In this paper, we compare and analyze demand forecasting results of existing methods and proposed method using K-Nearest Neighbor method in predicting irregular demand patterns of Self-propelled Artillery Repair Parts. Our analysis shows that forecast...
In this paper, we compare and analyze demand forecasting results of existing methods and proposed method using K-Nearest Neighbor method in predicting irregular demand patterns of Self-propelled Artillery Repair Parts. Our analysis shows that forecasting error obtained from K-Nearest Neighbor method is smaller than those obtained from existing methods. This result is also verified through paired T-test. It is expected that our proposed method can be used to improve not only the effectiveness of defense budget utilization but also weapon system readiness in future military operations.