This paper presents a generalized framework to construct composite indicator, which can be used in static and dynamic analysis. By grouping DMUs (decision making units) first, the proposed approach is more flexible to derive weights for entities featuring diverse characteristics. A more neutral set of weights can be obtained through investigating the lower and upper bound of possible weights. Subsequently, we introduce a slack-based composite indicator from the perspective of distance function, which facilitates studying entities' improvement potential in sub-indicators. Furthermore, the slacks-based composite indicator is combined with the Malmquist index to conduct dynamic assessment, aiming to quantify the evolvement of composite indicator over time and the underlying driving forces. To illustrate the usefulness of the proposed approach, it is applied to construct the Sustainable Energy Index for 109 countries worldwide in 2005-2010. Our results show that the high-income country group has the best sustainable energy performance among all the three country groups in 2010. The dynamic assessment indicates the worldwide sustainable energy development level declined during 2005-2010, and the efficiency change was the main negative driving force. More discussions and implications are presented in the paper. (C) 2014 Elsevier Ltd. All rights reserved.