Optimization of Oxygen and Acetylene Gas Inventory Control Through Forecasting-Based ROP and ROQ Evaluation at PT Kaltim Prima Coal

Abstract

This internship aims to evaluate and optimize the inventory control of industrial oxygen and acetylene gas at PT Kaltim Prima Coal, which is managed under a closed-loop system. The main issues are the mismatch between the Reorder Point (ROP) and Reorder Quantity (ROQ) parameters in the Ellipse system and actual demand fluctuations, as well as an imbalance in bottle distribution that causes stockouts in the warehouse. The solution methods include statistical analysis of historical data from 2024–2025, forecasting using Exponential Smoothing, recalculation of ROP/ROQ based on three lead time scenarios, and determination of the ideal number of cylinders using the Demand During Lead Time approach. Analysis results indicate that the best forecasting method for oxygen is Exponential Smoothing with α=0.3 (MAPE 15.86%) and for acetylene, α=0.5. Recalculations using actual lead times resulted in an oxygen ROP of 22 bottles (4-day lead time) and an acetylene ROP of 29 bottles (11-day lead time), significantly different from the existing parameters (oxygen ROP=3, acetylene ROP=0). The ideal number of bottles in the system is 95 bottles for oxygen and 91 bottles for acetylene, while the actual inventory of 103 oxygen bottles and 87 acetylene bottles indicates that the root cause is not a shortage of bottles, but rather an imbalance in distribution across cycle points. The primary recommendation is to accelerate bottle returns and even out distribution without needing to increase the total bottle inventory.