Disasters are events that cause destruction of the local infrastructure and impact thousands of people every year, causing material and human damages and significant economic losses. When this happens, humanitarian aid is required to be sent to affected locations and rescue survivors as soon as possible. In order for a disaster response operation to happen efficiently, it is necessary that the transport network to be operational and for hospitals to be able to receive the victims. To mitigate the impact of these events, a linear programming model is proposed that seeks to identify the urban roads and hospitals that need investments to better serve the population in case of a disaster. The model identifies which pathways are critical and therefore should receive priority investments and which hospitals need to have their capacity increased. In order to verify the application of the proposed model, scenarios for landslide catastrophes and disasters were created in the city of Rio de Janeiro. On the first case, it was considered a situation where all the districts considered were affected at the same time. On the latter, more likely to occur scenarios were set, where only one or a few areas were impacted. In this way, decision-makers start to rely on a tool to guide the allocation of limited resources, in a careful and preventive way. The direct benefit is a better service to the population of the city in the event of a disaster, but also in a non-crisis situation.