A novel approach for developing composite eco-efficiency indicators: The case for US food consumption

Eco-efficiency analysis can provide useful information for sustainability benchmarking of products and sectors while assessing and monitoring their economic and environmental performances. The eco-efficiency is defined as a ratio between economic performance and environmental impact. With multiple environmental and economic metrics, the eco-efficiency assessment is computationally complex. One common aspect of this complexity is associated with the importance (a.k.a. Relative weights) of sustainability indicators in the presence of high multicollinearity. A novel weighting method integrating two well-established methods for reducing the multicollinearity consequence during the aggregation process is presented in the study. The proposed method's mathematical and operational procedures, called Weighted Penalized Maximum Likelihood Estimation (W-PMLE), are demonstrated for the eco-efficiency analysis of U.S food consumption. The eco-efficiency analysis results revealed that the CO2 emissions, the level of consumption of the metallic mineral, and water were the most critical to the eco-efficiency performance of U.S. consumption. (C) 2021 Elsevier Ltd. All rights reserved.

Automated summary

Eco-efficiency analysis is a method for evaluating and monitoring the economic and environmental performance of products and industries, involving multiple environmental and economic indicators, and considerations of the importance of sustainability indicators under high multicollinearity conditions. A novel weighting method called Weighted Penalized Maximum Likelihood Estimation (W-PMLE) is proposed to reduce the impact of multicollinearity, and applied to eco-efficiency analysis of U.S food consumption, revealing key indicators such as CO2 emissions, consumption of metallic minerals, and water resources.