Resumen de Systematic tools based on data envelopment analysis for the life cycle sustainability evaluation of technologies
Moving towards a more sustainable energy system has been a main aim of modern societies during the last decades. The minimization of the fossil fuels dependence in order to improve sustainability has become the goal for investigators and policy makers jointly with the development of new strategies and policies.
The presented thesis introduces the combined use of data envelopment analysis (DEA) and life cycle assessment (LCA) technique as a suitable tool to achieve these challenges. The environmental performance (eco-efficiency) of the electricity mix of the top European economies has been studied and the environmentally efficient and inefficient countries have been identified.
For the inefficient ones, the targets have been provided in order to make them efficient. When making decisions for the future electricity mix other considerations of economic and social nature should be considered. For example, the specific character of electricity mix of Norway has been classified this country as an outlier. After removing Norway from the analysis, the following seven countries out of 26 have been found eco-inefficient: Czech Republic, Greece, Croatia, Hungary, Luxemburg, Slovenia and Slovakia and the reduction values of their impact (compared to the current level) should be: 6.94%, 24.31%, 9.97%, 22.23%, 6.73%, 2.27% and 22.57%, respectively.
The main proposed changes implied reductions of different magnitude in the share of fossil fuels, which cause significant environmental impacts.
Moreover, the DEA+LCA method has been extended to deal with the uncertainty, through the implementation of the Pedigree matrix approach and Monte Carlo simulation to enable the eco-efficiency assessment under uncertainty. The consideration of uncertainty is crucial to allow the suitable generation of practical and good quality management for decision makers. Numerical results show that the efficiency scores in the nominal and the stochastic cases can differ significantly, and the same applies to the target values established for the inefficient units. Moreover, the differences in efficiency values affect the ranking of alternatives. These results support the need to incorporate uncertainties into the LCA+DEA framework in order to provide further insight into the problem and assess the robustness of the results obtained.
Our results provide valuable insight for governments and policy makers that aim to satisfy the electricity demand while minimizing the associated environmental impacts and considering the uncertainties of the data.
