Design of mining under seismic threats implies choice from among alternatives. For each alternative solution the designer should assess the associated risk. It requires synthesis of many kinds of information. The decision process demands a weighing of the costs, risks and benefits of the alternatives, and ultimately the chioce of the single. This risk is a composite of both the likelihood od various levels of seismic threat and the uncertainty in the response of the object behaviour. Other words, the total risk is the combined effect of the different possible seismic events that might occur and the different ways in which a particular object may respond to a given event. In this paper the object is exposed to a probabilistically defined seismicity. It is assumed that the event occurrence constitutes a Poisson process. It means that the energy events are independent and identically distributed. This assumption implies that the distribution of the renewal time of event occurrence is not modified by physical process involved. In some instances, seismic activity tends to be more periodic and a renewal model can represent Poissone'a process. It is possible for events having very long returned periods - rather weak seismicity - or when high seismicity produces difficulties in recognising of events clustering in space and time. The response of object is related with mining performance. It can be predicted considering uncertainties represented in purely technical terms, which can be combined with economic consequences and hence expressed in economic terms. The decision making under uncertainty is the rool not used frequently in mining planning. The basic reason for this seems to prevail that the procedure requires an excessive effort in data collectionn and preparation.