CAR 44 | Prevention of Overtaking Accidents on Rural Roads

Home , Overtaking

German Insurance Association Nr. 44

Prevention of overtaking accidents on rural roads Compact accident research

Insurers Accident Research 3

Content

Preliminary remarks 4

Introduction 5

Methodology 5

Accident occurrence 5

Effect of prohibiting overtaking and setting up a speed limit 6

Effect of bends, cambers and sight distances 7

Detailed analysis of the accident occurrence 9

Detailed analysis of overtaking behavior 10

Summary and recommendations 11

References 12 4 Compact accident research 44 | Prevention of overtaking accidents on rural roads

Preliminary remarks

9% of the people killed on roads outside of built-up areas are the victims of overtaking accidents. These accidents are thus among the accidents with the most serious consequences on roads outside of built-up areas.

The purpose of this now completed UDV research project was to ascertain the infrastructure-related, traffic-control and traffic-related factors influencing overtaking behavior and overtaking accidents in order to be able to identify suitable measures to prevent these severe accidents.

500 sections of road in five different German federal states that were found to have the highest numbers of accidents served as the basis for this study. It was ascertained where overtaking was prohibited (by means of road markings and signs) and what speed limits were impo- sed. On 100 of these sections of road, the sight distances at the accident locations were ascertained and included in the analysis of the accidents. In a detailed investigation, overtaking maneuvers were recorded on video and analyzed.

This compact accident research report summarizes the findings of this study. You can obtain more detailed infor- mation from research report no. 24, entitled „Untersu- chung von Maßnahmen zur Prävention von Überholun- fällen auf einbahnigen Landstraßen“ (a study of measures designed to prevent overtaking accidents on single-carriageway roads outside of built-up areas). You can down- load this report free of charge at www.udv.de.

Insurers Accident Research 5

Introduction • Methodology • Accident occurrence

horizontal and vertical plan, and the sight distances were ascertained, in order to analyze any effect they might Introduction have on the accident occurrence. Finally, in a total of 50 selected stretches of road with conspicuously high Around 75,000 accidents involving injury on roads out- numbers of accidents, a microscopic detailed analysis of side of built-up areas were reported to the police in 2012. overtaking behavior was carried out by means of video 2,151 people were killed in them, and a further 25,766 peo- observations. ple were seriously injured. Around 6% of these accidents and 9% of the fatalities occurred as a result of an overtaking maneuver. Overtaking accidents are thus particularly serious.

The UDV therefore carried out a scientific investigation Accident occurrence to find out how, where and why overtaking accidents happen on single-carriageway roads outside of built-up The analysis of the accident occurrence consisted of a areas, and whether these accidents happen despite the nationwide analysis of the structure and consequences fact that overtaking is prohibited or in areas in which of overtaking accidents on roads outside of built-up are- overtaking is permitted [1]. Above all, the aim was also to as in 2009 and a detailed analysis of the accident oc- ascertain what can be done to prevent these accidents. currence in five federal states from 2007 to 2009. On 58,269 kilometers of single-carriageway roads outside of built-up areas, there were 85,345 accidents involving injury and serious property damage (category 1 to 4), inclu- ding around 6,200 overtaking accidents.

Methodology The results can be summarized as follows: • Two-thirds of the overtaking accidents involving The basis of the study was a comprehensive analysis of injury were accidents in longitudinal traffic, a further sections of road outside of built-up areas where a dispro- 21% were turning-off accidents, and the remaining portionate number of overtaking accidents in five se- 10% were driving accidents (i.e. the driver lost control lected German federal states occurred. The road informa- of the vehicle). tion databases and accident databases were prepared • In half of the overtaking accidents the driver overtook and combined with this purpose in mind. Based on the despite the fact that there was oncoming traffic or the calculation of the accident indicators, 500 sections of traffic situation was unclear. 24% of the overtaking road were identified in the area covered by the study that accidents involving injury were frontal collisions, a were most affected by overtaking accidents. A survey ve- further 24% were side-on collisions with vehicles hicle was driven along these sections of road. The sec- traveling in the same direction, 14% were rear-end tions where overtaking was prohibited (by means of road collisions with the vehicle in front, and 11% were markings and signs) and speed limits were set up were accidents with vehicles turning into or crossing a road. ascertained. The overtaking accidents were assigned to • 76% of the people involved in the accidents who these different sections of road. In addition, for the 100 were partially responsible for the accident were sections of road with the most conspicuously high driving passenger cars, 16% were riding motorcycles, numbers of accidents, the axes were redesigned both in and 8% were driving trucks. 6 Compact accident research 44 | Prevention of overtaking accidents on rural roads

Effect of prohibiting overtaking and setting up a speed limit

The accidents that occurred in these sections were analyzed, and the following results were obtained: Effect of prohibiting overtaking and • 74% of the overtaking accidents occurred in sections setting up a speed limit where overtaking was not prohibited. • 72% of the overtaking accidents occurred in sections with a speed limit of 100 km/h. As a result of further filtering, 500 sections of road out- • There was a lower risk of accidents occurring on side of built-up areas with a conspicuously high number stretches of road where overtaking was prohibited, of accidents and a total length of 2,235 kilometers were and the accidents were less serious (lower accident identified, on which a total of 1,557 overtaking accidents rate and accident cost rate). had occurred. These 500 stretches of road were driven in • There was an even lower risk of accidents and lower order to ascertain whether the accidents occurred on accident severity on stretches of road with speed sections where overtaking was prohibited. limits of less than 100 km/h (Figure 1).

On average, overtaking was prohibited on 31% of the stretches of road driven. If overtaking was prohibited in both directions, this was generally indicated by a broken middle marking line (in 56% of cases) or by traffic signs (in 22% of cases). However, the combination of a broken middle marking line and the traffic sign, which is actually not permissible, was also frequently found (in 22% of cases). On around a third of the sections of road included in the study, there was a speed limit of less than 100 km/h.

Risk and consequences of overtaking accidents on two-lane rural roads by speed limit

Overtaking regulation Accident rate Accident cost rate Accident rate Accident cost rate [A/(106•veh•km)] [€/(1000•veh•km)] [U/(106•veh•km)] [€/(1000•veh•km)] Figure 1: Accident rate and accident cost rate 0.12 19.0 0.09 14.5 depending on overtaking regulations and speed limits

0.09 15.8 0.06 8.9

1) 0.10 12.0 0.07 9.4

0.07 13.9 0.07 9.7

1) partly removal slow vehicles Insurers Accident Research 7

Effect of bends, cambers and sight distances

100 s 90 80 97 n = 333 overtaking accidents included* 67 70 Bend to the right 60 Bend to the left 47 50 37 40 23 Figure 2: 25 28 Overtaking accidents Number of overtaking accident 30 22 by bend radius and 13 20 12 13 13 direction of bend 30 9 8 24 5 7 10 6 15 10 12 7 7 3 2 5 2 *It notovertaking accidents was possible to pinpoint of five the location 0 3 7 0 0 0 ∞ 20 ≤ > 1,800 201 – 400 401 – 600 601 – 800 801 – 1,000 1,001 – 1,201,201 – 1,4001,401 – 1,601,601 – 1,800 Bend radius [m]

ted at over two-thirds (71%) of all cambers at which a conspicuously high number of accidents occurred (Figure 3). Effect of bends, cambers and sight distances Figure 3 shows whether signs and what kind of road markings were in place at typical accident locations on roads outside of built-up areas. Overtaking accidents also oc- The axes of the 100 stretches of road with the most cons- curred at intersections, and overtaking was not prohibi- picuously high numbers of accidents were redesigned ted in around half of these cases. for this detailed analysis. It was possible to pinpoint the locations of 333 overtaking accidents over a total road Around 70% of the overtaking accidents occurred in loca- length of 350 km. In addition, on the basis of the redesign tions where the sight distances were shorter than those the sight distances around the accident location were stipulated for safe overtaking in the guidelines for the ascertained. design of roads outside of built-up areas [2]. Moreover, 24% of these accident locations had shorter sight di- Around 71% of the accidents occurred within the sphere stances than the limit values stipulated in the guidelines of influence of bends. The tighter the bends, the more for the road markings [3]. overtaking accidents were found (Figure 2). Moreover, overtaking was not prohibited at around 73% of the Figure 4 essentially shows two trends. First, both the bends at which conspicuously high number of accidents number and severity of overtaking accidents are lower occurred. where sight distances are shorter. Second, in general fe- wer accidents occur on stretches of road where overta- Around 38% of the overtaking accidents happened in the king is prohibited than on stretches that have broken direct vicinity of cambers, and overtaking was not prohibi- middle marking lines. However, there are two exceptions 8 Compact accident research 44 | Prevention of overtaking accidents on rural roads

Effect of bends, cambers and sight distances

n = 236 n = 114 n = 39 100 5.93 4.19 12.82 15.68 20.18 80 5.08 4.39 38.46 Sign/road marking 11.86 13.16 60 Road marking

2.56 Sign Figure 3: 40 61.44 10.26 Signing and marking Hazard line at typical accident locations Relative frequency Relative 57.89 Broken middle 20 35.90 marking line

0 Acidents Accidents Accidents at bends at cambers at intersections (as defined by RAS-L)

Figure 4: Accident indicators by sight distance ranges and overtaking regulation

0.08 16 n = 70 n = 138 n = 92 n = 70 n = 138 n = 92

0.06 12 0.072 0.073 14.72 14.74 0.069 13.04 13.25 •veh•km)]

6 0.062 0.04 8

9.30 8.23 0.039 0.040 0.02 4 Accident cost rate [ € /(1000•veh•km)] Accident cost rate Accident rate [O-A/(10 Accident rate

0 0 No sight Sight distances SIght distances No sight Sight distances SIght distances distances between stipulated by RAS-L distances between stipulated by RAS-L RAS-L and RMS RAS-L and RMS

Overtaking prohibited Overtaking allowed Insurers Accident Research 9

Detailed analysis of the accident occurrence

to the general conclusions. The accident risk, as measured by the accident rate, is very similar for all three sight distance ranges for the sections of road where Detailed analysis of the accident overtaking is allowed. Where overtaking is prohibited, on occurrence the other hand, the accident risk decreases with shorter sight distances, in line with the general trend. Conside- ring the accident severity measured by accident cost A detailed analysis of the accidents that occurred was rates and accident cost density on sections of road with carried out for 43 sections of road on which there were adequate sight distances there are no differences bet- 166 overtaking accidents during the period studied (2007 ween sections of road where overtaking is allowed and to 2009). The key findings were as follows: sections where overtaking is prohibited. Although the • The overtaker was generally responsible for causing accident risk is lower where overtaking is prohibited, the accident (in 97% of the cases). when accidents do happen, they are similar in terms of • The accidents were predominantly caused by car severity to accidents on sections of road where overta- drivers (82%). king is allowed. • One in five overtaking accidents occurred at intersections or junctions with agricultural roads. • In 31% of cases more than one vehicle was overtaken. • 48% of the overtaking accidents occurred during the overtaking maneuver, 27% when pulling out and 19% when pulling in again. • Inexperienced drivers were at particular risk of causing these accidents. Around half (46%) of the people responsible for causing overtaking accidents were younger than 30 (Figure 5). • 85% of the people primarily responsible for causing Figure 5: People primarily responsible for causing overtaking accidents were men (Figure 5). overtaking accidents by age and gender

50 n = 130 overtaking accidents included 45 40 Women 40 15% Men 35 27 30 85% 25 11% 19 20 17 21% 13 15 18% 89% 15% 10 79% 5 5 82% 85% 4 5 40% 20 % 100%

Number of overtaking accidents Number of overtaking 60% 80% 0 16-19 20-19 30-39 40-49 50-59 60-69 70-79 80-89 Age of person causing the accidents 10 Compact accident research 44 | Prevention of overtaking accidents on rural roads

Detailed analysis of overtaking behavior

• Overall, the most important causes of accidents were problems with assessing sight distances, distances and speeds; failure to recognize a driver‘s intention to Detailed analysis of overtaking turn off; loss of control; a poor sense of orientation behavior with regard to following vehicles and conflicts on pulling out and back in; and failure to maintain adequate safety clearances. In the detailed analysis, a total of 15,173 overtaking maneuvers were investigated using video recordings (Ta- ble 1). Around a third of these involved cars overtaking other cars, 37% involved cars overtaking trucks, 17% involved cars overtaking light motorcycles or other vehicles, and 7% involved motorcycles overtaking cars.

The longer the sight distances, the more overtaking maneuvers took place. However, many overtaking maneuvers also took place where sight distances were poor or inadequate.

The analysis of overtaking accidents (Figure 6) shows that the accident risk on sections of road outside of built- up areas with shorter sight distances than stipulated in the guidelines for road markings (RMS) [3] was five times

Table 1: Overtaking maneuvers investigated

Overtaking Vehicle overtaken vehicle

Car Truck Light motorcycle Motorcycle Other vehicles

Car 5,143 5,560 1,490 111 1,127

Truck 35 62 94 2 96

Light motorcycle 9 5 2 0 5

Motorcycle 1,099 294 19 2 11

Other vehicles 4 1 2 0 0

Total number of overtaking maneuvers = 15,173 Insurers Accident Research 11

Summary and recommendation

0.06 (27.5 km) Summary and recommendations 0.056 Overtaking accidents are generally particularly serious. They occur predominantly in locations where overtaking 0.04 is permitted. However, the analyses show that the majo- rity of these accidents take place in locations where drivers cannot see far enough ahead. Drivers are evidently not always able to recognize whether the free stretch of 0.02 (89.9 km) road ahead is long enough to overtake safely. The follow- 0.017 (96.9 km) ing recommendations are therefore made: 0.010 • The level of risk is high when overtaking. On sections Accident risk of overtaking [(A/3a)/(Overtaking/10h)] Accident risk of overtaking 0 No Sight Sight Sight of road where sight distances are inadequate or of distances distances distances medium length, overtaking should therefore be between as stipulatd RAS-L and by RAS-L prohibited in order to prevent driving errors. RMS • Prohibiting overtaking reduces the accident risk and Sight distance ranges on section of accident severity. road investigated (each driving direction) • A speed limit further reduces the risk and severity of overtaking accidents. Fig. 6: Accident risk of overtaking maneuvers • Overtaking must be clearly and effectively prohibited in different sight distance ranges near intersections.

In addition, the provision of an alternating central overtaking lane for both directions (2+1 roads) reduces the higher than on sections of road with good sight di- pressure to overtake. Lower speeds also have the effect stances, as stipulated in the guidelines for the design of of reducing the pressure to overtake. The development of roads (RAS-L) [2]. suitable driver assistance systems could also make overtaking safer in future. Where overtaking was prohibited, the number of overtaking maneuvers on roads outside of built-up areas was The dangers of overtaking and the correct overtaking be- lower, but they were not eliminated. On sections of road havior (practice) should be emphasized during driver with speed limits of less than 100 km/h, significantly lo- training and in targeted safety campaigns, because over- wer overtaking rates were recorded. taking in locations where it is prohibited is not a minor offense. 12 Compact accident research 44 | Prevention of overtaking accidents on rural roads

References

[1] Richter, T.; Ruhl, S.: Untersuchung von Maßnahmen zur Prävention von Überholunfällen auf einbahnigen Landstraßen (A study of measures designed to prevent overtaking accidents on single-carriageway roads outside of built-up areas). Research report no. 24. German Insurers Accident Research (UDV), Berlin, 2014.

[2] German Road and Transportation Research Associati- on (FGSV). Richtlinien für die Anlage von Straßen - Teil Linienführung (RAS-L) (Guidelines for the design of roads – section on alignment), Cologne, 1995.

[3] German Road and Transportation Research Associati- on (FGSV). Richtlinien für die Markierung von Stra- ßen, Teil 2: Anwendung von Fahrbahnmarkierungen (RMS-2) (Guidelines for road markings, part 2: Use of road markings), Cologne, 1980. Insurers Accident Research 15

Imprint

German Insurance Association Insurers Accident Research

Wilhelmstraße 43/43G, 10117 Berlin Postfach 08 02 64, 10002 Berlin E-mail: [email protected] Internet: www.udv.de Facebook: www.facebook.com / unfallforschung Twitter: @unfallforschung YouTube: www.youtube.com / unfallforschung

Editors: Dr.-Ing. Jean Emmanuel Bakaba, Dipl.-Ing. Jörg Ortlepp

Layout: pensiero KG; www.pensiero.eu

Photo references: Cover photo: Fotolia, UDV

Published: 01/2016 German Insurers Association

Wilhelmstraße 43 / 43G, 10117 Berlin PO Box 08 02 64, 10002 Berlin

Phone: + 49 (0) 30 2020 - 5000, Fax: + 49 (0) 30 2020 - 6000 www.gdv.de, www.udv.de