Usefulness of 'floating car speed data' for proactive road safety analyses : analysis of TomTom speed data and comparison with loop detector speed data of the provincial road network in the Netherlands. Currently, data on road crash casualties provide local authorities with insufficient starting points for their road safety policy. The regional road safety bodies of the Dutch provinces of Gelderland (ROVG, currently the ROV Eastern Netherlands), Fryslân (ROF) and Zeeland (ROVZ) and the province of Utrecht have therefore asked SWOV to join them in monitoring a number of alternative indicators. In literature these alternative indicators are known as ‘safety performance indicators’ abbreviated as SPIs. SPIs are indicators for risk factors that exhibit a strong causal relationship with road safety. Use of such indicators in policy ties in with a number of developments in, among others, provinces. One of the most important developments is the desire of provinces (and increasingly also of municipalities) to make more proactive policy. Recently a tool has been developed that allows measuring and prioritizing road safety in a more proactive way (ProMeV). Starting point for the monitoring pilot was to begin the study with a limited set of indicators for risk factors that were on the ‘wish list’ of the abovementioned provinces and that also meet the definition of an SPI. This yielded indicators for a) drink-driving, b) speed, c) quality of road infrastructure, and d) bicycle safety risks. SWOV has previously made suggestions about how to create a decentralised monitoring network for these four indicators. For speed this suggestion can be found in the report of Goldenbeld & Aarts (2013). For the indicator ‘speed’, we first investigated the usefulness of floating car data (FCD) products, more in particular the TomTom speed profiles as a road safety indicator. This follows from the fact that several provinces are currently using this product. It also provides speed data of almost the entire road network, as opposed to existing traditional monitoring networks and measurements that mostly focus on the main road network. This report discusses the following questions: 1. To what extent can speed data from navigation equipment be used as a source for SPIs in the field of ‘speed’, and if so, how? 2. Which recommendations can be made in addition to those of Goldenbeld & Aarts (2013) about a speed monitoring network for the secondary road network in particular on the basis of this exploration? This study concludes that speed data derived from FCD can to some extent be used for gaining an insight into speeds of road traffic as one of the SPIs. The main advantages of using FCD are: - The data covers a much greater proportion of the road network than is the case with more traditional methods such as loop and radar measurements. - The data not only concerns measurements at one point on a trajectory (as is the case with more traditional methods), but provide insight into the average speeds for an entire trajectory. - Collected data offers a usable idea of the average traffic speeds on a trajectory, especially for the hours between 04:00 AM and 09:00 PM. - Historical data is available at the moment that it is decided to start collecting speed data. The main limitations of FCD in relation to road safety are: - FCD cannot be used to calculate average speeds for roads with too little traffic. - As yet, no distinction is possible between speeds of different groups of road users and speed differences between individual road users. - The exact relationship between average speed as measured with FCD methods and road safety is not known. Peak speeds of a single vehicle or on a small part of the trajectory are not expressed after averaging. - The continuity of measurements is not guaranteed and is not within the direct sphere of influence of policy makers. For use in a monitoring network we recommend taking continuity in the method (s) of collection as a starting point. FCD may be interesting because it covers a greater part of the road network, but it is less suitable for specific issues (high speeds, groups of road users). A decentralised monitoring network for indicators of the risk factor speed make use of the initiative of, for example, the Dutch National DataWarehouse (NDW) to enrich the existing data file with speed information of FCD. Further exploration of the possibilities is therefore to be recommended.
Samenvatting