气候变化领域集成服务门户

Indicators are commonly used for evaluating relative sustainability for competing products and processes. When a set of indicators is chosen for a particular system of study, it is important to ensure that they vary independently of each other. Often, the number of indicators characterizing a chosen system may be large. It is essential to select the most important indicators from a large set so that a dependable bias-free analysis can be done using the reduced set of indicators. In this paper, we propose the use of principal component analysis (PCA) along with the partial least square-variable importance in projection (PLS-VIP) method to ensure that the explicit or tacit assumption of the independence of the chosen indicators is valid. We have used two case studies to demonstrate successful use of these two methods for parsimonious use of indicators for sustainability analysis of systems.