A Comparison of Data Mining Approaches for Forecasting Sales of FMCG Food Products

Sale forecasting is considered a key factor for balancing risk of over-stocking and out-of-stock. Its performance is largely influenced by the input it provides to supply chain activities. A competitive advantage for producers and retailers is the ability to predict sales of fast-moving consumer goods with greater accuracy, particularly in the food industries. If the accuracy of the sale prediction is poor, even with significant effort and financial investment, supply chain activities will still be unreliable and inefficient. The objective of the study is to evaluate the effectiveness of various data mining algorithms for predicting the future sales of FMCG food products. The study analyses four popular data mining techniques: Cluster Analysis, Regression Analysis, Visualization and Tracking Patterns on a dataset of historical sales data. The study compares the accuracy and efficiency of these algorithms using various performance metrics. The research aims to help the different stakeholders, like various businesses were involved in wholesale, manufacturing, marketing, logistics, and retailing. This study provides valuable insights for FMCG food product sales forecasting, and can be useful for businesses to make informed decisions about their sales strategies.