search for


Data mining based army repair parts demand forecast
Journal of the Korean Data & Information Science Society 2019;30:429-44
Published online March 31, 2019;
© 2019 Korean Data and Information Science Society.

HyungTae Kim1 · SuHwan Kim2

1Joint Chiefs of Staff Strategic Planning Headquarters, 2Department of Defense Science, National Defense University
Correspondence to: Professor, Department of Defense Science, National Defense University, Nonsan 33021, Korea. E-mail:
Received January 11, 2019; Revised March 12, 2019; Accepted March 16, 2019.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License ( which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Recent the development of science and technology, weapon systems have been upgraded and the cost of introducing, operating and maintaining the weapon systems has greatly increased. However, defense spending is limited, and the operational burden of army equipment is continuously increasing in order to cope with North Korea’s constant threats. Therefore, the ARMY needs precise demand forecasts for the spare parts in order to maintain proper operational availability under budget constraints. This is a study to develop a model for predicting of the 3-equipment including k-9. In order to do this, we collected the data that affected the demand for the spare parts using DELIIS (defense logistics integrated information system). The objective variable is the quantity of spare parts in 2017 and RMSE (root mean squared error), MAE (mean absolute error) are used as the predictive power measure. To construct an optimal demand forecasting model, regression tree, random forest, neural network and linear regression model were used an data mining techniques. The model construction results showed that RMSE, MAE value was the best in the random forest model and the predicted quantity was also highest in the scatter plot.
Keywords : data mining, repair parts, random forest, multiple regression