Predictive Maintenance and Time Series Forecasting for Industrial Equipment Health Monitoring

Ahmad Farrel Sirajudin Zaidan

Abstrak

This study compares two forecasting methods used in the assessment of industrial equipment health: ARIMA and Monte Carlo simulation. ARIMA is a time series analysis method that focuses on historical patterns to make future predictions, while Monte Carlo simulation uses random sampling to account for uncertainty across various scenarios. Both methods were applied to historical data to evaluate their predictive performance. The results of the study indicate that ARIMA excels in short-term forecasting, providing predictions with high accuracy and minimal error under stable data conditions. In contrast, Monte Carlo simulation, although more reliant on random modeling, offers greater reliability in long-term forecasting, especially when dealing with high variability and uncertainty. Therefore, Monte Carlo is more suitable in scenarios where predictions must consider numerous uncertain factors, while ARIMA is more effective when clear and consistent patterns can be identified from historical data. This research highlights the importance of selecting a forecasting method that aligns with specific needs, based on the application context and time horizon. The findings contribute significantly to the development of better forecasting practices and help inform decision-making in the management of industrial equipment health. Through this in-depth analysis, the study also introduces a new approach to understanding forecasting dynamics by considering the strengths and limitations of each method.

© 2026 UPA Perpustakaan POLINEMAStatus LayananKebijakan PrivasiSyarat & Ketentuan

v2026.11.17.4 · 26b24a0