Dealing with uncertainty in sustainability assessment : report on the application of different sensitivity analysis techniques to fieldspecific simulation models

Assessing sustainability is more and more becoming a common practice in products, policies and institution appraisals. However, increasing concern has been recognized in the scientific community regarding whether the various available examples of sustainability assessment are really comprehensive and able to judge in a robust and reliable way if new developments to "meet the needs of the present without compromising the ability of future generations to meet their own needs". It is possible to identify three main sources of uncertainty: i) the "sustainable development" concept and the definition of boundaries (physical, economic and social) to assess it, ii) the intrinsic subjectivity of many assessment tools, iii) the incapability of many available modelling activities to mimic our world. This report tries to define a conceptual framework in order to deal with the uncertainty within a sustainability assessment describing some statistical techniques that can be fruitfully applied for this purpose. The conceptual framework proposed is centred on sensitivity and uncertainty analysis. A Monte Carlo framework is suggested in both the cases, and examples of its application to field specific models are provided. Results clearly show the powerfulness of such techniques in providing the analyst with useful information to manage the uncertainties hidden behind any assessment exercise.