Purchase-driven Classification for Improved Forecasting in Spare Parts Inventory Replenishment

Abstract

Performance of inventory management depends on the accuracy of demand forecasting. There are many techniques used for forecasting demand in retail sale. Advances in data mining application systems have given rise to the use of business intelligence in various domains of retailing. The current research captures the knowledge of classification of the customers using the purchase-based data of customers for improved forecasting. The model developed in this work suggests a technique for forecasting of demands which results in improved performance of inventory. The suggested forecasting model with the inventory replenishment system results in the reduction of inventory level and increase in customer service level. Moreover, the model makes use of purchase driven information instead of customers" demographic profile or other personal data for developing the decision tree for forecasting.

Modified replenishment policy for multi-item inventory of imperfect quality items using ABC classification and cross-selling effect

“…Clustering-based demand forecasting (Bala, 2012) and classification-based demand forecasting (Bala, 2010a(Bala, , 2010b systems have been developed for improved forecasting and enhanced inventory management. Various other techniques such as statistical inference, rule induction, data visualisation and applications of hop field neural network are defined in Kumari et al (2013).…”

Collaborative predictive business intelligence model for spare parts inventory replenishment

“…Bala proposed an inventory forecasting model which use of purchase driven information instead of customers' demographic profile or other personal data for developing the decision tree for forecasting [13]. The methodology combines neural networks, ARIMA and decision trees.…”