ABC Classification in Spare Parts for Inventory Management using Ensemble Techniques

The aim of this work is to present the data mining techniques for classifying stock keeping units to ABC classification in spare parts warehouse in a factory of Thailand. ABC analysis is an “inventory categorization method” that divide items into three categories (A, B, and C), which are the label of data. The data were collected from SAP inventory management in the year of 2015-2019. There were totally 21,222 instances and 28 attributes. The dataset is processed with five popular ensembles techniques including Randomforest, Bagging, Adaboost, Dagging and Decorate. The results of this study give insight into techniques for accurately classifying ABC categorization for inventory management, compare the precision, recall, f-measure and accuracy of five ensemble algorithms that can be deployed to the inventory process for the factory.