Consumable materials in the electricity industry are the important materials needed to meet consumer demand for both new electrical installations and power addition. Stock-out on consumable materials often occur in the company used in this research case study. The purpose of this paper is to analyze and determine the safety stock, reorder point, and lot size policies to manage and control consumable materials in the case study. The forecasting method used in this research is ARIMA and Artificial Neural Network (ANN), as well as two types of lot-sizing rules, Economic Order Quantity (EOQ) and Periodic Order Quantity (POQ). Based on last year’s demand data for new electrical installations and power addition, the ANN forecasting method is shown to be more appropriate because it has a smaller error than the ARIMA method. Twenty consumable materials were analyzed in this paper. Thirteen materials need to use the POQ lot-sizing rule, while the remaining seven need to use the EOQ rule correctly. The result shows that a change in the lot size rules used for each material, cause an increase in the service level.