Claim Missing Document
Check
Articles

Found 2 Documents
Search
Journal : Operations Research: International Conference Series

Investigating Long-Run and Short-Run Dynamics of Palm Oil Production with Key Factors Using the VECM Method Lathifah Zahra; Gustriza Erda
Operations Research: International Conference Series Vol. 6 No. 4 (2025): Operations Research International Conference Series (ORICS), December 2025
Publisher : Indonesian Operations Research Association (IORA)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47194/orics.v6i4.429

Abstract

This study investigates the long-run and short-run relationships among palm oil production, rainfall, the number of bunches per palm (NOB), and average bunch weight (BTR) using the Vector Error Correction Model (VECM). Monthly data from 2015 to 2024 obtained from PT Perkebunan Nusantara IV (PTPN IV) Regional III, Sei Rokan Estate, were analyzed. Descriptive statistics indicate high variability in rainfall and relatively balanced distributions for production, NOB, and BTR. The Augmented Dickey-Fuller (ADF) test confirmed that all variables became stationary after first differencing, and the Johansen cointegration test identified three cointegrating relationships, suggesting both short-run and long-run linkages among variables. The VECM estimation results reveal positive long-run relationships for palm oil production (ECT = 0,052), rainfall (ECT = 0,090), and NOB (ECT = 0,042), indicating that these variables move toward long-run equilibrium in the same direction. In the short run, previous rainfall significantly affects both current palm oil production and NOB, with coefficients of 0,203 and 0,178, respectively, highlighting the critical role of rainfall fluctuations in influencing short-term productivity and fruit development. Model evaluation using the Root Mean Square Error (RMSE) shows low prediction errors across all variables, with rainfall having the highest RMSE (1,334) and NOB the lowest (0,962), confirming the model’s strong predictive performance. Overall, the findings demonstrate that the VECM approach effectively captures both long-run equilibrium and short-run dynamics among key determinants of palm oil productivity in the Sei Rokan plantation.
Palm Oil Production Forecasting Using the SARIMA Model at the Terantam Plantation of PTPN IV Regional III in 2025 Eky; Erda, Gustriza
Operations Research: International Conference Series Vol. 6 No. 4 (2025): Operations Research International Conference Series (ORICS), December 2025
Publisher : Indonesian Operations Research Association (IORA)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47194/orics.v6i4.430

Abstract

Palm oil is one of the important plantation commodities that plays a major role in the Indonesian economy because it contributes to state revenues, making palm oil production crucial. Forecasting palm oil production is essential to support effective planning and decision-making in plantation management. This study aims to forecast palm oil production at the Terantam Plantation of PTPN IV Regional III for the year 2025 using the Seasonal Autoregressive Integrated Moving Average (SARIMA) model. The data used consist of monthly production data based on volume (kg) from January 2014 to December 2024. The results of the analysis indicate that the best model obtained is SARIMA(0,1,4)(0,1,1)12 with the smallest Akaike Information Criterion (AIC) value. Diagnostic tests show that the model residuals behave as white noise and are normally distributed, indicating that the model is suitable for forecasting. The Mean Absolute Percentage Error (MAPE) value of 8.02% indicates a very good level of accuracy. The forecasting results reveal a seasonal pattern in palm oil production, with the highest production in September 2025 amounting to 15,108,145 kg, and the lowest in February 2025 at 9,347,573 kg. Overall, the SARIMA model is able to capture both trend and seasonal patterns effectively, making the forecast results useful as a reference for production planning and operational management at the Terantam Plantation. Furthermore, the findings of this study are expected to serve as a reference for applying similar forecasting methods to other plantation commodities.