This paper outlines a methodological approach utilizing quantitative risk analyses and cost-benefits evaluation in order to select the most cost-effective options for risk reduction in industrial plants including structural changes and preventive maintenance. The methodology is based on quantification of top events resorting to fault trees, a sensitivity analysis of the probability of occurrence of top events, and a comparison of corrective measures based on cost and effectiveness criteria. The method is applied to a case study represented by the pseudocumene purification plant of the Borexino experiment in order to show its capability. The experiment is located in the Gran Sasso National Laboratory operated by the Italian National Institute for Nuclear Physics. (C) 2004 American Institute of Chemical Engineers Process.