Complementary ways of assessing network's accessibility

In a large meaning, accessibility can be defined as the capacity to access the best way to (a set of) points or destinations on a given territory. It also relies to mobile persons or objects and to the existence of resources, facilities, amenities and points of interest. Beyond this definition, many linked aspects and issues of this property can be addressed. They all mix together to provide a more or less “good accessibility”.

A first aspect concerns the spatial network structure which supports the movement. Made of sections and intersections, a road network or a transport network can be designed as a mathematical graph of connected edges and nodes. Its structure and topology strongly influence the accessibility level, because it imposes a spatial footprint to the city and to its inhabitants and obviously determines their capacity to explore their surrounding environment. We will illustrate this fact with a few examples, that clearly show how much weights the network organisation in the access efficiency. This will reveal that the network design is one of the key levers to deal with urban mobility, giving the planners a determinant role in city design and sustainability. On the other hand, this will leave only a rather narrow degree of freedom for people to move on this constraining structure, even using new mobility services that can help in providing congestion information in real-time. In this research field, some theoretical issues from other disciplines and models, such as Wardrop equilibrium and Braes Paradox, will once again highlight the strong impact of the network on mobility location and spreading, especially in the context of urban sprawl.

The second question involves the way this topological structure can be handled and what different methods and metrics are able to measure accessibility on it. Depending on sites history and topography, but also location (especially which continent is considered), urban networks can show slightly different patterns. From Manhattan networks to radio-concentric networks and even to fractal networks, one can observe a huge variety of topological patterns. All of them should have a rationality possibly based on specific accessibility purpose. Moreover, since Human uses to concentrate in cities and to group facilities in the same vicinity, space is no longer isotropic and there obviously appear hot spots of attractiveness in the town, such as transport stations, supermarkets, city centres, etc., that polarise houses, population and activity. To locate those facilities, different metrics can be used according to different objectives, from equity to efficacy. All of them induce a particular impact of the metric on the centre location and sustainability and also a spatially differenciated influence of the demand points (population) on a given facility. Those metrics belong to graph theory, operations research and space syntax. A comparison of a few of these metrics will be done on several typical and theoretical networks. It will emphasise how much sensitive can the accessibility measurement be to chosen metrics and will enhance the necessity to use mathematical indices with a particular caution keeping in mind possible robustness deviations.

A third issue is thus related to the people who use this network. Indeed, what could be the interest of a good network accessibility if nobody gets profit from it? This means that accessibility can not only be assessed using geometrical or mathematical metrics. A good accessibility must serve the people the best way. It is thence strongly related to mobility practices and urban rhythms. Homo Mobilis can have a peculiar behaviour during his motion, depending on how he perceives the urban environment, his habits and the reasons why he needs to move. This draws a framework of study that makes very relative the simple concept of shortest path in distance or time that is provided in all the GPS of the market. In reality, the best way to get to a given destination can strongly differ from an individual to another. For instance, commuters do not have any time to lose and will prefer to follow the shortest time route, if they are in a car, in a bus, or in any other public transport service. A pedestrian could choose the best way according to both the best walk ability and an ensured personal security. Blind people will carefully prepare there trajectory, by avoiding noisy places, transient roads and narrow pavements to minimise the risk, even if this criterion increases the walked distance, in a maximum bearable time. We will provide examples of such applications from works we made with students and an association of visually impaired people and guide dogs education.

A last concern about accessibility is currently related to the climate change and the imperious necessity to keep the air breathable and finally to save the planet. This will pass by the development of sustainable transportation systems that strongly encourage people to share their motion services, using different novel transportation means such as carpooling, carsharing and demand responsive transport. This leads to optimisation objective functions that explicitly include an environment protection purpose. In such a context, we will provide several results on optimal demand responsive transport and more recent research progress in carsharing. They show that one still has non-negligible possibilities to improve transport efficiency, keeping correct accessibility capabilities for people in cities. Another condition to reduce the air pollution with transport optimisation is indeed for people to lose a little time in their activities. Some reflections will be provided about this specific aspect of the society that somehow contradicts all the speed rules to gain in economic efficiency and productivity. But fortunately what is lost in time is gained in quality of life and sociability.

In conclusion, these combined aspects of accessibility suggest to comprehend this global concept by taking into account its different facets altogether, including several important dimensions that planners and scientists have to keep in mind: urban infrastructures sustainability, access equity and environment protection.