (d) All non-intersection accidents. Regression analysis confirmed that there was a strong correlation between accident density and traffic volume for each of these classifications. A statistical method, based on the Poisson distribution, was used to compare the actual number of accidents for the two year period with the expected number for each counting section and to decide whether the difference between the actual and expected numbers was significant at the 95% level in respect of each of the four classifications. A counting section shown up by this method in respect of one or more of the four classifications was then designated as having "significantly high accident density" or "significantly low accident density". The use of this method resulted in the designation of 16.7% (769 km) of the network as having significantly high accident densities and of 15% of the network (678 km) as having significantly low accident densities. The advantages and disadvantages of the approach used were described. Since Volume I was produced the Road Inventory was completed for the National Route Network designated by the Minister for Local Government in August 1970. Each traffic counting section was divided into sub-sections which are physically homogeneous or have uniform geometric design characteristics. In addition a further year's accident data (1970) became available. These changes were used to analyze more closely accident distributions on the National Routes. A similar analysis to that described in Volume I was performed on the data. Volume II presents and discusses the results of the analysis and describes briefly the statistical methods used.|
|Full-text||Beschikbaar Niet beschikbaar, klik om contact op te nemen voor een digitalisatie verzoek|
|Suggestie?||Neem contact op met de SWOV bibliotheek voor uw opmerkingen|