Exploring the Drivers of Food Waste in the EU: A Multidimensional Analysis Using Cluster and Neural Network Models.

Food waste poses a significant global challenge with profound economic, environmental, and social implications. Within the European Union, socioeconomic conditions, food affordability, and sustainability initiatives create a complex framework for understanding and mitigating food waste. This study examines how economic and sustainability factors shape food waste patterns across EU member states, employing advanced statistical techniques to uncover underlying dynamics. The analysis focuses on five key variables: the Harmonized Index of Consumer Prices for food, food waste, food retail sales, the Sustainable Development Goals Index, and GDP per capita. Factorial analysis and a general linear model were used to investigate linear relationships, and multilayer Perceptron (MLP) neural networks were employed to model the non-linear relationships driving food waste. At the same time, hierarchical cluster analysis identified four distinct country groups, each characterized by unique combinations of these variables. The results reveal that higher GDP per capita and stronger sustainability performance are associated with lower food waste, whereas higher food prices and increased retail activity present more nuanced influences. The findings underscore the importance of customized policies that address the EU's diverse socioeconomic and sustainability contexts, offering a pathway toward more sustainable food systems and reduced waste.