The paper is providing “experimental evidence that a rankboost algorithm is effective to achieve a good trade-off between the accuracy of the final rank and the elicitation effort”. The test field is Trentino (Italian northern region) and the problem: “ranking a set of areas that a preliminary and rough analysis has recognized as risky from the hydrogeological point of view”. The goal “[…] was to implicitly elicit a rationality for planning further studies of deeper analysis and the related emergency plans. In particular this process of assessment was required to take into account the systemic vulnerability, i.e. the capability of the social organizations to deal with unexpected events”.