Macro-scale vulnerability assessment of cities using Association Rule Learning

Ismael Riedel (1) , Philippe Gueguen (1) , François Dunand (2) , Stéphanie Cottaz (1)

1 ISTerre - Institut des Sciences de la Terre
2 GEOTER - GEOTER

In this paper, a datamining method based on Association Rule Learning (ARL) is applied to define a vulnerability proxy between the elementary characteristics of buildings and the vulnerability classes of the European Macroseismic Scale EMS98 (Grunthal, 1998). The method was applied to the Grenoble city test-bed described in the first part of this paper. The ARL method is then presented and a vulnerability proxy was derived for a Grenoble city-like environment. The vulnerability proxy is tested in Nice in the third part, a city that has been the subject of a vulnerability study (Spence and Lebrun, 2006). Finally, the damage produced by historic earthquakes was computed, considering the (equivalent) earthquake-era and the present-day urbanization for simulating seismic damage.

Mots clés: sismology; vulnerability; risk; association rule learning; Lambesc

Domaines: Géophysique [physics.geo-ph]; Géophysique [physics.geo-ph]; Milieux et Changements globaux

Ismael Riedel, Philippe Gueguen, François Dunand, Stéphanie Cottaz. Macro-scale vulnerability assessment of cities using Association Rule Learning. Seismological Research Letters, 2014, 85 (2), pp.295-305. ⟨10.1785/0220130148⟩. ⟨hal-01015723⟩