Black spot management and safety analysis of road networks : best practice guidelines and implementation steps. Deliverable D6 of the RiPCORD-iSEREST project (Road Infrastructure Safety Protection - Core-Research and Development for Road Safety in Europe; Increasing safety and reliability of secondary roads for a sustainable Surface Transport).
20090183 ST [electronic version only]
Sørensen, M. & Elvik, R.
[Brussels, European Commission, Directorate-General for Transport and Energy (TREN)], 2008, 99 p., ref.
|Samenvatting||Black spot management (BSM) has a long tradition in traffic engineering in several countries in the European Union. In the last 5 to 10 years, more and more countries have supplemented this work with network safety management (NSM). However, the current approaches and quality of both BSM and NSM differ very much and the work can be characterised by a lack of standardised definitions and methods. Thus, the objective is to describe and develop state-of-the-art approaches and best practice guidelines for BSM and NSM and to describe the necessary implementation steps. State-of-the-art approaches are defined as the best currently available approaches from a theoretical point of view while best practice guidelines are the best approaches from a more practical point of view and can be used when the data and resources for developing, implementing and using a national method are limited. No standard definition exists of either black spots or hazardous road sections. However, from a theoretical point of view black spots and hazardous road sections should be defined as any location that has a higher expected number of accidents than other similar locations as a result of local risk factors. Black spots should be identified by reference to a clearly defined population of roadway elements as for example curves, bridges or four-leg junctions, while hazardous road sections should be identified by reference to 2-10 kilometres homogeneous road sections. This makes it possible to estimate the general expected number of accidents by use of an accident model. The identification of hazardous locations should rely on a more or less advanced model based method, ideally speaking the empirical Bayes method. The argument for that is that model based methods are the best to make a reliable identification of sites with local risk factors related to road design and traffic control, because systematic variation and partially random fluctuation are taken into consideration. To make the road division and develop the accident model it is necessary to have data about accidents, traffic volume and road design. These data have to be unambiguously located and be immediately interoperable with each other. The state-of-the-art approach for accident analysis consists of two stages. The first stage is, by means of detailed examination of accidents, to suggest hypotheses regarding risk factors that may have contributed to the accidents. The second stage is to test the hypotheses. This can be done by a double blind comparison of each black spot or hazardous location and a safe location. According to the best practice guidelines, the analysis stage should as a minimum consists of a general accident analysis, a collision diagram, a road inspection and relevant traffic and road analyses. In NSM results from the general accident analysis and the collision diagram should be combined into an extended collision diagram. Evaluation of the effects of the treatment should employ the empirical Bayes before-and-after-design, because it controls for local changes in traffic volume, long term trends in accidents and regression-to-the-mean. When it is not possible, the evaluation should be made as a simpler before-after-study controlling for long-term trends in the number of accidents, local changes in traffic volume and regression-to-the-mean by use of correction factors. (Author/publisher)|
|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|