Improving Inventory Management Decisions by Outliers Detection and Elimination

Abstract

The presence of outliers in the real databases of many applications is often unavoidable, and one of the most important of these applications is the data in the inventory management sector and its role in the continuation of production. In this paper, a Z-Score Modified method is proposed to detect outliers in the demand data for raw materials used in the cement industry in one of Iraq's factories, and using the method of deletion and reusing the custom model for inventory management. The results showed that deleting outliers contributed to reducing the deficit rates for raw materials (oil, limestone, dust and iron) by 39.84%, 68.62%, 89.02% and 41.71%, respectively, it also contributed to an increase in profit by 614062600 ID during two years as a result of avoiding the deficit that caused the suspension of cement production.