The volume of influence of mining works is essentially determined from in situ observations, field measurements (particularly surface subsidence measurements) as well as from numerical investigations. Seismic monitoring is the primary requirement for mining in rock-burst environment. A micro-seismic network is systematically used to monitor coal extraction in the Provence colliery (south-eastern France). A significant database of seismic events is now available. A further advancement in this sequence was added by the development of Doublets technique or multiplet selections. The technique uses with the database to identify families of seismic events and to work out the relative locations of these events. The analysis of location of these multiplet families showed a particular spatial distribution. Locations of the epicentre follow the direction in which the coalface is advancing. Numerical models are used in the paper to show the overstress distribution. By comparing the seismic activity and results of the numerical modeling of the advancing coalface with observations, a close correlation can be established between the location of seismic activity and induced stresses in the ground surface of the working areas. By combining micro-seismic monitoring with the numerical modeling, the volume of influence of long-wall mines is determined. Results show that such a coupling can be an efficient way for detecting areas subjected to rock-burst hazards. This can also prove to be a powerful tool for assisting the planning of underground workings in complex geological and mining conditions.