Top: Home Page
Up: Table of Contents

germany1.gif (169 bytes)Auch auf deutsch erhältlich!
This page is a translation of the document (in PDF format)
Verkehrssicherheit in Einbahnstraßen mit gegengerichtetem Radverkehr

Traffic safety on one-way streets
with contraflow bicycle traffic

Alrutz, D./ Angenendt, W./ Draeger, W./ Gündel, D.
In: Straßenverkehrstechnik, 6/2002

Translation by John S. Allen
who accepts responsibilty for any errors of translation
Posted by permission

1. Existing conditions

Urban one-way streets serve primarily to channel and regularize motor traffic. For bicycle traffic, on the other hand, they interrupt many direct connections and make the use of low-traffic local streets more difficult, leading bicyclists to travel illegally on one-way streets in the wrong direction, or to move over to parallel arterial streets.

The topic of opening one-way streets to contraflow bicycle traffic was controversial for years in Germany (see. DRAEGER 1997). Questions about the effect on traffic safety and of the proper approach under the traffic laws led local governments to various solutions, or generally to avoid contraflow installations.

Contraflow sidepaths and, less often, bike lanes, have, however, been installed in some places for decades. In many places, the "false one-way street" solution has been most widely used, mainly to avoid problems of conformity with the traffic law. Before 1997, some cities nonetheless opened true one-way streets to contraflow bicycle traffic, at least in some selected districts, by way of exceptional applications of the traffic law. Figure 1 illustrates these.

These developments, and the mostly positive experience to date, led to a provision in an amendment to the traffic law (StVO) that went into effect on September 1, 1997, allowing contraflow bicycle travel on one-way streets under certain conditions.

The provisions in the StVO and in rules for application of the StVO (VwV-StVO) for opening of one-way streets were, however, not put into effect until December 31, 2000, due to lack of sufficient research support. Experimental evidence was to be collected to determine whether, all in all, and particularly as they affect traffic safety, such measures are justifiable.

Figure 1: Main ways of opening one-way streets to contraflow bicycle traffic

1. Sidepath fig01a.gif (3684 bytes)
2. Bike lane fig01b.gif (4036 bytes)
3. Bicycle street fig01c.gif (2867 bytes)
4. False one-way street fig01d.gif (2364 bytes)
5. One-way street as in the 1997 amendment to the traffic law fig01e.gif (3482 bytes)

The goal of the research plan of the Federal Road Research Institute (Bundesanstalt für Straßenwesen), completed in 2000 (PGV/BIS 2001) was to look more deeply into the issue of traffic safety on one-way streets where contraflow bicycle traffic was permitted according to the new StVO provision, and to evaluate these streets in comparison with other one-way streets where contraflow bicycle traffic was prohibited. In this way, a well-founded decision could be reached as to whether the experimental provisions could stand after the year 2000, and whether modifications to them should be recommended.

In what follows here, the most important results and conclusions from the research are reported. These have already led to a permanent status for the StVO provision.

2. Research phases

  • An initial inquiry among cities with more than 20,000 inhabitants provides the first overview of the application of the amendment to the law and how it is received by the municipalities.
  • All crashes involving a bicyclist on a one-way street in 1999 in several provinces and additional selected regions are examined. A special collection of crash data and additional information about how crashes occurred is sent directly by local police departments.

    611 crashes on one-way streets were recorded, of which 436 were on local streets, which are of the greatest interest in this inquiry. The crashes were categorized according to the direction of travel of the bicyclist and the traffic laws that apply on the one-way street, and were evaluated in terms of the most important characteristics (e.g., outcome, other party involved, location, type of crash, age of the bicyclist etc.).

  • 15 cities were then chosen based on the results of the inquiry. One-way streets were identified and categorized according to their cross-section, parking arrangements etc. The crash records for a total of 669 streets were collected and correlated with the characteristics of the streets. In this way, the distribution of crashes, as well as other information, was made available, and one-way streets with crashes could be compared with others that were crash-free in the same city (and where the same legal provisions had been applied). A before-and-after comparison of crash incidence in 1997-1999 also was carried out on the selected one-way street segments in the cities which were opened as in accordance with the StVO provision. Bicycle counts according to direction of travel and the bicyclist's position in the corridor were used to extend the analysis.
  • Using traffic analysis methods, the traffic and interaction sequences were determined, and it was established which constellations of problems and safety risks occur with different configurations for contraflow bicycle traffic on one-way streets. Twelve situations were chosen in which bicycle and motor traffic were heavier and an increased likelihood of problems could be anticipated.
  • The results of all of these steps were compiled, and conclusions were drawn regarding traffic safety in relation to the legal, traffic and geometric conditions on the one-way streets.

Recommendations for amendments to the law and for measures that might be taken in order to make contraflow travel on one-way streets safe, were discussed with experts, i.e., in the context of work of committees of the Road and Traffic Research Society (Forschungsgesellschaft für Straßen und Verkehrswesen).

3. Conclusions of research

Amendment to the law and its reception by the municipalities

During the research phase, many cities and towns were wary of opening one-way streets. In part, they feared having to undo measures that had already been put into place, in case the new provision was not made permanent. But there were other reasons as well:

  • The burden of proof required by the rules for application of the traffic law, and in some cases the costs and staffing associated with them, were regarded as too troublesome.
  • Cities which had already installed false one-way streets before the 1997 amendment to the law mostly kept them as they were, to avoid the cost of new signage.

Many municipalities began to open one-way streets in the network of local streets immediately after the amendment. An area-wide opening, however, was only carried out in one case.

The documentation and evaluation of traffic and crash situations required in the rules for application of the traffic law are only rarely fully carried out by the authorities responsible for the streets. Only one example of withdrawal of an opening of a one-way street because of crashes involving contraflow bicycle traffic is known.

Results of the counts and behavioral observations

Depending on their location in the street network, one-way streets in 30 km/hour zones carry as many as 220 motor vehicles and 140 bicyclists per hour. As a rule, however, the traffic volumes are significantly lower.

In the one-way streets which have been opened, 40 – 45% of the bicyclists, on average, travel in the contraflow direction. This quantity is not significantly lower in the one-way streets which have not been opened; however, many more bicyclists traveling in the contraflow direction use the sidewalks (60% as opposed to approximately 20% in opened one-way streets, see figure 2). For pedestrians, then, there were markedly fewer restrictions and hazards on the sidewalk. Opening a one-way street also can shift bicycle traffic from main streets to local streets.

Figure 2: Distribution of contraflow bicycle traffic
with respect to traffic rules and location of bicyclists.

fig02.gif (4054 bytes)

Motorists reduce their speeds significantly when encountering bicyclists if a street is less than 3.50 m wide. Even when the street is narrow, the encounters are problem-free, because visual contact is good. When the streets are wider, the shy distance between bicyclists and motor vehicles increases, but the motor vehicles also travel faster.

Critical conflicts between motor vehicles and bicyclists in the one-way direction are more frequent than with bicyclists in the contraflow direction. Besides overtaking maneuvers, stationary vehicles are the most important cause of problems. Pedestrians, on the other hand, more commonly get into critical conflicts with bicyclists in the contraflow direction. These conflicts were, however, mainly caused by the pedestrians, who came from the right out of view of the contraflow bicyclists and failed to notice the bicyclists when crossing. These problems are most common in streets heavily used for business trips.

Behaviors of motorists and bicyclists at intersections with priority to the right range from a very cautious approach to a bold entry and passage through the intersection without any reduction in speed. Some turning movements involve "cutting corners", even when sight conditions are poor. Motor vehicles illegally parked near the intersection sometimes worsen the sight conditions.

Results of the crash analysis

In general, very few bicycle crashes occur in zones with a 30 km speed limit. More than 80% of the 669 streets had no crashes during the 3- to 4-year observation period. Only 3% of the streets had more than one crash (Figure 3). In the comparison between the one-way streets opened as provided in the StVO (669 segments) and those which were not opened, the crash densities were equal. The crash densities for one-way streets in 30 km/h zones were below those for nearby two-way streets (Figure 4). Following the opening, the crash densities even decreased slightly (Figure 5). The severity of the crashes also decreased. It was no worse for contraflow bicycle traffic on streets opened as provided in the StVO than for bicycle traffic traveling in the one-way direction.

Figure 3: distribution of crashes on the streets
in the test area during the 3- or 4-year
experimental period

fig03.gif (3229 bytes)

Figure 4: Crash densities with respect to the type of installation

fig04.gif (5835 bytes)

Figure 5: Before-and-after comparison of crash numbers

fig05.gif (3491 bytes)

Crashes in which bicycles and motor vehicles are traveling in opposite directions are rare on one-way streets opened as provided in the StVO, but represent a larger fraction on one-way streets which have not been opened. Safety problems were still most common at intersections (including those where the rule of priority for traffic on the right applies), involving motor vehicles turning onto or crossing the one-way streets – often in connection with sight lines restricted by parked vehicles. Motor vehicle drivers here appear often not to have anticipated bicyclists traveling opposite the one-way traffic. Parked vehicles between intersections have no recognizable effect on safety of contraflow bicycle traffic. Bicyclists traveling with the one-way traffic have a larger proportion of crashes involving parked vehicles. Crashes on the sidewalks, and generally with pedestrians, occur less often when one-way streets have been opened, and so the safety (and the quality of travel and of waiting) for pedestrians is improved.

The percentage of children and youth (up to age 17) involved in crashes while riding in the contraflow direction is significantly higher on streets which have not been opened. On streets in business districts, the greater traffic volume results in more problems for bicyclists, regardless of their direction of travel. Here, particular emphasis should be placed on appropriate speeds for motor vehicles.

All in all, the opening of one-way streets has no negative effects on traffic safety. The trend in crash numbers, densities and severity is even somewhat positive. Additional positive effects can be expected due to the shifting of bicycle traffic from arterial streets to the network of local streets. One example (city of Frankfurt am Main 1998) includes detailed before-and-after observations, and shows that the crashes involving contraflow bicycle traffic after the opening are concentrated on a short stretch of one major street in a business district, the Leipziger Straße, while very few crashes occur elsewhere in the test area (Figure 6). In this connection, it should be noted that the motor vehicle and bicycle traffic volume is highest by far on the main, business street, and that the contraflow bicycle traffic increased from approximately 250 to 400 bicyclists in a 6-hour period, an increase out of proportion to those in other study areas. Including the main streets, the total number of bicycle crashes in three residential study areas decreased (Figure 7). This decrease is indicative of the desired shift of bicycle traffic from the main streets, with their safety problems, to local streets.

Figure 6: Bicycle crashes in Bockenheim
before and after opening of one-way streets
(3 years each), from City of Frankfurt am Main, 1998.

[Note: the barely-legible legend at bottom of each image reads:

Crashes on bordering streets
Crashes in the research area
Contraflow crashes on one-way streets]


fig06a.gif (17090 bytes)


fig06b.gif (15764 bytes)

Figure 7: Change in crash types in the three test areas
in Frankfurt am Main, from: City of Frankfurt am Main 1998

fig07.gif (5455 bytes)

General notes on crash analysis

Some problems also of interest for other work on traffic safety occurred in connection with the crash analysis, and they will be noted here:

  • The bicycle-related content of the reports by local police departments differs. This is even more true of the tables in computer databases [EDV]. It is therefore often not possible to compare even basic information such as that about severity of crashes. More standardization would be helpful in order to allow comparative and evaluative analysis.
  • When crashes are entered into a computer database, sometimes indexes by street are not created, and the locations of crashes are no longer related to other information. It therefore becomes necessary to conduct a laborious referral to the original crash reports. Computer tracking of crashes also should allow rapid access to crash data for selected streets, for the sake of deeper analysis.
  • Crashes on one-way streets may be impossible to identify due to the lack of a corresponding indication in the crash data, as there commonly is no list of one-way streets in a city.

4. Consequences of the opening of one-way streets

The research shows that, at least, the opening of one-way streets to contraflow bicycle traffic has no negative effects on traffic safety. A tendency toward positive effects can even be shown. Even an area-wide application of the rule in 30 km/hr speed limit zones leads to no additional problems. A simultaneous opening of one-way streets in a section of a city also promotes efficiency in a public awareness campaign (Figure 8).

Figure 8: Pamphlets about one-way street opening –
Aachen, Brühl, Dorsten, Mainz

fig08.jpg (7408 bytes)

The main consequences and recommendations of the research report are:
  • Measures may be taken to improve existing problem spots; before-and-after testing is in order in each case.
  • A speed limit of 30 km/h will be a prerequisite for an opening according to the provisions of the StVO. Narrow alleys, in some cases less than 3 meters wide with low traffic volume, need not, however, be excluded, as long as locations for passing (for example, merging off the roadway, or empty spaces between parked vehicles) exist or can be created and/or the probability of a head-on encounter is very low due to the shortness of the one-way segment or low traffic volume (Figure 9).

If the roadway is wider (over 3,75 m) and motor traffic volume is higher, separating stripes may also be used, as long as a sufficient distance from parked vehicles can be maintained.

Figure 9: Marking of a passing location for head-on encounters
on a one-way street with a narrow travel lane (Köln).

fig09.jpg (6972 bytes)

  • Curves tend to lead to problems if the contraflow bicycle traffic is on the inside of the curve and motor traffic may cut across the bicyclists' path. The most important measure to improve sight lines is to prevent parking on the inside of the curve, sometimes in connection with markings including arrows indicating the direction of travel, or shoulder or bike-lane stripes. If sufficient roadway width is available, a bike lane may be separated by bollards or curbs (Figure 10).

Figure 10: Measures to increase safety of contraflow
bicycle traffic on a curving roadway (Hannover)

fig10.jpg (10151 bytes)

  • Intersections are the most important potential problem locations. Assuring sufficient sight distances and making it clear that contraflow bicycle traffic will be approaching are of special importance. In addition to the usual prohibition of parking near intersections (§ 12 StVO), clearly-recognizable signage indicating the opening for contraflow bicycle traffic is important for all travelers entering the intersection. In special cases, marking of exit or entry locations for contraflow bicycle traffic can clarify the situation (Figure 11). Structural elements to improve safety at the entry and exit locations may be considered where bicyclists must yield right of way or motor traffic is heavier, as the risk of motor vehicles' cutting corners is high here (Figure 12).

Figure 11: Marking for a slip lane with a stop line
for contraflow bicycle traffic at an intersection
with priority for the vehicle on the right (Köln)

fig11.jpg (8155 bytes)

Figure 12: Structurally separated exit location for
contraflow bicycle traffic transitioning to a street
which has priority for traffic on the right (Bremen)

fig12.jpg (8693 bytes)

5. Legal Situation

Since January 1, 2001, the possibility of opening one-way streets to contraflow bicycle traffic has been incorporated without limitation in the StVO. Additional changes, particularly concerning a stiffening of the requirements in the rules for application of the StVO, are presently under discussion and will be addressed in a forthcoming amendment to the law. At present, the following changes are anticipated:

  • Traffic law [Straßenverkehrs-Ordnung (StVO)]

    In place of the sometimes misunderstood supplementary sign with the vertical arrows, used with the horizontal-arrow one-way street sign 220 (Figure 13), a sign with the same intent but with horizontal arrows is anticipated, and it is already included in the forthcoming amendment to the law, "Weniger Verkehrszeichen" (reduction in the number of traffic signs). Sign 353 will entirely go out of use following this amendment, and signs for opened one-way streets will be as in Figure 14.

Figure 13: Possibly-confusing combinations of signs
at an intersection. The arrows on the supplementary sign
point to the cross street, a one-way street which has
not been opened to contraflow bicycle traffic.

fig13.jpg (10529 bytes)

Figure 14: Recommendation for signage of one-way streets
on which contraflow bicycle traffic is permitted

fig14.gif (3386 bytes)

  • Rules for application of the traffic law [Verwaltungsvorschrift zur Straßenverkehrs-Ordnung (VwV-StVO)]

    A considerable simplification is intended in the rules for sign 220 in the rules for application of the traffic law for contraflow bicycle traffic. The following requirements for opening one-way streets will be imposed:

    - The speed limit will still be 30 km/h or less.

    - There must be sufficient width for passing, except for short narrow segments. There will be no specified narrowest dimension, other than 3.50 m for streets with significant truck or bus traffic.

    - The traffic pattern should be clearly laid out. If it can not be, a partial separation might be considered.

    - Where the location and traffic require, a safety space should be created for bicycle traffic.

    - Sign 220 with the supplementary sign should be placed at the beginning of the one-way street and at all streets and entrances, so that it is visible from all vehicles entering the one-way street.

In the future, some current requirements in the rules for application of the StVO will be discarded, for example that for area-wide bicycle traffic planning, testing of other options such as false one-way streets, short stretches for head-on encounters, and pre-evaluation of stopped vehicles. The requirement for evaluation of traffic and crashes that existed in the test phase has already been eliminated. These steps result in increased flexibility for jurisdictions to open one-way streets. The opportunities to open them are broadened, and the process is to do it is simplified. .

6. Summary

As the appeal of bicycling can be increased by opening one-way streets without negative effects on traffic safety or on other travelers, there need be no concern about wider application of the rule in the municipalities. The solution is generally advantageous for the users and the municipalities:

- Bicyclists can access residential areas area-wide and without the need for detours.

- Bicyclists have an increased ability to avoid arterial streets. That is safer, and generally more pleasant as well because of the reduced impediments from motor traffic.

- Through connections in the bicycle travel network are easier to achieve. However, attention must, as a rule, be paid to safety of crossings of arterial streets.

- The solution can be implemented quickly and relatively cost-effectively.

The proposed changes in the rules for application of the traffic law make it easier for municipalities to open one-way streets. With continuing positive experience in practice, a future reduction in the need for signage, and a general opening of one-way streets in connection with the 30 km/h speed limit rule (Z 274.1 StVO) may be considered, excluding only specific problem streets.


PGV/BIS (Planungsgemeinschaft Verkehr [Traffic planning collaborative], Hannover/ Büro für integrierte Stadt und Verkehrsplanung [Office for Integrated urban and traffic planning], Bonn): Verkehrssicherheit in Einbahnstraßen mit gegengerichtetem Radverkehr [Traffic safety on one-way streets with contraflow bicycle traffic], Berichte der Bundesanstalt für Straßenwesen [Reports of the Federal Road Traffic Institute] vol. V, 83; Bergisch Gladbach 2001

Alrutz, D./ Stellmacher-Hein, J.: Sicherheit des Radverkehrs auf Erschließungsstraßen [Safety of bicycle traffic on local streets], Berichte der Bundesanstalt für Straßenwesen, Heft V 37; Bergisch Gladbach 1997

Alrutz, D./ Tebbe, H./ Willhaus, E.: Führung des Radverkehrs in Einbahnstraßen [Accommodation of bicycle traffic on on one-way streets], in: Handbuch der kommunalen Verkehrsplanung [Handbook of Municipal Traffic Planning]; Bonn 1999

Draeger, W.: Die StVO-Novelle - Konsequenzen für die planerische Praxis [Consequences for Planning Practice], in: "Straßenverkehrstechnik" [Traffic Engineering]; vol. 12/1997

Stadt Frankfurt am Main: Modellversuch „Radfahren gegen Einbahnstraßen" [Prototype test of contraflow bicycling on one-way streets]; Frankfurt/Main 1998

Addresses of the authors of the present document:

Dipl.-Ing. Dankmar Alrutz, Dipl.-Ing. Detlev Gündel,
Planungsgemeinschaft Verkehr,
Große Barlinge 72 a,
30171 Hannover;

Dipl.-Ing. Wilhelm Angenendt, Dr. Werner Draeger,
Büro für integrierte Stadt- und Verkehrsplanung,
Meckenheimer Allee 67-69,
53115 Bonn

Top: Home Page
Up: Table of Contents