A Waste Reduction Journey: Data-driven and Machine Learning on Perishable Food Waste Optimisation in New Zealand Retail Supermarkets https://doi.org/10.34074/rere.00403

By: Nenie Tamarisca Petrisia Indrapriya Kularatne

Abstract

he issue of fresh food waste in retail supermarkets globally has become increasingly problematic, with New Zealand
reporting alarming levels of waste primarily in vegetable, bakery, meat, fish, and fruit products. The retail supermarkets in
NZ contribute more to food waste than any other sector, with an estimated 60,500 tonnes of food wasted annually due to
customers’ increasing expectations for fresh food, which is highly perishable. Despite numerous advanced technologies
aimed at mitigating food waste, including deep learning techniques for fruit waste reduction and machine learning for
food donation optimisation, perishable food waste remains a persistent challenge. This study employed a systematic
literature review from scholarly journals to explore the potential of data-driven approaches combined with machine
learning in preventing perishable food waste in retail supermarkets. The scholarly articles were extracted from online
research databases. Drawing upon the principles of Sustainable Consumption and Production which aims to halve global
food waste at the retail level and reduce waste through prevention strategies. This review focused on analysing existing
research studies that investigated the application of machine learning for demand forecasting based on sales data and
influencing parameters. Through a thematic analysis of the literature, this review identified key insights, challenges, and
opportunities associated with implementing data-driven and machine learning strategies to mitigate perishable food
waste in New Zealand retail supermarkets. By synthesising findings from the literature this study highlighted the urgency
of the food waste prevention stage by utilising data-driven approach and machine learning for predicting accurate demand
benefiting managers on fresh food purchase order decisions. Therefore, it helps improving operational outcomes,
profitability, and customer satisfaction, specifically in lowering discarded unsold fresh food products.

Keywords:
Data-driven, Machine Learning, Sustainability, Fresh food, Perishable food, Food waste reduction, Retail supermarkets.

Download the article (pdf)