A decision-analytic framework for impact assessment
- Cite this article as:
- Hertwich, E.G. & Hammitt, J.K. Int J LCA (2001) 6: 265. doi:10.1007/BF02978787
- 252 Downloads
In the first part of this paper, we showed how life-cycle impact assessment can be described as an exercise in decision analysis. We developed a structure for how to decide on the relative importance of different environmental stressors. In this second part, we offer criteria for the grouping of stressors into impact categories and for the development of impact indicators. Facts to be included in a characterization method should be selected according to their relevance and combined following established scientific models. Facts should be included only if they are informative, that is, if sufficient and sufficiently certain information is available for all stressors that should be evaluated by this method. Abstract, constructed indicators at the ‘midpoint level’ are better suited to compare similar impacts than indicators reflecting ‘observable environmental endpoints’ if there is a large uncertainty about the effects on observable endpoints. We argue that midpoint modeling should be retained. The additional evidence introduced by endpoint methods should be used to support ‘judgments about facts’ needed to evaluate the importance of different impact categories (or means objectives) in the means-ends objectives network.