X Международная студенческая научная конференция Студенческий научный форум - 2018

Abstract

When considered in complex urban environments, the improvement of road safety through urban design has to face two kinds of difficulties : First “safety” only constitutes one component of the whole urban design task, implying that by making street designs for safety reasons, one may stimulate some unexpected effects over accessibility, commercial activities, pedestrian practices and other urban uses... effects which may even be contrary to the initial purpose of road safety.

Second it is also very difficult to pass, even within the restricted framework of road safety concerns, from the elements of the diagnosis and preliminary studies to some design scenarios. Indeed, a “sum of safety misfits to be avoided” is not sufficient to constitute a street design whereas, in the contrary, a street design necessarily implies more components and decisions than the few parameters taken into account by any diagnosis. The analytical results of a diagnosis (be it a diagnosis of road safety, of commercial welfare, of environmental health ... or even the sum of all of them) clash with the essentially synthetic character of the design activity. This article will show how the "design models" tool may offer some insights and perspectives to remedy this situation.

THE DIFFICULT TRANSITION FROM SAFETY DIAGNOSIS TO DESIGN PROPOSALS

Before the implementation of a design, it is necessary, in order to correctly take safety matters into account, to perform an appropriate diagnosis. In France, a safety diagnosis forms part of a larger general process, which unfolds in three stages that are conducted prior to the actual design of the project:

1. Study of safety issues: what types of accidents occur (bodily, material, etc.), who are the victims

(pedestrians, cyclists, etc.) and at what location (schools, intersections, etc.)? What are the priority issues for infrastructure administrators?

2. Diagnosis: what factors related to behaviour or to the environment are involved in these accidents? What scenarios emerge from the analysis of accident reports?

3. Design proposals: how should the configuration of the existing project area be modified to reduce accident risks in the future? What are the design principles to be adopted?

Depending on the specific situation, all or part of the stages of this process will be carried out.

In the case of a new project, only the third and last stage applies: there is no existing situation and hence no particular accidents to be analysed. It is thus possible, during all the project phases, to simply evaluate the impact of design decisions and choices on safety so as to avoid the creation of

new configurations known to be dangerous. It is this process that is implemented in connection with the Control of Safety relative to Road Projects (CSPR) on French national highways.

In the case of localised operations, after a significant accident for example, stages 2 and 3 will be carried out. The issues (victims, location and type of accident) are in fact already known: what remains to be determined are the circumstances of the accidents before going on to the design proposals. The analysis and understanding of accident circumstances (players, behaviours, interactions with the infrastructure, sociological context, traffic, etc.) enables classification into families (accident scenarios, associated with accidentological and aggravating factors). Project proposals will then consist of disabling the occurence of situations and behaviours similar to those having led to the studied accidents. It will thus be possible, for example, to prevent a litigatious u-turn if such an operation is found to be the cause of recurrent accidents.

In the case of urban design projects for a whole route or an entire district, the three stages of the

process must naturally be completed. The first stage devoted to the definition of issues takes on its full significance when the extent and complexity of the study area are such that a certain ordering of purposes is required: which publics have to be protected more particularly? Which places have to be handled with the greatest care?

In the latter case, as in the case of localised operations and new projects, the approach to road safety by urban design comes up against two difficulties, which are moreover encountered to a greater extent in the case of complex urban project (urban areas with multiple users, functions, spaces, traffic controlsystems, etc.):

(1)Safety constitutes only one component of urban design, so that there is a risk, when providing

“safety design” improvements, of causing unexpected consequences on accessibility, commercial activities, conditions for performing certain practices, etc. Some outcomes may not have been expected and could even be counter to the initial aims of safety. Such is the case, for example, when the design reserves part of the space for certain types of users in order to protect them: this protection builds the confidence of such users, which may lead to less attention being paid to others...

(2)It is moreover very difficult, even within the restricted framework of safety studies, to progress from the elements of the diagnosis to the design proposals. An enumeration of problems to be avoided may in fact not be enough to comprise a design project whereas, by contrast, a design inevitably entails more components and decisions than the few parameters allowed for in the diagnosis. Except in very simple cases in which it is decided to physically prevent contact between certain types of users (implementing traffic barriers, for example), a logical link is often difficult to establish between a spatial design and its propensity to provoke or to allow certain types of accidents.

The analytical results of the diagnosis thus come up against the more synthetic nature of design activity.

To solve the first difficulty, the study of safety issues and diagnosis are customarily complemented by other studies and preliminary diagnoses: parking, traffic, lighting, signing, environmental impact, pedestrian ways, commercial activities, etc. However, this overlapping of diagnoses in “layers” independent of each other further compounds the difficulty of progressing from diagnosis to design proposals: it is not, in fact, by extending the already long list of constraints, criteria and objectives to be “taken into account” that the elements of diagnosis will be integrated more reliably in the design. Quite the contrary, the disorganised multiplication of these prior observation elements, and standards of all kinds, rather tends to heighten the conceptual complexities affecting the ability of designers to fully use this knowledge and understanding of the existing situation.

In practice, the second difficulty is thus only partly resolved, and a sort of “conceptual jump” is observed between analysis results and the make-up of design proposals. Two types of elements nevertheless support this passage:

-A set of “road safety good practices:” road designs, which have proven, empirically, that they were not significantly accident-prone.

-A set of “safety design principles:” general rules assumed to have a positive effect on the lowering of accidents, such as “visibility” and “legibility.”

We shall thus see that the “design models” method developed within the framework of NR2C research enables significant advances in the resolution of these two difficulties: the incorporation of safety concerns among all the concerns underlying any transformation of the urban environment, on the one hand, and the solidity of the transition from diagnosis conclusions to design proposals, on the other.

We shall also see that these advances can be achieved by simple extensions of existing tools: by the methodical transformation of “good practices” and “design principles” as they are currently used today.

FROM “ROAD SAFETY GOOD PRACTICES” TO “URBANISTIC DESIGN MODELS”: CYCLEPATHS, ROUNDABOUTS, PEDESTRIAN PLATEAUS, ETC.

Knowledge of the danger level of specific design solutions is mainly empirical: what is known, from observations in a large number of cases, is that some types of design are likely to prevent accidents, whilst other types are, on the contrary, liable to cause some.

The accumulation of observations relative to a large number of accidents has shown the existence of recurrent and general themes which lead to the preparation of special solutions. Such is the case with regard to speed restrictions in town. It can take on the form of general specifications (lowering of speed limit to 50 km/h for example) or special designs which tell the driver what the appropriate speed is (traffic calming devices, narrower lanes). The conversion of a street into a 30-zone (an urban area where circulation speed is limited to 30km/h) combines these two aspects, among others.

The special treatment of intersections is another example of the application of empirical accident knowledge: intersections are crossflow points for traffic of different kinds and speeds (automobiles, motorcycles, cyclists and pedestrians). Conversion into a “roundabout” (to favour vehicle traffic) or a “raised plateau” (to favour soft travel modes) makes it possible to greatly reduce the gravity and occurrence of accidents.

The provision of street configurations specifically dedicated for soft travel modes (cyclists, pedestrians, reduced mobility persons, etc.) enables a reduction in the number of accidents involving them. The same applies to paths or lanes reserved for cyclists: “cycle paths on a widened sidewalk”, for example, protect cyclists from automobiles whilst also keeping pedestrians off the roadway.

Nonetheless, these designs, viewed from the road safety viewpoint (examples of “road safety good practices") may also be viewed as special occurrences of full-fledged “urbanistic models” whose properties, objectives and intentions can not be summarised by those concerning road safety.

The model of the “cycle path on a widened sidewalk" for example, separates, by its configuration, pedestrians from the edge of the roadway; it thus has a tendency to exclude the use of the sidewalk border for stopping, and steers pedestrian traffic towards the building side. This effect could be desirable in the case of a shop-lined street with attractive surroundings: shopowners will have nothing against the attention of pedestrians being turned towards shop windows. Along unappealing high buildings, by contrast, the application of this cycle path model will ultimately have a tendency to favour local and transit traffic, limiting stopping spaces since sidewalks borders usually constitute a privileged spot for this type of use.

In light of the high complexity of all possible causes and effects of urban phenomena, the contribution of “design models” consists in:

-Deciding to formalise “urbanistic models” in the general sense of the term, which do not favour in principle a given objective, criterion or discipline (safety, traffic, hydrology, lighting...).

-Distributing all these urban considerations “idea by idea:” as it is clearly impossible to group them all under a same general model which would be valid everywhere, the method consists, first, in

acknowledging the existence of “design ideas” or “design partis” (such as the cycle path on widened sidewalk) to group and to agregate, then, all knowledge elements and viewpoints that seem relevant for the understanding of the studied phenomena.

By progressing from the concept of “good practices” to that of “urbanistic models”, we bring in mechanisms and phenomena involved in the explanation of accidents into a more general understanding of the life of studied urban environments: depending on the models considered (cycle paths, bus lanes, sidewalks, arcades, avenues, boulevards, pedestrian ways, alleyways, bus shelters,etc.), safety issues will be mixed with those of aesthetics, water flow, parking, commercial activities, local residences, public transport, city image, etc.

On a first level, this passage from safety good practices to urbanistic design models allows a designer wishing to use a given model for safety purposes to determine its implications from different viewpoints: it enables him to work within a more general understanding of the urban environment while conceiving the place and role of the components he is handling. Doesn’t a “line of posts along the edge of a sidewalk” simultaneously represent, depending on the urbanistic model within which it is inserted, a means of preventing litigious parking, a system protecting pedestrians against vehicles, one less circulation lane for pedestrians, a rhythmic composition influencing the aesthetics of the street, a support for persons stopping on the edge of the road before crossing?

On a second level, this passage from safety good practices to urbanistic design models encourages us to consider that, in the final analysis, there is theoretically no design model that is “safer” than another but, rather, that each design model can be conceived of as a special means suited to certain types of situations, making it possible to deal with the issue of safety as an aspect of the general organisation of urban uses. It is not the “roundabout,” the “cycle path on widened sidewalk” or the “levelled street” (a street in which roadway and sidewalks are put at the same level) that are in themselves design solutions safer or less safe that others but, more accurately, it is the use that is made of these urbanistic models in certain contexts.

On a third level, this ongoing from safety good practices to design models considerably facilitates the transition from diagnosis to programmes and then to the definition of design proposals. While the “cycle path on widened sidewalk,” the “roundabout,” the “bus corridor,” etc. taken as safety “good practices” are rather abstract concepts, taking them not from this special viewpoint but as themselves, as several design ideas, makes them much easier to grasp. The fact that these ideas, once taken from the general viewpoint and not from the special standpoint of safety, offer both the capacity to express needs (draw the attention of passersby to shops and other businesses, provide cyclists with more space, etc.) as well as possibilities of solutions (create a reserved path, setup on sidewalk edge, for example) is a twofold advantage:

-Firstly, for the inhabitants, infrastructure owners and all non-expert persons, who customarily express their desires and objectives by means of terms refering to solutions (need for a roundabout rather than the slowing of vehicles, for a cycle path rather than good conditions of circulation for cyclists, etc.).

-Secondly, for the designers themselves, who handle, combine and deform naturally urbanistic ideas of this kind when imagining different ways of transforming an existing site and current functioning patterns.

Briefly stated, going from “good practices” relative to safety (or lighting, traffic, parking, etc.) to the study and definition of real “urbanistic models” means opting for the organisation of all available urban knowledges (including that related to safety) around ideas which can be worked during the definition of design programmes and projects.

In this respect, the method of design models (1) makes it possible for all these elements of know how to be effectively incorporated in the urban design processes and (2) constitutes a solid support for the improvement of the organisation, distribution and consolidation of the urban design process itself, thus considered as an iterative expressing of needs and design solutions.

FROM “ROAD SAFETY DESIGN PRINCIPLES” TO “URBANISTIC DESIGN OPERATORS”:

VISIBILITY, LEGIBILITY, COHERENCE, ETC.

The tools available for the improvement of road safety by design are not limited to good practices. Acertain number of more fundamental studies on the question of safety have in fact made it possible to support certain principles or certain general rules to guide designers in the definition of projects. These studies are based on a systemic approach to safety matters. They consider that the occurrence of accidents can not be summarised as the responsibility of one or more players. Accidents are regarded as complex phenomena involving interactions between various users and their environments at different scales.

On the local scale, it is the man-vehicle-environment interaction which is considered. An accident is the malfunctioning of this man-vehicle-environment system or of the pedestrian-environment system. In this context, three major concepts take part, among other things, in the functioning of the system:

-“Visibility:” which is obtained when a design enables users to gain access sufficiently early and

accurately to those part of the visual information of their environment that may allow them to adapt their behaviour in order to avoid causing an accident: pedestrians, in the proximity of a protected passage, must be afforded a sufficiently distant visibility to avoid vehicles as they cross the roadway.

-“Legibility:” which is obtained when a desgin enables users to understand those apsects of the

functioning of their environment that may allow them to adapt their behaviour in order to avoid causing an accident: types of permissible use, speed to be adopted, etc.

-“Coherence:” which is obtained when the elements of a design do not produce any contradiction in the reading of the information being conveyed that would be capable of causing accidents: a very wide roadway with multiple lanes in an unbuilt environment might appear contradictory, in the information it transmits to drivers, with signs indicating a speed limit of 50 km/h for example.

Nevertheless, while these concepts are meaningful from the viewpoint of the safety diagnosis, they

have only a rather limited impact when used in the design process for a relatively complex urban

project. When accidents are diagnosed, “visibility” and “legibility” will always be relevant explanatory factors, for as they were defined in this context, they will cover every possibility of the safety system malfunctioning:

-“Visibility for safety” consists in providing access to information which, if not transmitted, could lead to accidents.

-“Legibility for safety” consists in making comprehensible certain functioning aspects which, if not understood, could lead to accidents.

However, it would be possible to define other concepts of visibility and legibility of equal legitimacy from the viewpoint of urban design:

-“Visibility for commercial activity” could consist in giving access to information which, if it were not transmitted, could cause a drop in the sales of local businesses, for example: shops obviously want to make their display windows and signs visible.

-“Legibility for commercial activity” could consist in making understandable certain functioning aspects of locations which, if not understood, could cause a decline in the number of people who patronize local businesses: proper understanding of access and parking availibility, for example, are not negligible in this regard.

The same would be true of “visibility/legibility for aesthetic pleasure” or “visibility/legibility for waste treatment,” etc...

When an urban project is being designed, the concepts of legibility and visibility can no longer be used in a sectoral manner, because new forms are not imagined and conceived by simply verifying whether they are conform or not from the safety viewpoint (visible, legible, etc.), from the aesthetic viewpoint (visible, legible, etc.), or from viewpoint of shops, businesses, etc.. To perform these verifications, and the adjustments they imply, it is in fact necessary to have already conceived these forms, by means of concepts, which are not of the same nature than those used to verify their post-conformity.

In practice, designers also use these “visibility” and “legibility” concepts. However, these terms no longer have the same meaning: "visibility” and “legibility” are taken as “design operators” which may be specified, depending on circumstances, into different modes of interaction between the urban spatial arrangements on one hand and the uses they support on the other: any message may be communicated, for example, on the mode of visibility (size, recurrence, contrast, etc.) or on the mode of reliability, (honesty, uniformity, certainty, etc.).

A good part of the design work precisely consists in forming certain modes of interaction between the spatial arrangements of a city and its uses: in determining the ways these arrangements could support certain social organisations. It has been seen, for example, that the “mode of visibility” could relate to both road safety and health of commercial activities along a street. It is in fact possible for the visibility of shops to be contrary to road safety and, conversely, for visibility oriented to road safety to be detrimental to the visibility of shops and businesses. Many of our cities entrances experience, for example, over-visibility due to signs dealing alternately with commercial activities and traffic regulation.

This mode of “visibility” is however not the only way to ensure road safety or to enhance the appeal of shops and businesses. It would be possible to dedicate visibility to shops and businesses, for example, and ensure road safety through “reliability”: simplicity, regularity, predictability. Or, on the contrary, to confine visibility to road safety and ensure the health of shops and businesses through reliability:

parking places always available, extended opening hours, easy and regular access, etc.

While in the narrow system of road safety studies, visibility and legibility are always to be increased (they always tend towards improved accident prevention), this is absolutely not the case from the viewpoint of the urban design process which consists, on the contrary, in choosing (depending on the circumstances and on the prioritised purposes of people) certain modes, rather than others, of interaction between the urban spatial arrangements and their uses. Yet, visibility and accessibility are only some among other possible modes of organisation that make it possible to ensure road safety but also, as we have seen it, the health of businesses, traffic fluidity, aesthetic appeal, etc.

This is what differentiates, to our eyes, the road safety “principles” or “rules” from the “urbanistic

design operators”. Just as we went from safety “good practices” to “urbanistic models,” by defining design models which can be used to pass from diagnosis to design proposals, our approach has consisted here, with regard to general concepts, in going from this notion of “sectoral design principle” (useful for verifying the conformity of a project or for adjusting certain project configurations) to that of “urbanistic design operator”, useful for implementing and ensuring progress in the design process.

“ROAD SAFETY” AND THE “FIVE ELEMENTARY QUALITIES OF A STREET”

We shall now provide, in concluding this article, an example of an urbanistic design operator that we have developed in connection with NR2C research. Of course, what is involved here is just one among many other possible operators for dealing with safety questions, also making it possible to approach questions of ambience, image, commercial activity, etc. These operators are thus more general than urbanistic design models, which focus for their part on specific urban forms or design ideas of a specific nature (cycle paths, bus corridors, avenues, arcades, etc.). But they still make it possible to consider the articulation between different viewpoints, trades, intentions and purposes which are led to cohabit during the project design process.

We thus propose the operator of the “5 elementary qualities” which comprise a way to describe and to consider in a systemic manner the qualities and performances that society expects from streets and public spaces. This design operator seeks to ensure the grasping of not only the most important dimensions of the human experience of the urban environment, but also the relations which intrinsically link these dimensions to each other.

We base ourselves, unlike the analytical approaches, on the postulate that all the qualities of concern to us are, in urban areas at least, basically dependent on each other: our premise is that the “safety” and “reliability” of a street are not independent of its "animation" and of its "convivial" character, and that the “adaptable” or “embraceable” characteristics of a street are not independent of its “practical” and “legible” aspects.

We thus postulate that all the qualities that may be expected from a street or a public space, which are also different ways of using any spatial arrangement, can be formulated in the terms of 5 elementary qualities which are “vitality”, “reliability”, “firmness”, “accessibility” and “sympathy.”

These 5 elementary qualities are moreover organised so that there are two types of processes which explain their production:

1.A relation of generation (in a circle and clockwise): vitality generates reliability which generates firmness which generates accessibility which generates sympathy which generates vitality, and so on.

2.A relation of mastering (star fashion and clockwise): vitality masters firmness which masters

sympathy which masters reliability which masters accessibility which masters vitality, and so on.

It is important to understand that according to this systemic and qualitative mode of thinking, what is involved for each project or for each model is not the choice between “reliability”, “vitality”,

“accessibility”, etc.. What is involved, quite the contrary, is the understanding that if “vitality” is wished, then at least some “sympathy” is necessary while care must be taken to temper excessive “accessibility.” It must be understood again that if it is “accessibility” which is of interest to the infrastructure owner, some “firmness” is at least necessary and care must be taken to avoid excessive “reliability,” etc.

This design operator makes it possible to consider safety or security, for example, not as phenomena that may be measured in themselves but rather as properties of certain modes of organisation of streets uses: the vitality of the “levelled street” model (complexity, density) is not the vitality of the “multi-functional street” model (rapidity, simplicity); the reliability of an avenue is not the reliability of a boulevard; the sympathy of an arcade is not the sympathy of a fountain; and the safety of a buslane corridor is not the safety of a roadway open to all circulations.

Thus, no design model of a street or a public space is more “reliable” or more “sympathetic” than any another. Rather, the operator of the 5 elementary qualities makes it possible to imagine the different ways by which each of these design models can achieve “sympathy”, “reliability, “accessibility,” etc.

According to the diagram of the 5 elementary qualities, we find the hypothesis according to which

safety and security are elements of “reliability,” something everybody already knew. What then is the contribution of this kind of diagram for the analysis of these questions? The diagram of the 5 qualities in fact proposes 4 hypotheses on safety and security:

1.That safety and security are produced by vitality, i.e. that without vitality, safety and security are impossible to achieve. This is easily verified with some examples. It is known that pedestrian streets are perceived at night as less safe than streets handling automobile traffic. At least some animation, agitation and people (i.e. according to the diagram, at least some level of vitality) appear necessary so that the feeling of safety or security emerges from a situation. It is moreover well known that the presence of a crowd can offer more safety, to the extent for example that the crowd constitutes a sort of continuous mutual watch.

2.That safety and security produce firmness, that without safety or security, firmness and its related qualities (respect, serenity, calm, tranquillity, etc.) can not arise. These are things that appear verifiable, once again: the feeling of safety has a tendency to provoke, in people experiencing it, a certain serenity and a certain tranquillity.

3.That sympathy masters safety and security, that too much sympathy (familiarity, comfort, sweetness, softness) annihilates both safety and security. Once again, the reasoning appears correct: it is in the daily and familiar routes that most highway accidents occur, doubtless owing to a certain relaxing of attention of drivers who find themselves in a known area, in a situation which can become so comfortable that it will become dangerous.

4.That safety and security master accessibility, that too much safety and security annihilate any possibility of accessibility (rapidity, fluidity, ease, discretion, etc.). It is also possible to cite some examples which confirm this situation: the safest places are obviously not those which are the easiest to access, and it is evident that certain locations become inaccessible when a certain safety or security threshold is exceeded; it is also remarkable that safety devices slow the movements, and that they introduce checks which do not facilitate mobility, discretion or ease, which are all qualities related to accessibility.

What is important to consider, in the verification of these few hypotheses, is, on the one hand, each of these hypotheses in themselves, despite the fact that they appear verifiable most of the time in an intuitive manner by simple reasoning and, on the other hand, the articulation of these hypotheses with each other.

However, what must be especially kept in mind, from the design process viewpoint, is how the questions of urban safety and security are linked to other phenomena which are apparently distant but which are in fact directly related as factors or as products of these questions: that safety and security are produced by a certain vitality of the situation, and that they can suffer by excess sympathy, for example...

Let us take another one: what are the main components of the question of a buslane safety?

1. Firstly, a certain vitality, i.e. a certain minimum traffic on the buslane so that it does not appear deserted; it is in fact for such reasons that buslanes can be violated by automobiles or motorcycles when they can not resist the temptation to use this free space which is empty most of the time. A sufficiently increased use of the buslane will thus be a safety factor.

2. Secondly, care must be taken to avoid causing excessive sympathy, in particular in bus drivers who drive in a space which is dedicated entirely to them and which thus becomes relatively comfortable. As the space is reserved and configured for the buses, they can drive in full tranquillity and at a good speed which could, in certain cases, give rise to dangerous situations. What is important is not to give the bus driver the idea that the situation is very easy, likeable and comfortable, which the potential danger of any buslane is indeed.

Finally, a last example: what are the important components of the question of safety of a “levelled street”, i.e. a street designed with sidewalks lowered to the level of the roadway and on which pedestrians are induced to take over a large part of the roadway when circumstances allow?

1.Firstly, a certain vitality, i.e. in the case of a levelled street, a certain diversity of uses and of travel modes as well as sufficient density of use so that the situation appears to be vivacious and animated, and so that the different users pay attention to each other. What produces the safety of a level street is this necessary liveliness which keeps all the users on the alert and without which the levelled street, despite its lines of bollards, is rather less safe than an ordinary street with sidewalks.

2.Secondly, care must be taken to avoid excessive sympathy, in particular for pedestrians who are to be able to walk easily over the entire width of the street, while they are not allowed to stay anywhere along the street: the level street is designed to facilitate movements; it maintains a sufficiently low level of comfort to dissuade stopping or grouping.

Hence, that is how this operator of the “5 elementary qualities” makes it possible to formulate some solid hypotheses regarding phenomena which keep together the different qualities of a street.

We see clearly that this sort of tool is not usable to check whether a given design complies with safety standards, but rather to direct thinking along the lines of a logic of design, for the progressive definition of urban forms and uses modes of organisation which are to structure the transformation of one specific site.

Regarded both as bases and as complements to urbanistic “design models”, these “design operators” could contribute to the integration of knowledge, studies, diagnoses and issues of road safety into global urban design projects. The designs resulting from this kind of process may not involve improvements simply complying with safety rules, but configurations based, in part only, on the spatial and social organisation of road safety.

REFERENCES

NR2C – Deliv. 1.2: Specifications and preliminary concepts for the design of multimodal streets.

NR2C – Deliv. 1.3: Design models for the design of multimodal streets.

BRENAC Th. et al., Scénarios types d’accidents impliquant des piétons, Lavoisier Editions, 2003.

CERTU, Etudes de sécurité des déplacements en agglomération, 2006.

CERTU, Villes plus sûres, quartiers sans accidents, 1990.

CETUR, Sécurité des routes et des rues, 1992.

Код для цитирования: Скопировать