Name
Abstract | ||
|---|---|---|
Le Havre agglomeration (CODAH) includes 16 establis hments classified Seveso 3 with high threshold. In the literature, we construct vulnerability maps to help decision makers assess the risk. Such approach es remain static and do take into account the population displacemen t in the estimation of the vulnerability. We propos e a decision making tool based on a dynamic vulnerability map to evaluate the difficulty of evacuation in the different sectors of CODAH. We use a Geographic Information system (GIS) to visualize the map which evolves with the road t raffic state through a detection of communities in large graphs algorithm. The population of the Seine estuary is exposed to s everal types of natural and industrial hazards. It is included in the drainage basin of the "Lézarde " and is also exposed to significant technological risks. The modeling and assessment of the danger is useful when it intersects with the ex posed stakes. The most important factor is people. Recent events have shown that our agglomerations are vulnerable in fro nt of emergency situations. The examination of impa cted populations remains a difficult exercise. In this c ontext, the Major Risk Management Direction team (DIRM) of Le Havre Agglomeration (CODAH) has developed a model of spatial and temporal population exposed allocatio n PRET- RESSE; the scale is the building (Bourcier and Mall et 2006); thus by distinguishing their day and nigh t occupation. The model was able to locate people during the day both in their workplace and their residence (the unempl oyed and retirees). Although the model was able to locate th e diurnal and nocturnal population, it remains stat ic because it does not take into account the daily movement of people and the road network utilization. For a better evacuation of people in the case of a major risk, we need to know the state of road traff ic to determine how to allocate the vehicles on the road network and mo del the movement of these vehicles. In fact, the pa nic effect of some people can lead to accidents and traffic jams, whic h may be too grievous with a danger that spreads qu ickly. The panic generally results from the lack of coordination and dialogue between individuals (Provitolo, 2007). In the literature, several models were developed to calculate a score of the vulnerability related to the road network utilization. This score may depend on social, bioph ysical, demographical or other aspects. Most of the se models adopt a pessimistic approach to calculate this vulnerabilit y: this case is met when a group of people in a haz ardous area decide all to take the same route to evacuate this area, w hich unfortunately happens quite often in the real world evacuation situations. Although it helps decision-makers to es timate the risk by a census vulnerability map, this approach remains static and does not take into account the evolution of the road network traffic. In this paper, we have to simplify the representati on of the population displacement, which is a compl ex phenomena. We also propose a dynamic and pessimistic approach related to the access to the road network. To this end, we model the road network by a dynamic graph (the dynamics i s due to the traffic evolution). A simple model bas ed on traffic flow will also be proposed. Then, we apply a self-o rganization algorithm to detect communities on the graph belonging to the collective intelligence algorithms. The algo rithm allows us to define the different vulnerable neighborhoods of the agglomeration in the case of an evacuation due to a potential danger, while taking into account the ev olution of the road network traffic. The result of this algorithm will be visualized into a GIS on a dynamic vulnerability map which categorizes various sectors depending on the diffic ulty of access to the road network. The map will he lp decision- makers in a better estimation of risk in the commun es of the CODAH. It will enrich PRET-RESS static model developed at the CODAH, taking into account the mobility of the population. |
Year | Venue | Keywords |
|---|---|---|
2009 | Clinical Orthopaedics and Related Research | dynamic vulnerability map,dynamic graph,traffic flow.,self organization,hazard,gis,decision making,communities detection,decision maker,risk management,collective intelligence,traffic flow,geographic information system |
Field | DocType | Volume |
Graph,Data mining,Population,Geographic information system,Economies of agglomeration,Computer science,Road traffic,Vulnerability | Journal | abs/0911.1 |
ISSN | Citations | PageRank |
International Symposium on Risk Models and Applications, Ki\`ev :
Ukraine (2008) | 0 | 0.34 |
References | Authors | |
3 | 5 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Michel Nabaa | 1 | 0 | 0.34 |
| Cyrille Bertelle | 2 | 81 | 20.42 |
| Antoine Dutot | 3 | 20 | 4.53 |
| Damien Olivier | 4 | 42 | 6.95 |
| Pascal Mallet | 5 | 1 | 0.77 |