The report details the safety benefits achieved by the installation of median barriers on the New Zealand motorway system. Before and after analysis, using log-odds, regression and multiplicative models, was applied to determine the safety benefits.
Each of these models produced evidence of similar reductions in fatal accidents of around 75%. The estimated reduction in non-fatal accidents was inconclusive, with analysis showing the possibility of either a major increase or minor decrease in the number of non-fatal accidents.
Benefit/cost calculations were performed by applying both the 1986 and the 1991 methods for evaluating the benefit/cost of roading projects used in New Zealand. Median barrier installation on average had a benefit/cost ratio between 12 and 16. Because of changing accident patterns the recommendation is that, rather than developing a Median Barrier Installation Warrant based on previous experience for New Zealand motorways, each median barrier installation proposal should be judged on its own merits.
Keywords: accidents, before & after analysis, benefits, evaluation, log-odds model, median barriers, modelling, motorways, multiplicative model, regression model, roads, safety