Safety Management of Road Infrastructure – Tools and Applications

Safety Management of Road Infrastructure – Tools and Applications


Hagen Schueller, PTV Group, Paulo Humanes, PTV Group


This paper deals with tools and applications that help in implementing standard procedures of road safety management like Black Spot Management, Network Safety Management and Road Safety Impact Assessment into practice. Recommendations, data requirements, possible benefits and use cases are discussed. The aim is to show some kind of a road map on how a more effective safety management of the road infrastructure can be achieved.


The EC-Directive 2008/EC/96 describes several procedures for improving road safety in the planning and maintenance process of the road infrastructure. Some of the procedures do rely on good accident data, either by counting number and severity of similar accidents at defined locations or on basis of accident models that are derived from large accident samples. Often it is necessary to combine accident data with other information like maps, network data, traffic volumes or information on design and operation. The mass of data needs support by software tools because of comprehensive data sets or the calculations regarding the entire road network.
The Black Spot Management (BSM) is the most widely applied procedure that is even integrated into the national road administration law in Germany. Since the first uses of electronic accident databases and maps like EUSKA in 2002, there has been an increase in data quality of the police recorded accidents, a decrease in the number of severe road accidents as well as a wider use of accident data for all kind of applications for improving road safety. BSM itself identifies locations with numerous severe accidents and analyzes relevant accident contributing factors. A management of the identified black spots also means the implementation of measures and evaluating the effects afterwards. BSM is a microscopic approach involving police and road administrations to work together to improve single spots in the existing network that did prove to be successful.
The next step will be the consideration of the current road safety situation in new planning schemes and even the assessment of future scenarios. It will need a combination of the available accident data (that was also used for BSM) with other datasets regarding traffic and infrastructure information. One example would be the integration of accident data in the transportation modeling e.g. in the tool VISUM that is commonly used by traffic planners.
There is a need to bring several small site measures in a larger context of strategic planning. The procedure called Network Safety Management (NSM) follows this idea. NSM evaluates the entire road network in a jurisdiction and identifies potentials for improvements. This is done by the calculation of safety parameters derived from section length, traffic volume and accident costs. The results are compared to an optimal safety level. This optimum is what the road section could achieve if it was designed and operated regarding best practice design and has a corresponding road condition. The difference between the actual and the desired safety level is called safety potential (measured in accident costs per kilometer road). It can be used for benefit-cost-analysis in strategic transportation planning of new roads or maintenance projects. Sites with promise or rather a high safety potential indicate that the money spent for infrastructure projects results in the highest outcome regarding accident savings. Because of the high amount of data (traffic volumes and network attributes) software tools are needed. VISUM as a standard tool supports this task and is already in use at many road administrations.
If NSM defines the “need for action” then the next step will be finding the most effective and efficient measures or projects supporting the improvement of road safety. It needs reliable safety indicators that can be used in economic assessments that will describe future changes in number and severity of accidents (accident costs). Whereas traffic forecast based of demand models are standard today, the combination with accident models for prediction of future safety levels would be the obvious next necessary step.
Often the calculation of accident costs is done roughly by using simple accident cost rates and based on a new section without considering the adjacent road network. Road Safety Impact Assessment (RIA) aims at showing changes in road safety for the entire road network that is affected by the infrastructure project. RIA can be based on more complex accident models, forecast of traffic volumes and network attributes. Most of this data is available in transportation models that are created in tools like VISUM. Integrating accident prediction models into the transportation model enhances the results significantly by not only showing how future traffic is distributed but also how it will affect road safety.
Today it is often not possible for the traffic engineers to objectively consider safety aspects in the planning process. One reason for that is that some of the existing procedures are relatively new and the existing tools have no sufficient road safety features yet. There is a need for finding simple and cost efficient ways for improving that situation.


Association for European Transport