Fs Driving Strips

SWOV Fact sheet New type of layout for 60 km/h rural roads Summary In a sustainably safe traffic system, uniformity of traffic facilities is a point of special interest. Uniformity ensures recognizability and predictability of (critical) traffic situations. The uniformity of rural access roads can be increased by applying a narrow single lane for motorised vehicles in the middle of the carriageway: a driving strip. Its marking consists of broken lines, and on each side of the driving strip some space is left. These 'edge strips' can be used by cyclists if they are sufficiently wide. Studies have shown that this type of marking slightly increases road safety. For this reason, but particularly because it improves the recognizability and predictability of the road environment, SWOV recommends this type of marking for rural access roads. Background When redesigning the rural roads according to the Sustainable Safety guidelines, the 80 km/h roads with a minor traffic function in rural residential areas are converted into access roads in a 60 km zone. This road type is intended to be used by all transport modes and has a speed limit of 60 km/h. The motivation for this 60 km/h speed limit is that it lies in between the current general speed limits of 50 and 80 km/h, is clearly lower than 80 km/h and, in spite of this, does not result in an unacceptable extra journey time in rural areas. In a sustainably safe traffic system, uniformity of traffic facilities is a point of special interest. Uniformity is a way of ensuring recognizability and predictability of (critical) traffic situations. The uniformity of rural access roads can be increased by applying a 'driving strip' in the middle of the carriageway (see Figure 1). Its marking consists of broken lines, and on each side of it there is an 'edge strip' that is sufficiently wide for cyclists to use. These edge strips are called non-compulsory bicycle lanes.

Edge marking

Edge strip

Driving strip

Edge strip

Carriageway width

Figure 1. Driving strip with edge strips; cross section (left) and view from above (right). How unsafe are access roads? SWOV has estimated the numbers of casualties on access roads with speed limits of 60 km/h and 80 km/h (see Table 1; Janssen, 2007)1. In 2000 there were 23 deaths and 141 in-patients on 60 km/h roads. In 2005 this had increased to 56 deaths and 768 in-patients. Beside this increase on 60 km/h roads, there was a decrease in the number of casualties on 80 km/h roads. These developments are partly the result of the steady transforming of 80 km/h roads to 60 km/h roads, and therefore of a change in road length. The change in traffic volumes on these road types also played a part.

1

60km/h roads are mainly rural access roads that have been integrated into a 60 km zone (AR60). With 80 km/h roads we here mean both distributor roads with a speed limit of 80 km/hour (DR80) and access roads where there is still the old speed limit of 80 km/h (AR80).

SWOV Fact sheet Reproduction is only permitted with due acknowledgement

1

© SWOV, Leidschendam, September 2007

Casualties Deaths

In-patients

Road type

2000

2001

2002

2003

2004

2005

60 km/h

23

24

27

33

44

56

80 km/h

534

500

479

482

368

331

60 km/h

141

283

399

529

651

768

80 km/h

5,174

4,815

4,762

4,743

4,379

3,984

Table 1. Estimation of the numbers of deaths and in-patients in the 2000-2005 period (Janssen, 2007). Influence of the change in road length SWOV estimates that in the period up to 2000 approximately 5,300 km of access roads had been transformed into 60 km zones. In 2005 this had increased to a total length of about 13,800 km. The road length of 80 km/h roads had decreased from about 49,500 to 42,800 km (Janssen, 2007). If we now divide the number of fatalities by the road length, the fatality density (number of road deaths per 1000 km) remained about the same between 2000 and 2005 (see Table 2). Generally speaking, the density on 60 km/h roads is approximately half that of 80 km/h roads; but not exactly: in the 2000-2005 period, the density decreased with about 28% on the 80 km/h roads. Road type

2000

2001

2002

2003

2004

2005

60 km/h

4.3

3.3

2.9

2.9

3.5

4.1

80 km/h

10.8

10.5

10.4

10.8

8.4

7.7

Table 2. Estimation of fatality densities (fatalities per 1000 km road length) in the 2000-2005 period (Janssen, 2007). Influence of the change in kilometres travelled Between 2000 and 2005 the traffic volume on both road types increased (Janssen, 2007).In spite of their total road length decreasing, the annual number of motor vehicle kilometres on 80 km/h roads increased by about 7%, from an estimated 53.0 to 56.8 billion motor vehicle kilometres. On the 60 km/h roads the traffic volume even more than doubled, going up from approximately 630 to 1650 million vehicle kilometres. The main cause for this is the increase in road length. If we now divide the number of casualties by the estimated traffic volume, we arrive at the casualty rates as given in Table 3. On 60 km/h roads the death rate has remained more or less the same. On 80 km/h roads this rate decreased by about 40% between 2000 and 2005. The death rate on 60 km/h roads is considerably higher than that on 80 km/h roads, a factor 6 in 2005, and the decreasing trend is much smaller. For in-patients, the casualty rate on 60 km/h roads has risen. According to Janssen (2007) the increase in the number of in-patients can therefore not only be explained by the increase in road length, but is also due to an increase in crash rate (see Table 3). This increase in casualty rate is striking and does not run parallel with the national decreasing trend. However, more research is required in order to make any more statements. Casualties Deaths

In-patients

Road type

2000

2001

2002

2003

2004

2005

60 km/h

36.0

27.7

24.7

24.7

29.4

34.2

80 km/h

10.1

9.3

8.7

8.6

6.5

5.8

60 km/h

224.5

326.5

360.6

393.5

435.3

466.3

80 km/h

97.5

89.5

86.4

84.9

77.5

70.2

Table 3. Estimation of casualty rates (casualties per billion motor vehicle kilometres) in the 2000-2005 period (Janssen, 2007). Which measures can be taken? The rural access roads can be subdivided into roads of type I (carriageway width of 4.50-6.20 m) and type II (carriageway width of 2.50-4.50 m) (Table 4). From the Sustainable Safety point of view, the 60 km/h speed limit on access roads is too high for the mixing of different transport modes. For safe mixing the maximum speed should not be higher than 30 km/h. To make the situation as safe as possible, we recommend applying a driving strip for car traffic in both directions in the middle of the

SWOV Fact sheet Reproduction is only permitted with due acknowledgement

2

© SWOV, Leidschendam, September 2007

carriageway on type I access roads (CROW, 2002; see also Figure 1). Such a visual narrowing is intended to make motor vehicles drive slower and more in the middle of the driving strip. The driving strip is marked by broken lines in the ratio of 1-3 (1 metre line and 3 metres blank). The spaces between the driving strip and the edge of the carriageway on either side are called edge strips. These strips can be used by motor vehicles when overtaking or passing oncoming vehicles; they are then referred to as overtaking/passing strips. In addition, edge strips provide more room to rectify steering, thus reducing the risk of going off the road. We expect that this will reduce the damage to road shoulders, which in turn will substantially lower the maintenance costs. In principle, cars and cyclists use the same lane on access roads, with large differences in driving speed when passing and approaching each other. The recognizability of this situation increases by using uniform markings which clearly indicate where cars should drive. In their Road design manual (CROW, 2002), the chapter on access roads recommends an edge strip width of 25-50 cm. In later years, this strip width was reduced to a maximum of 40 cm to prevent road users using this relatively narrow edge strip as a cycling facility. If the carriageway is sufficiently wide (> 5 m) the edge strip can be designed as a non-compulsory bicycle lane of at least 1.25-1.50 m wide with a broken edge marking in the 1-1 ratio (CROW, 2002; 2004). Characteristic*

Access road type I

Access road type II

Speed limit

60 km/h

60 km/h

Number of lanes

One

One

Carriageway width

4.50 – 6.20 m

<4.50 m

Width of driving strip

3.00 – 4.50 m

Same as carriageway width

Marking

Broken edge marking (10 cm wide)

No marking*

Width of edge strip

0.25 – 0.40 m (overtaking/passing lane) 1.25 – 1.50 m (non-compulsory bicycle lane)

n.a.

Proportion of road length**

30 -40%

70 - 60%

Table 4. Features of rural access roads (* CROW, 2002; 2004; **Schoon, 2000; and Janssen 2002). In the Design manual for bicycle traffic (CROW, 2006), the width of the non-compulsory bicycle lane was reduced to a minimum of 1.5 m. The non-compulsory bicycle lane has no legal status; motor vehicles may use it, stop on it, and park on it. Up till now this non-compulsory bicycle lane has sometimes been given extra emphasis by giving it a red colour. However, in the Design manual for bicycle traffic this red colour is strictly reserved for the bicycle lane. Bicycle lanes are further distinguished from non-compulsory bicycle lanes by a bicycle symbol on the road surface every 500 metres and after every intersection. Bicycle lanes have the legal status of a bicycle facility (Traffic Rules and Signs Regulation, article 10, 1990). Drivers of motor vehicles and moped riders are permitted to use the bicycle lane, but they are not allowed to stand still or park on or next to it. The Design manual for bicycle traffic advises against positioning parking spaces next to the bicycle lane. Separate bicycle and moped tracks do not fit the Sustainable Safety character of the access road: mixing of all transport modes is one of its important features. Only if the daily traffic volume exceeds 2000-3000 motor vehicles per day, does CROW (2006) recommend a separate bicycle track. When redesigning the current situation with a smaller traffic volume, bicycle and moped tracks can be converted into non-compulsory bicycle lanes, with mopeds on the carriageway. According to the recommendations in the Road design manual, access roads of type II are too narrow (< 4.5 m) for a marked driving strip and therefore have no edge marking. A broken edge marking can be applied here (in the ratio of 3-1; 3 metres line, 1 metre blank) in dangerous bends or when vehicles regularly go off the road. On both types of access road, an axis marking may only be applied very exceptionally: only in a bend for safety reasons and only on short stretches (CROW, 2002). What are the results of recent studies? SWOV has made a comparison of access roads with and without a driving strip (Van der Kooi & Heidstra, 1999). The conclusion was that the average driving speed was slightly lower on roads with a SWOV Fact sheet Reproduction is only permitted with due acknowledgement

3

© SWOV, Leidschendam, September 2007

driving strip than on roads without edge strips, and that the space between cyclists and passing cars was slightly smaller on roads with edge strips, than on those without. Then SWOV carried out before-and-after studies of traffic situations on roads which initially had no driving strip, but to which a driving strip with non-compulsory bicycle lanes had been added later (Van der Kooi & Dijkstra, 2003). At each location the driving speeds of cars were measured and their place on the road i.e. their lateral position. This study showed that a driving strip with edge marking has a canalising effect and that both cyclists and motorists seem to accept it. These are positive effects. The cyclists use 'their own' strip and usually keep further from the edge of the road than before there was such a strip. Generally, the motorists also keep slightly further away from the edge in case of a driving strip. When passing a cyclist they often choose not to cross the edge strip on the other side of the road. At the same time, when motorists overtake a cyclist on a non-compulsory bicycle lane they are often closer to the cyclist. To what extent the distance being smaller by a few centimetres is dangerous is not yet known. In most cases, the average driving speed goes down by a few km/h as a consequence of the edge lanes. This is a positive development that, however limited it may be, has a positive effect on nearly all crash types. It must also be noted that most non-compulsory bicycle lanes in the SWOV study were narrower than the advised minimum width of 1.25 m. We expect a wider lane to have more effect on the speed and the distance between car and cyclist, but this has not yet been confirmed by research. A meta-analysis of the effects of changes in road marking, both in the Netherlands and abroad (Davidse et al., 2004), has shown that continuous lines are noticed better than broken marking. The good visual guidance has positive effects on the lateral position of cars and lowers the risk of running off the road and of damage to road shoulders. An effective line marking reduces the mental burden and the concentration required to stay on the road. This could have the unintended effect of drivers increasing their speeds. Broken lines offer slightly less visual guidance, but give a better impression of the speed driven. When the aim is to reduce speeds, such as on access roads, broken lines are preferred. Recent SWOV studies of the recognizability and predictability of road types show that red noncompulsory bicycle lanes and bicycle lanes do indeed increase the distinction between road types. Expectations about the presence of other road users, in this case cyclists and moped riders, proved to be more correct when these red lanes were present (Aarts & Davidse, 2007; see also the SWOV Fact sheet Recognizable road design). As an assessment of the 60 km/h projects, the Association of Water Boards commissioned a largescale accident analysis, with before-and-after studies plus control areas, in 20 zones with a total road length of 850 km (Beenker, 2004). This study showed that the introduction of 60 km/h zones had resulted in a decrease of the number of crashes on both road segments and intersections. The number of casualty crashes2 on road sections had decreased significantly by 17%. The roads in this study had a width of between 3 and 5 m. Only those roads with a carriageway of 4.5-5 m wide will have edge strips. It cannot yet be said with certainty which part of the decrease can be attributed to roads with edge strips. The total number of injury crashes on intersections decreased by as much as 47%, probably due to the construction of raised junctions at dangerous intersections. The reduction in the number of injury crashes on the combined road sections and intersections was 24% and the reduction in the number of severely injured3 casualties was 28% (Beenker, 2004). What are the costs and benefits? For converting a road into a type I rural access road the following costs are estimated: installing 60 km/h road signs, removing the axis marking, and applying broken edge markings on both sides of the road. This adds up to a total of €1,880 per kilometre (NMB, 2004). About 46,000 of the approximately 54,000 km of rural non-motorways will be converted into access roads (Janssen, 2005). SWOV had previously estimated that 30-40% of these roads (road width of > 4.5 m) would have a driving strip with edge markings (Schoon, 2000; Janssen, 2002). This requires an investment of €25-35 million. We assume here that the old (axis) marking will only be removed on access roads of type I and that any existing markings on type II roads will disappear from natural wear and tear. As there is no extra budget as yet, the conversion is necessarily carried out as part of 2

Casualty crashes are crashes with deaths, in-patients, Accident & Emergency department patients, or slightly injured (Beenker, 2004). 3 Severely injured casualties are deaths and in-patients.

SWOV Fact sheet Reproduction is only permitted with due acknowledgement

4

© SWOV, Leidschendam, September 2007

management and maintenance. This causes the process to continue to not later than 2020. The water boards aim to have implemented all their 60 km/h areas before 2010. The water boards' study (Beenker, 2004) showed that the average costs per kilometre will amount to €10,106 for all road sections and intersections in all areas taken together. Assuming a 28% reduction of the number of severely injured casualties, the water boards achieve a cost-effectiveness of €17,600 per saved severe casualty in a 60 km zone. As long as it is not known which measures have contributed to the 28% reduction in severely injured casualties, unfortunately nothing can be said at the moment about the expected benefits of applying a driving strip with edge strips. Conclusion The uniformity of rural access roads is increased by applying a driving strip in the middle of the carriageway. The marking of the driving strip consists of broken lines. The remaining space is used for two edge strips that can be used by bicycles if they are sufficiently wide. Such strips are called noncompulsory bicycle lanes. If a bicycle symbol is painted on the lane surface, it is called a bicycle lane. Bicycle lanes have a legal status and are preferably carried out in red asphalt. Meanwhile, the driving strip with edge markings has been included as access road type I in the Dutch Road design manual and in the Dutch guidelines Essential recognizability features. Research shows that this type of marking slightly improves road safety. For this reason, but especially to improve the recognizability and predictability of the road image for a road type, SWOV recommends this type of markings for rural access roads. Publications and sources (SWOV reports in Dutch have an English summary) Aarts L.T &. Davidse, R.J. (2007). Recognizable layout of roads; Final report of the predictability projects in SWOV's 2003-2006 research programme. R-2006-18. SWOV, Leidschendam. [in Dutch] Beenker, N.J. (2004). Evaluatie 60 km/uur projecten; Eindrapport. In opdracht van de Unie van Waterschappen. VIA Advies in verkeer & informatica, Vught. CROW (2002). Handboek wegontwerp wegen buiten de bebouwde kom: erftoegangswegen. Publicatie 164d. C.R.O.W kenniscentrum voor verkeer, vervoer en infrastructuur, Ede. CROW (2004). Richtlijn essentiële herkenbaarheidkenmerken van weginfrastructuur: wegwijzer voor implementatie. Publicatie 203. C.R.O.W kenniscentrum voor verkeer, vervoer en infrastructuur, Ede. CROW (2006). Ontwerpwijzer fietsverkeer. Publicatie 230. C.R.O.W kenniscentrum voor verkeer, vervoer en infrastructuur, Ede. Davidse, R.J., Driel, C.J.G van & Goldenbeld, Ch. (2004). The effect of altered road markings on speed and lateral position: a meta-analysis. R-2003-31. SWOV, Leidschendam. [In English] Janssen, S.T.M.C. (2002). A method for calculating changes in accident rates on roads made sustainably-safe. R-2002-23. SWOV, Leidschendam. [in Dutch] Janssen, S.T.M.C. (2005). The Road Safety Explorer used regionally; The calculation method and its assumptions. R-2005-6. SWOV, Leidschendam. [in Dutch] Janssen, S.T.M.C (2007). De veiligheidsverkenner voor het wegverkeer. Risico's van wegen in de periode 1987-2005 met prognose voor 2006-2020. R-2006-35. Stichting Wetenschappelijk Onderzoek Verkeersveiligheid SWOV, Leidschendam. [To be published]. Kooi, R.M. van der & Dijkstra A. (2003). Some behavioural effects of non-compulsory (bicycle) lanes on narrow, rural roads. R-2003-17. SWOV, Leidschendam. [in Dutch] Kooi, R.M. van der & Heidstra, J. (1999). Effect of edge strips on traffic behaviour. R-99-25. SWOV, Leidschendam. [in Dutch]

SWOV Fact sheet Reproduction is only permitted with due acknowledgement

5

© SWOV, Leidschendam, September 2007

NMB (2004). Document Essentiële Herkenbaarheidkenmerken Duurzaam Veilig. TT02-084, versie 15. In opdracht van CROW uitgevoerd door Traffic Test BV. Nationaal Mobiliteitsberaad. RVV (1990). Reglement Verkeersregels en Verkeerstekens RVV . Staatsblad van het Koninkrijk der Nederlanden, No. 459. 's-Gravenhage. Schoon, C.C. (2000). Verkeersveiligheidsanalyse van het concept-NVVP, Deel 1: Effectiviteit van maatregelen. D-2000-9I. SWOV, Leidschendam. [in Dutch; no English summary] SWOV (2007). Recognizable road design . SWOV-factsheet, mei 2007. SWOV, Leidschendam.

SWOV Fact sheet Reproduction is only permitted with due acknowledgement

6

© SWOV, Leidschendam, September 2007