The Effectiveness of Using Fuzzy Neural Networks in Predicting the Sales Volume of Perishable Products

Nataliya Mutovkina

doi:10.47852/bonviewAIA52026341

Authors

Nataliya Mutovkina Department of Management and Social Communications, Tver State Technical University, Russia

DOI:

https://doi.org/10.47852/bonviewAIA52026341

Keywords:

sales volume, demand, forecasting, dynamics series, neural network, neuro-fuzzy network, average error

Abstract

This article discusses methods for predicting the sales volume of perishable products to identify the best method that provides the most accurate forecast value of the indicator under study. The initial data for forecasting are the volumes of daily perishable products that we sell. For a trading company, the volume of products purchased for retail sale must match the volume of demand. Therefore, the presence of product residues at the end of the trading day is unacceptable. That is especially true for products with a short shelf life (chilled meat, dairy products, fruits, vegetables, etc.). We must write off the products as losses if the sale period has expired. Therefore, identifying the most accurate forecasting method in this area is a significant task. Solving this problem will reduce losses from trading activities. For comparison, the following three methods were used: alignment of the obtained dynamics series according to the Fourier series, forecasting using a two-layer neural network, and forecasting using a neuro-fuzzy network. We determined the accuracy of each method by calculating the average approximation error. The prediction method using a neuro-fuzzy network showed higher efficiency. The discrepancies between the actual and theoretical values of the implementation volume were minimal. Therefore, the developed neuro-fuzzy model can be used to predict the sales volume of perishable products. We are further testing the neuro-fuzzy model on new sales data.

Received: 3 June 2025 | Revised: 30 July 2025 | Accepted: 14 September 2025

Conflicts of Interest

The author declares that she has no conflicts of interest to this work.

Data Availability Statement

Data sharing is not applicable to this article as no new data were created or analyzed in this study.

Author Contribution Statement

Nataliya Mutovkina: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization, Project administration.