Navigation – Plan du site
11ème Colloque Européen de Géographie Théorique et Quantitative, Durham, Royaume-Uni, 3-7 septembre 1999

daily mobility and adequacy of the urban transportation network a gis application

La mobilité quotidienne et l'adaption du réseau de transports urbains

Véronique Mondou


L'objet de cet article est de définir avec précision la population desservie par un réseau de transports en commun. L'analyse prend en compte à la fois le niveau spatial et social. Des inégalités sont observées dans l'accès au réseau par la population en fonction de critères liés à la distance au centre mais aussi en fonction du profil socio-professionnel de la population.

Haut de page

Texte intégral

1Mobility plays today an essential part in the urban structure. It is one of the main action of the development of the city. Motorized transport meets the main part of the demand but cars being preferred, the use of public transport is stagnating or decreasing. However, urban transport has a specific role. On the one hand, it is marginalized and use by a part of the population as a last resort. On the other hand, the political authorities give it priority over cars. It even has a great influence on urban planning and a social role since it allows everyone to move.

2The main idea is to confront the urban transport and the mobility of the population. Is there an homogeneous service on the whole of the officially covered territory ? Does it offer the same possibilities for all to move within the urban space ? Does it favour a specific population group ?

3To try to give an answer to the questions we used a GIS. The GIS presents various interests for this analysis on urban transport. Among the functions of the GIS, some have been used.

4The survey has been done in Rouen (France) and the surrounding towns (32 towns). This area corresponds to the perimeter where the intra-urban network is located (figure 1).

5The network is made of 32 bus routes and two tramway lines. Outside this perimeter, there is a coach service but it has a very low frequency and there is a railway network.

6One of the properties of the GIS is the possibility to combine several layers of information to create a new information. The objective is to determine which population group are best serviced by the network.

7We started from several layers of information : on the urban transport network, we have only used bus stops and not bus lines (figure 2). On this figure, we can see the localization of the 900 bus stops that constitute the network. They are distributed over on the whole territory, with the exception of woods.

8Then, we took the hypothesis of transport companies : a person is ready to walk for about 300 meters to get to a bus stop. Around every bus stop, buffer-zones with a 600 meter diameter have been defined.

9In the centre, where they are separated by less than 300 meters, they form a continuum. In the suburbs, they have a linear form where the bus stops don't part from the main roads. There are only a few isolated areas.

10This information has been cross-analysed with the population localized at the level of the enumeration district so as to know the attraction area of the network. The enumeration district is the smallest statistic unit available (Insee). Usually, the diagnosis of the transport organism rests on the analysis of the entire network, based on indicators with no spatial reference (such as number of kilometers covered, number of persons carried, for example). At the best, the analysis uses the level of an urban district and considers that the population is served when there is at least one bus stop in the urban district. But, it's not enough for the people to have a few bus stops, they have to be where they live. In reality, the whole population is not served in the same conditions.

11We consider that the population living inside the buffer-zone is served by the public transit, and that the population living outside is not served (figure 3). This figure shows the process that allows us to define the population served. The buffer-zones are determined from the bus stops and then they are crossed with the population density known at the enumeration district level. We obtain two new informations : the areas served shown in red shades and the areas that are not considered as served in blue shades.

12A double reading of the map is necessary to understand. First, we can isolate the population served in red and the population not served in blue. The colour gradations allow us to visualize the low densities (light blue/light red) and the high densities (dark blue/dark red).

13The second criterion allowing us to define the quality of the network is the frequency of the bus services. We will examine the effects caused by these factors.

14The first distinction (distance to the bus stop) corresponds to the delimitation of the buffers explained before (figure 4). If we take all the bus stops, the areas in red, the served areas predominate. It is 58 % of the territory that is served but, most important, 89 % of the population.

15This criterion doesn't have an important discriminating effect : the whole population is served. So we will test the second constraint, the frequency of the bus services. We used the timetables and we divide the time range (last departure minus first depart) by the number of bus. So we obtain a frequency for each bus stop. Three categories of bus stops have been established. Bus stops where busses run :

  • at least every 15 minutes,

  • at least every 30 minutes,

  • every 30 minutes and more.

16We can consider that an average wait of 7 minutes and a half (the first category) shows a break between two uses of the network : beyond a maximum of 15 minutes, people have to know the timetables in order to minimize their wait. The study of several frequencies (figures 5 & 6) shows a different vision of the service of the urban area. The map of the population served at least every 30 minutes doesn't present important differences. 79 % of the population is served. But, the examination of the population served at least every 15 minutes shows a lot of differences. It is only 21 % of the territory that is served and only 57 % of the population. Several areas appear in dark blue, that indicates high densities.

17The served areas are limited to the downtown area : since, in Rouen, the bus routes are radial or diametral and they converge towards the urban core. Around the town centre, the bus routes follow the main roads, which depend on the relief. The valleys are a constraint on the right bank, whereas there is none on the left bank. The distance to the centre seems to be the first criterion to explain the variation of the served population. Beyond 8 km from Rouen, the frequencies of the bus services are superior to 15 minutes. However, we can see areas near the centre which are not very well served, for example, the northern plateaus : Mont-Saint-Aignan, Bois-Guillaume, Bihorel. To explain these differences, we have studied other factors. The analysis of the characteristics of the population (socio-professional categories, age, sex, unemployment, car ownership rate) give us some answers. We will take into account only the socio-professional categories, namely two of them : the workers and the executives (figure 7). Areas with population of very different profiles are distinguished according to the presence of one of these categories.

18The two maps show two opposite distributions : a very clear opposition appears, each one excluding the other. Workers live rather on the left bank, or along the former industrial valleys of the right bank. The executives are concentrated on the right bank : in the centre or on the northern plateaus. These are the areas we have already located, they do not belong to the best served sectors. So we put forward this hypothesis : the network favours specific types of population.

19The analysis of recent mobility surveys shows that according to some characteristics, people do not have the same practices regarding movement. Clerks are the first users of public transport, followed by workers and middle ranking executives. On the contrary, senior executives, craftsmen and shopkeepers represent a low percentage of the users. Urban transport services focuses on specific types of population because they represent numerous potential users.

20In order to confirm this hypothesis on the whole perimeter, we have compared the service for each of the socio-professional categories with the whole active population. If we only keep areas served at least every 15 minutes, it corresponds to a radius of 8 km around the centre of Rouen. To measure the distance, we have drawn rings every 1 km (figure 8). The 8 km circle is the limit of the area served at least every 15 minutes. Beyond, the frequencies are lower.

21Figure 9 presents the curve for each category. In black, we can see the curve for the active population. This curve slopes down. In the town centre, 97 % of the population is served, and 8 km away the proportion drops to 30 %. All the curves have the same shape, but we can see two groups of curves : one group on each side of the active population's curve.

22The first group is made up of curves for workers, clerks and middle ranking executives. In blue shades appear workers and clerks. On the whole perimeter, these categories seem to be better served than the others. In red or green, are shown the curves for middle ranking executives, craftmen and shopkeepers and senior executive, they are not so well served. Workers and senior executives are the two extremes : workers appear always better served than the active population when the service of senior executives is deficient.

23Beyond the 8 km limit, the whole population, regardless of social position or occupation, is not so efficiently serviced (cf. figure 7). To attract this population, time spent by people in the public transport should be reduced. In the urban core, public transportation vehicles drive on bus lanes ; being out of traffic jams, they can compete with cars. The representation by isochrons (curve of equal time) shows us the space accessible to the urban transport users or the motorists (figure 10). The common key highlights the variation of time between the two means of transport. The amplitude of time for the urban transport is 80 minutes with a maximum of 1h25 to reach the town of Belbeuf, the variation is only 20 minutes for cars. The space seems to be more "iso-accessible" by car than by urban transport. In some areas, busses can rival with cars, in the urban core and along the tramway lines, times are similar. But, when we go away from the centre, in term of time, the urban network is disadvantaged. Near the centre, we can estimate that time is superior by 10 minutes, but for the suburbs the deviation is more than 30 minutes. However, it has to be noted that people don't always take into consideration the time spent in the search of parking spaces and the fares involved. The urban transportation network has to offer advantages (vehicle cost, car-park fares and parking difficulties).

24The urban transportation services allegedly homogeneous in the urban area (every town has a bus stop) appears very differentiated ; we can highlight two types of population for which the intra-urban network is objectively deficient.

25The urban transport services depend on the distance and the characteristics of the population. Senior executives do not seem to be very well served in the 15 minute frequency areas. But beyond 8 km, we can see population which could be users (clerks and workers), they are served but with low frequencies.

26For these two categories, a change for other means of transport, mainly cars, seems inevitable. There are some risks to improve the service for these populations, in fact, these households own several cars and that does not represent a financial constraint above all for senior executive and for the second category, the car ownership rate increases over public transport.

Haut de page


BEAUCIRE F. (1996, a), Les transports publics et la ville, Edts Les essentiels Milan, Toulouse, 64 p.

CANEPA B., MATHIEU M. (1992), Les transports collectifs pour libérer la ville, Celse, Paris, 137 p.

ÉMANGARD P.H. (1994), "L’inadaptation de la ville aux transports collectifs", Transports Urbains, n° 82, janvier-mars, pp. 23-29.

FISHMAN R. (1990), "Metropolis unbound : the new city of the twentieth century", Flux, n° 1, printemps, Paris, pp. 44-55.

FLINK J. (1990), The Automobile Age, The MIT Press, Cambridge, 456 p.

GARREAU J. (1991), Edge City, Anchor Books, 548 p.

GIULIANO G., SMALL K. A. (1993), "Is the Journey t oWork Explained by Urban Structure", Urban Studies, vol. 30, n° 9, pp. 1485-1500.

NEWMAN P., KENWORTHY J. (1996), "Formes de la ville et transports : vers un nouvel urbanisme", Les Cahiers de l'IAURIF, L'automobile dans la ville, n° 114-115, Mai, Paris, pp. 99-109.

OFFNER J.M. (ss la dir. de) (1992), Les déplacements urbains, Edts La documentation française, Problèmes politiques et sociaux, n° 690, Octobre, Paris, 68 p.

PÉNY A., WACHTER S. (1999), Les vitesses de la ville, Editions de l’aube, La Tour d’Aigues, 195 p.

Haut de page

Pour citer cet article

Référence électronique

Véronique Mondou, « daily mobility and adequacy of the urban transportation network a gis application », Cybergeo : European Journal of Geography [En ligne], Dossiers, document 192, mis en ligne le 18 juin 2001, consulté le 19 juillet 2018. URL : ; DOI : 10.4000/cybergeo.990

Haut de page


Véronique Mondou

Laboratoire MTG -UFR des Lettres et Sciences Humaines Université de Rouen- 76821 Mont-Saint-Aignan Tèl. : 02 35 14 68 88

Haut de page

Droits d’auteur

© CNRS-UMR Géographie-cités 8504

Haut de page