Integrated cost-benefit analysis and prescriptive decision tree model for a flood risk management problem

Climate changes and urbanization trends increase occurrences of extreme flood events, which generate relevant disasters worldwide and motivate the need to set up adequate flood risk management plans. These plans should be able to develop comprehensive planning approaches and to provide decision support evaluating damage mitigation alternatives. The European Directive 2007/06/CE establishes that the flood-risk evaluation should include Cost-Benefit Analysis (CBA) on a long-term planning horizon, drawing attention to the assessment of flood mitigation measures. The CBA applied to the flood risk management can estimate the strength and the weakness of mitigation projects. The effectiveness of flood risk mitigation measures could be evaluated through economic criteria (e.g.: net benefit, rate of return, benefit-cost ratio, payback period evaluations, etc.) according to objectives motivating the choice among different alternatives. Moreover, the decisions selecting mitigation measures must be made before observing data, which are affected by high level of uncertainty. To deal with this problem, an approach based on prescriptive decision trees (PDTs) is proposed in the paper. They are analytical models to systematically structure and analyze decision problems under uncertainty. The integration of CBA and PDT can provide a decision-support for the optimal planning of flood mitigation measures. The paper reports the outcomes obtained in a real case-study for the Coghinas river flood plain (North-West Sardinia, Italy).