According to the Turkey Waste Report of 2018, an alarming 26 million tons of food are squandered annually in our country, representing a staggering financial loss exceeding 555 billion Turkish Lira, equivalent to 15% of the national income. This wastage results in the loss of food items and squanders water, energy, time, and valuable resources. The fact that 53% of the fruits and vegetables produced in Turkey are wasted from the field to the consumer due to faulty marketing practices is particularly concerning. Unlike in developed countries, where contract production agreements between producers and modern retailers minimize losses, Turkey's agricultural sector suffers from inefficient inventory management. However, implementing predictive models utilizing machine learning algorithms offers a promising solution. These models enable accurate demand forecasting with the analysis of historical data, facilitating optimal ordering practices to reduce shelf time and maintain product freshness. Recognizing the pivotal role of pricing in driving perishable goods demand, integrating sales price as a decision variable in practical scenarios is crucial. Even slight price reductions can stimulate demand, underscoring the necessity for nuanced pricing strategies aligned with demand patterns to mitigate food wastage while enhancing market sustainability. In this study, we aim to supply a categorical analysis of the machine learning models implemented for preventing waste in perishables and the fast-moving consumer goods industry. ORCID NO: 0009-0005-0508-3874, 0009-0003-7491-3988, 0000-0002-2589-3568
Anahtar Kelimeler: FMCG, Perishables, Machine Learning Models
|
