Optimizing Food Demand Forecasting in the Supply Chain for Shelf-Life Management

Accurate food demand forecasting plays a critical role in optimizing supply chain operations, reducing waste, and ensuring effective shelf-life management of perishable goods. Its applications span from retail inventory management to large-scale food distribution, enabling businesses to maintain an optimal stock of products such as bread, butter, and other perishables. By anticipating demand fluctuations, organizations can better align production schedules, reduce overstocking and understocking issues, and minimize financial losses. Effective forecasting also supports sustainability by reducing food waste and enhancing consumer satisfaction through improved product availability. Traditional demand forecasting systems often rely on manual approaches or static statistical methods, which are limited by their inability to adapt to dynamic market conditions and complex time-series data. Manual methods, in particular, are prone to human error, delays, and inefficiencies, making them unsuitable for high- stakes decision-making in the supply chain. Furthermore, these approaches struggle to account for multiple influencing factors, such as seasonality, market trends, and external disruptions, resulting in inaccurate demand predictions and poor shelf-life management. To address these limitations, this paper proposes the use of a novel algorithm called the Nonlinear Autoregressive Exogenous Neural Network (NARXNN) for food demand forecasting. NARXNN is a recurrent dynamic network characterized by feedback connections that encompass multiple layers, enabling it to process complex and nonlinear time-series data effectively. Derived from the linear ARX model, NARXNN leverages exogenous inputs to enhance its predictive capabilities. By applying NARXNN to supply chain products such as bread and butter, the model showcases its potential to optimize demand forecasting, improve inventory management, and reduce wastage, thereby setting a new standard for shelf-life management in the food industry.