Clustering Hierarchical Disjoint Principal Component Analysis for environmental impact assessment of food products

This study evaluates the environmental impacts of food production by proposing a new methodology named Clustering Hierarchical Disjoint Principal Component Analysis that allows for a deep understanding of multivariate data and simplifies their interpretation by using clustering and a hierarchical set of Specific Composite Indicators (SCIs) that arrive to a General one (gCI). In fact, by utilizing data encompassing 43 food items, the research constructs SCIs for environmental dimensions such as land use, water consumption, and greenhouse gas emissions, which are synthesized into a gCI representing the overall foods environmental impact. The methodology also identifies significant clusters of food items, revealing variations in environmental burdens and offering insights into sustainable dietary choices. The proposed methodology aims to guide policy and consumer decisions towards reduced environmental footprints in food production.