Data Management for Sustainability Evaluation

Industrial systems are evaluated for relative sustainability of competing products and processes with quantitative indicators. These indicators should be independently variable. Also, the number of indicators defining the system may be large. Economy of effort for dependable decision making suggests that a reliable method be used for reducing the indicator set to a necessary and sufficient number.
To this purpose, two multivariate statistical analysis methods were used. We propose the use of Principal Components Analysis (PCA) to ensure that the indicators are linearly independent. We also propose the use of Partial Least Square – Variable Importance in Projection (PLS-VIP) method as a supervised model where an overall pattern of the datasets is used to obtain the most important indicators from the given set of indicators. This work provides a description of these two methods and their use in making decisions on relative sustainability with case studies from the literature.