Traffic crashes have a major impact to European society, in 2008 over 38,000 road users died and over 1.2 million were injured. The economic cost is immense and has been estimated at over 160 billion for the EU 15 alone. The European Commission and National Governments place a high priority on reducing casualty numbers and have a series introduced targets and objectives. The experience of the best-performing countries is that the most effective policies are based on an evidence-based, scientific approach. Information about the magnitude, nature and context of the crashes is essential while detailed analyses of the role of infrastructure, vehicles and road users enables new policies to be developed. The EU funded SafetyNet project established the European Road Safety Observatory to bring together data and knowledge to support safety policy-making. The project developed the framework of the Observatory and the protocols for the data and knowledge, the ERSO is now a part of the DG-Move website: http://ec.europa.eu/transport/road_safety/specialist/index_en.htm. The DaCoTA project will add to the strength and wealth of information in the Observatory by enhancing the existing data and adding new road safety information. The main areas of work include • Work package 1 - Policy-making and Safety Management Processes • Developing the link between the evidence base and new road safety policies • Work package 2 — In-depth Accident Investigations • Setting up a Pan-European Accident Investigation Network • Work package 3 — Data Warehouse • Bringing a wide variety of data together for users to manipulate • Work package 4 — Decision Support • Presenting analysis results and data to policy makers • Work package 5 — Safety and eSafety • Intelligent safety system evaluation • Work package 6 - Naturalistic driving observations This deliverable is a production of Work package 4. The aim of WP4 is to bridge the gap between research and policy to enable knowledge-based road safety management. To support road safety decision makers, this Work Package will: (1) exploit the data available for analysis by providing forecasts of the road safety situation in the different member states and, possibly, the whole of Europe; and (2) work on the development of ready-to-use instruments. Tools that were well-appreciated in the past will be standardised and complemented by new tools. This will be done in close communication with the end-users themselves. The end-users mainly concern the policy makers, but may in some case also concern power-users from research and the industry. The expected outcomes of WP 4 are • National forecasts: To enable target setting and monitoring of the road safety progress in the different countries, forecasting models will be implemented. • European forecasts: To identify common trends in different European countries, the accident outcomes will be analysed jointly. • Web texts: Web texts are already provided on the ERSO website that give compact, impartial information on important road safety issues. These are updated and web texts on complementing issues will be added. • Browser tool for data warehouse: A browser tool will allow easy access to information stored in the Data Warehouse that will be developed in Work Package 3. • Country overviews: These will give an overview of the road safety situation in each country. The overviews will address final road safety outcomes, performance indicators, policy performance and background characteristics of the countries. • Country indices: To further this information even more, possibilities are investigated to summarize the information contained in the country overviews into one or a few country road safety indices. Objectives of present deliverable: Roads and road transport play a central role in Western societies, but the benefits they offer come at a cost. In addition to the obvious costs of building roads and vehicles and providing fuel, there are various less obvious costs: human and environmental. We focus here on road accidents and in particular on the resulting fatalities, which are the unintended consequences of the road transport system. The frequency of accidents and the number of fatalities evolve over time. In fact the number of fatalities has decreased in most European countries in recent years. It is important to monitor these developments, focusing on a number of key questions • Has there been a continuous, smooth development or were there abrupt changes? • If there have been changes, are they to be attributed to changes in the actual risk of having (fatal) accidents, or rather to changes in traffic volume? • Where does the present development get us (if continued)? The last issue is particularly important for the setting of political road safety targets. It has been shown that in countries that have an explicit target to be reached by a particular year - for instance the reduction of the number of fatalities - more concrete actions to improve road safety have been taken (Wegman et al., 2005). Such a target has to be SMART: specific, measurable, attainable, realistic, & timely (Doran, 1981). The European Commission has set the target to halve the number of road deaths in 2020 as compared to 2010. However, countries differ in the reductions that can realistically be expected. In some countries there is a long tradition of road safety oriented policy making and the risk is comparatively low already. In other countries, efforts to increase road safety have only recently begun and there is still a lot to achieve. In this case, a stronger reduction of the number of fatalities has been observed in past years and is also realistic to expect in the years to come. A good way to select realistic targets for the reduction of the number of fatalities is to extrapolate the past development into the future. Such an extrapolation gives an indication where the development goes if past efforts are maintained. For some countries, this constitutes an ambitious target already. For others, past efforts might be perceived as insufficient, and the target should be chosen below the number of fatalities forecasted in continuation of the present trend. In each case, a sound forecast for the target year should form the basis for setting the target and monitoring the achievements in the coming years. The present deliverable gives forecasts for 2020 for each European country. While the detailed methodology, including the definition of the statistical models employed was given in Deliverable D4.2 (Martensen et al., 2010), the focus of the present deliverable is on the actual forecasts. In Chapter 2, the principles that played a role in the selection of the statistical models to forecast the fatalities up to the year 2020 are described in a relatively non-technical way. This Chapter also gives a view on problematic issues like the data quality and forecasting in times of the recession. In Chapter 3, an overview of the resulting forecast models will be given. In the Appendix A, the full report on the time series analysis of each country is given. This is a technical description of the forecasting model and the process that lead to its selection is given for each country. The use or non-use of exposure in the final model (presented in the factsheets in Appendix B) is argued on the basis of additional analyses and different forecasting models are compared according to various quality criteria. These detailed country reports are written for experts and an understanding of the statistical principals underlying latent state modelling (see Martensen & Dupont, 2010, D4.2) might be necessary to read them. In Appendix B the forecast factsheet for each country is presented. These factsheets are meant to give a relatively untechnical description of the development of the fatalities (and of the exposure if available) so far. If known, the (possible) reasons for the developments are shortly described. The forecasts of the fatality numbers up to 2020 made under the assumption of continuation of past development continues are also provided. Whenever an exposure measure of the necessary quality was available (see Chapter 2), an estimation of the fatality risk is presented along with three scenarios based on different assumptions for the development of mobility in the next 20 years. (Author/publisher)
Samenvatting