Predictive Analytics for Production Line Downtime: A Comprehensive Study Using Advanced Machine Learning Models
A study found that a machine learning model called LSTM is the best at predicting production line downtime.
Researchers used advanced machine learning models to predict downtime in production lines. They analyzed three years of data on equipment failures, idle times, and runtime, and found that the Long Short-Term Memory (LSTM) model performed the best. The study used techniques like data preprocessing and cross-validation to ensure the accuracy of the predictions.
Abstract
This study delves into predictive analysis, employing advanced machine learning models to predict downtime in production lines. The research follows a detailed methodology, starting by exploring production lines and their conditions. The focal point is Overall Equipment Effectiveness (OEE), which includes availability, performance, and quality, with a special focus on downtime impacts. Through an OEE tracking system, downtime is categorized into planned and unplanned, offering insights into critical components affecting production line availability. The datasets cover three years, including runtime, unplanned downtime due to equipment failures, and idle times, emphasizing instances prone to frequent equipment failures.
Examining relationships through a heatmap reveals essential correlations crucial for accurate predictions. We outline the feature selection process and fundamental data preprocessing steps to maintain dataset integrity. Introducing machine learning models such as Xgboost, Prophet, Multi-Layer Perceptron (MLP), and Long Short-Term Memory (LSTM), each model is carefully configured with past window parameters and forecast horizons. To prevent overfitting, the study employs K-fold cross-validation and early stopping mechanisms, ensuring robust model performance.
Two datasets, comprising runtime, equipment-induced downtime, and idle time, are utilized for analysis. Data preprocessing involves a heatmap for characteristic-target correlation and basic steps like mean-based null value imputation. Results, presented in terms of Mean Absolute Error (MAE), underscore the superior performance of LSTM. The discussion interprets these findings, confirming the potential of LSTM for downtime prediction, supported by visualizations comparing actual and predicted downtimes. Future directions include comprehensive model comparisons, expanding metrics, and increasing datasets, building on the solid foundations established by this study.
References
- 1.Nakajima, S. (1988). An Introduction to TPM, Productivity Press, Portland, OR.
- 2.Next Generation Digital Factory Platform. SCW.AI. (2023, September 1). https://scw.aiLink
- 3.Tianqi, C. & Guestrin, C. (2016). Xgboost: A scalable and accurate implementation of gradient boosting machines. In Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining, pp. 785-794.
- 4.Taylor, S. J. & Letham, B. (2017). Prophet: forecasting at scale. In Proceedings of the symposium on advances in neural information processing systems, pp. 3293-3301.
- 5.Rumelhart, D. E., Hinton, G. E. & Ronald, J. W. (1986). Learning representations by back-propagating errors." Nature 323, no. 6088: 533-536.
- 6.Hochreiter, S. & Schmidhuber, J. (1997). Long short-term memory. Neural computation 9, no. 8, 1735-1780.
Saylam, A., Atlı, H. (2023). Predictive Analytics for Production Line Downtime: A Comprehensive Study Using Advanced Machine Learning Models. *The European Journal of Research and Development*, 3(4), 88-94. https://doi.org/10.56038/ejrnd.v3i4.354
Bibliographic Info
Indexing & License
More from The European Journal of Research and Development
Challenges in Maize Root Phenotyping: Preprocessing Limits and Class Imbalance in Deep Learning
Hüdanur Engin, Ali Murat Tiryaki
2026 · Vol 6 · Issue 1
The Bleaching of Woven Fabrics Using the Foam Application Technique
Aylin Kuşen, Onur Balcı, Koray Pektaş
2026 · Vol 6 · Issue 1
Automated Monkeypox Disease Classification Using Texture and Focus-Based Image Features
Tuğba Şentürk, Çiğdem Gülüzar Altıntop, Fatma Latifoğlu
2026 · Vol 6 · Issue 1
EEG-Based Assessment of Stress Levels Using Time–Frequency Features and Machine Learning
Sevde Samsa, Çiğdem Gülüzar Altıntop
2026 · Vol 6 · Issue 1
A Compact Non-Intrusive Measurement System for Critical Dimensions and Calibration Chart Generation of Underground Fuel Tanks
İlker Değirmencioğlu, Savaş Barış, Yusuf Kaya
2025 · Vol 5 · Issue 1
An AI-Based Question–Answering System for Corporate Documents: VK ArtiFin
Zeynep Örpek, Büşra Tural, Zeynep Destan
2025 · Vol 5 · Issue 1