ABSTRACTIn this article, we propose a methodology to evaluate the risk of environmental and life cycle impacts under data uncertainties that can be applied to a broad range of data availability assumptions. Specifically, we first propose a data uncertainty model that can accommodate scenarios where only a few data points are known, where data histograms are available, or where multiple, inconsistent data sources are present. An impact risk valuation model is then developed, based on the certainty equivalent of an exponential dis-utility function. We show that the evaluation of the impact risk value can be achieved using a closed-form expression and demonstrate an application in a food waste recycling alternatives comparison. We further extend the methodology to construct an impact safety index model that evaluates uncertain impacts using an impact tolerance level. We show that the proposed model is computationally tractable and can be used as an optimization criterion. Computational studies in an example of sustainable building material selection are then used to demonstrate the improvement of the proposed model compared with the standard approach of optimizing average impacts across several statistical criteria.