A predictive modelling approach to decoding consumer intention for adopting energy-efficient technologies in food supply chains

Abstract

The transition towards energy-efficient practices in the food supply chain (FSC) is essential for addressing the dual imperatives of sustainability and cost-effectiveness. As consumers become increasingly aware of the environmental impact of their food choices, their willingness to support energy-efficient technologies (EET) has become a critical factor in shaping the future of sustainable FSC. This study empirically investigates consumer intention and desire to pay for food products characterized by a reduced energy footprint, utilizing machine learning (ML) algorithms to predict consumer preferences within the FSC. Association rule mining (ARM) was employed to uncover key patterns in consumer intentions, while multiple ML algorithms were compared to identify the most effective algorithm for predicting willingness to pay. The results reveal that the Random Forest algorithm achieved the highest accuracy at 82%, significantly outperforming other models. These findings underscore the potential of ML to refine marketing strategies and operational decisions, facilitating the broader adoption of EET within the FSC (EET-FSC). The study offers valuable implications for industry professionals seeking to enhance sustainability efforts through data-driven decision-making. The research contributes to optimizing FSC through improved decision-making, resource allocation, and sustainability initiatives. Future research directions include expanding the dataset scope, exploring advanced ML techniques, and examining the economic impacts of EET-FSC.