Navigating uncertainty: intermittent demand forecasting in the furniture manufacturing industry through clustering and predictive - case study

This study addresses the challenge of forecasting intermittent demand in the furniture manufacturing industry, highlighting the combination of clustering analysis and predictive models based on neural networks. Distinctive demand patterns were identified through detailed sales data collection and preparation, followed by advanced clustering techniques. The research delves into using classification and regression LSTM neural networks to predict future demand accurately. Findings reveal that this integrated approach significantly enhances forecasting accuracy, providing a solid foundation for the sector's inventory management optimisation and production planning. This work underscores the importance of tailoring forecasting strategies to the specific characteristics of intermittent demand, offering valuable insights for the furniture industry.