SWOV
Publication

De stijging in het aantal verkeersdoden op rijkswegen in 2015

Statistische analyse, bestudering van ongevallen en verkenning van mogelijke verklarende factoren
Report number: 
R-2016-9
Publication type: 
rapport
Date published: 
2016
Publisher: 
SWOV
, Den Haag
Author(s): 
Stipdonk, H.L.; Bijleveld, F.D.; Davidse, R.J.; Weijermars, W.A.M.; Bos, N.M.; Wijnen, W.; Louwerse, W.J.R.; Goldenbeld, Ch.; Bruin, J. de
Topics:
Verkeersveiligheid in cijfers, Infrastructuur, Visie & beleid, Mens, gedrag & verkeer

Abstract (EN): 

The increase in the number of road deaths on national roads in 2015; Statistical analysis, examination of crashes and exploration of possible explanatory factors

In 2015, the Netherlands counted 621 road deaths. This is an increase of 9% compared to the 570 road deaths in 2014. The increase in the number of road deaths is most strongly visible among car occupants (+ 37; + 20%) and drivers of mobility scooters and vehicles for the disabled (+ 14; + 50%). In the last group the annual number of road deaths had already been increasing  for some time. The number of road deaths in registered crashes on the national road network was considerably higher in 2015 than in 2014; it rose from 63 in 2014 to 82 in 2015 (+ 30%). The national road network consists of all roads of which Rijkswaterstaat is the road authority. Rijkswaterstaat is part of the Ministry of Infrastructure and the Environment and is responsible for the design, construction, management and maintenance of the main infrastructure facilities in the Netherlands. The roads in the national road network are mainly motorways, and also a small set of other road types.

This increase, particularly on the national roads system, led Rijkswaterstaat to commission SWOV to perform a limited study. This study consisted of three components:

  1. a statistical analysis of, among other things, the statistical significance of this increase;
  2. a crash analysis to identify prominent features of the crashes on the national road network;
  3. an analysis of external factors that may explain why the number of road deaths increased in 2015, not only on the national road network in the Netherlands, but generally, in the whole of the country.

This report addresses each of the three questions above, each one in an individual chapter. The findings are summarized below.

Summary of the statistical analysis

  1. In 2015, the number of road deaths on all Dutch roads was significantly higher than was to be expected based on the trend until 2014. This was even more so for the number of deaths on the national road network. However, at this moment, i.e. after one year of data, it cannot be concluded that the increase in the number of road deaths in 2015 in the Netherlands or on the national roads system is the start of a new trend, although this cannot be ruled out either.
     
  2. The increase occurred mainly among car occupants, on all Dutch roads as well as on the national road network.

    a) On the main road network, a large part of the increase occurred in the south of the Netherlands. This increase is very significant. An explanation of this increase in the Province of Noord Brabant is important for being able to explain the number of road deaths in all of the Netherlands. However, no such explanation is presently available.

    b) The increase among car occupants in 2015, in comparison with 2014, is a strong deviation from the trend in the whole of the Netherlands. It cannot be established that this is the start of a trend break, but this cannot be ruled out.

    c) Analyses of reports by Grontmij (2016) and Arane (2016a; 2016b; 2016c) provide insights in the trends over years by speed limit. Among other things, a comparison was made of the number of road deaths, based on 3 years before and 3 years after the change in the speed limit. One of the findings was that particularly on road sections where the speed had been changed before September 2012 the number of deaths had increased, while a decline occurred on road sections where the speed limit was changed in after 2012. A clear explanation could not be given. To be able to say something more on the role of speed, SWOV recommends to investigate the development of the number of road deaths on the basis of road sections distinguished by the (unambiguous) current speed limit of 130 km/h at all times and by 130 km/h in time slots, and to compare the current numbers of road deaths with the numbers of road deaths on the same road sections in the past. Also information about the actual speed driven at the time of each accident has to be taken into account. There are ways in which this information can be obtained, for example from a road side speed detection system, if this is installed at the location of the crash. Presently this data is not available for our research. Information about the actual traffic situation at the time of the crash, such as congestion or work in progress, may also be helpful to make good analysis.

    d) The trend in the share of fatal single vehicle crashes on the national road network does not differ from that of the national total share of fatal single vehicle crashes.

    e) In 2015, as in 2014, there were approximately 1.05 road deaths per fatal crash in the whole of the Netherlands. This means that the trends of the numbers of fatal crashes are not substantially different than of the numbers of road deaths.

    f) The increase in the number of road deaths does not only occur on national roads, but also on the other roads. For national roads, the increase in absolute terms is slightly larger than elsewhere in the Netherlands. Relatively, the increase for national roads is much larger. However, the differences between the numbers in 2014 and 2015 are not exceptional from a statistical point of view. A difference of any two consecutive years, without taking the trend into consideration, can easily be somewhat larger by chance.

  3. The data provided by Statistics Netherlands on the actual number of road have been incomplete for years now (see Fact sheet Road deaths in the Netherlands; SWOV, 2016). While the police are working on improvement of this registration, and continue this improvement year after year, the data of 2014 and 2015 that were eventually registered in BRON, leave room for further improvement. In these years, not all characteristics were available for all fatal crashes. This complicates the analysis and the comparison between road death data for different years (for example, if the speed limit or the mode of transport of the casualty is not reported accurately). It has also transpired that possibly not all road deaths in BRON meet the formal definition (on public roads, not suicide). SWOV recommends reviewing these doubtful cases and, if necessary, removing them from BRON.

Summary of the crash analyses

On the basis of analysis reports of fatal crashes that Rijkswaterstaat has drawn up, the 75 fatal crashes on national roads that were registered in BRON in 2015, were further analysed. Where possible, additional information was gathered (VOA police reports and images using Google Maps and Globespotter).

The crash types

A crash type responsible for a large share in the total number of fatal crashes on national roads is a collision with an obstacle (37% of the deaths, 39% of the fatal crashes; this includes the motorcyclists who crashed into a light pole in a bend). Almost half of the non-crash friendly obstacles (46%) were placed within 10 m from the side marking. A similar proportion (42%) was situated between 10 m and 13 m. The current directive indicates that the obstacle free zone at newly constructed motorways and major maintenance of motorways with a design speed of 120 km/h, must be 13 m or more. In the old directive this distance was 10 m.

Another crash type was the one-sided crashes (single vehicle crashes without collision with an object, 7%). Three times this involved a car that ended up in the water at the roadside and twice it was a motorcyclist who fell on the road shoulder or road surface without contact with an obstacle or other road user. For the one-sided crashes in which a car ended up in the water, the water was situated at more than 13 m of the side marking. Despite this distance the vehicle hit the water and one of the occupants drowned.

A fifth of the fatal road crashes were rear-end collisions (21%). Slightly more than half of these crashes occurred at the tail end of a traffic jam (nine crashes). Matrix signs were present at three out of nine crashes. Compared to the other crash types there were many injuries in the rear-end crashes in addition to the road deaths. The cause probably lies in the large number of vehicles involved.

Half of the head-on collisions (8%) were due to a collision with a ghost driver. The other head-on collisions occurred mainly on a single lane trunk road (100 km/h) on which the driving directions were only separated by a double centre line marking with green filling.

In five of the nine accidents in which a pedestrian or cyclist (12%) was involved, it was unknown what the casualty was doing on the main carriageway of a motorway. The other four casualties were on the lane of a motorway because they were changing a tire, stepped out of the car after a fight, possible ignorance of the Dutch law (foreign cyclist) and inline skating on a parallel road maintained by a national road authority.

In two of the four crashes a motorcyclist slided and fell in a bend of an entry ramp (5%) and hit a light pole in the road shoulder of the outside bend.

Causes of the crashes

The analysis reports of Rijkswaterstaat contain little information about the causes of the fatal crashes. During the investigation the VOA reports by the road crash police analysts became available for a third of the crashes. Given the limited time available, these reports were not studied in detail. They showed, however, that the condition of the vehicle played a role in (at least) four of the crashes (tire defects or short circuit). The VOA reports occasionally mention alcohol, high speed and distraction as factors that have played a role in the occurrence of the crash. The influence of these factors, however, is not systematically examined and mentioned and it is therefore not possible to specify in which share of the crashes such high risk human factors played a role. This is also due to the fact that only a limited number of VOA reports were available for this investigation.

Factors that determine the severity of the consequences

The most important factor that determines the severity of the consequences is a road shoulder in which non-shielded obstacles are present within the prescribed obstacle-free distance. This is the distance within which a vehicle at the prevailing speed limit can come to a stop or can get back onto the roadway.

Other obstacles were shielded, but the ends of the guard rail did not have sufficient room for extension. In two crashes the performance class of the used guard rail was insufficient to stop a truck. Given the location of these guard rails they complied with the guidelines. A heavier performance class is only desirable on viaducts and other locations where a truck going off the road forms a threat for traffic on the underlying road. The heavier performance class can cause more serious injury to occupants of a passenger car (because of the stiffer construction of the guide rail).

The severity of the outcome of a crash is also determined by security measures to or in the vehicle. As it turned out, two drivers did not wear their seat belt correctly.

Recommendations for measures

Based on the available information, the recommendations are pretty much limited to measures that can reduce the severity – the fatal consequences – of crashes. The most important measure relates to making a barrier-free zone with a width of 13 m uniform for motorways with a speed limit of 120 km/hour or higher and shielding obstacles that are within this minimum barrier-free zone. Another solution could be to place a flexible guard rail over the entire length of the road, combined with an escape area of 2.5 m in addition to the emergency lane. At the same time, it is desirable to examine whether an obstacle free zone of 13 m is sufficient for a vehicle driving 130 km/hour to come safely to a halt.

The information that was available for this research provided insufficient handles for measures aimed at preventing the cause of the crash. There was too little information about the pre-crash phase – the last minutes or hours before the crash – to be able to specify which factors played a role in the car driving into the shoulder or the collision with another vehicle. A more detailed crash analysis can give more insight into the cause of the crashes, for example on the basis of the contents of all VOA-reports of fatal crashes on national roads in 2015 (just a few VOA reports could be made available for this research in the short available time and these could only be studied superficially ). An alternative is to carry out an in-depth study which for each crash systematically gathers and analyses information on the infrastructure and the behaviour, the backgrounds and the vehicles of the road users involved

This SWOV analysis of the fatal crashes in 2015 did not specifically look at the role of the 130 km/h speed limit in the occurrence of crashes; among other things because the driven speed at the time of the crash was unknown.

Summary of the analysis of external factors

The analysis of external factors focuses on factors that may have affected traffic and traffic behaviour everywhere in the Netherlands. This means that we did not just look at the national road network.

Only factors for which data was available were examined. Factors such as distraction by media devices (texting, phone calls) were not examined.

Analyses indicate that the increase in the number of road deaths can particularly be observed for:

  • men;
  • 30-to 39-year-olds, 50 to 69-year-olds and the over- 80s;
  • national roads and municipal roads;
  • crashes that occur outside the peak periods.

Possible explanations for the increase in the number of road deaths are:

  • Specifically for national roads: an increase in mobility on national roads. This increase can have made a small (2.2%) contribution to the increase (30%) of the number of road deaths on national roads. On national roads, therefore, the risk (number of deaths per km travelled) also increased.
  • Economic recovery and particularly a resulting decrease in unemployment. Both international literature and Dutch data over the period 2006-2015 show a relation between economic growth and unemployment, and the number of road deaths. In 2015, a clear decline in unemployment can be observed for the first time since six years and the growth of the gross domestic product per inhabitant is at its highest level since 2008. These developments can most probably explain part of the increase in the number of road deaths (see also the next bullet).
  • A possible increase in mobility of inexperienced drivers: in 2015 car ownership increased, after it had declined in 2013 and 2014. As a result of the economic recession, novice drivers may have postponed the purchase of a vehicle. If in 2013 and 2014 inexperienced drivers have postponed the purchase a vehicle until 2015, this will not only have led to an increase in the number of road deaths in 2015, but at the same time will also have had a damping effect on the number of road deaths in 2013 and 2014.
  • A possible worsening in traffic behaviour as a result of a decrease in the number of police roadside checks. The number of police fines in 2015 is halved compared to that in 2014. Whether this has indeed affected traffic behaviour cannot be investigated due to lack of sufficient reliable data on traffic behaviour and enforcement efforts.

 


Bron-URL: https://www.swov.nl/en/publication/de-stijging-het-aantal-verkeersdoden-op-rijkswegen-2015