預測性維護(Predictive Maintenance)是工業生產中很重要的一項發展課題，有別於其他應用機器學習技術純粹以達到準確度的最大化，在機台健康狀態預測中，除了準確率的提升，同時能夠得知機台何處損壞而造成整體健康狀態下降也是非常重要的課題。本研究欲探討機台健康狀態以特定模式變動的情況下(其中包含固定的維修模式或是固定的衰退周期)，使用機台中安裝於不同部件的偵測器收集之資料以及健康狀態時間序列資料，建立以時間序列預測方法和數據導向技術(Data-Driven Techniques)結合之通用預測性維護模型框架。透過以時間序列預測方法結合多種不同的數據導向分析技術之結果比較，本研究提出以融合模型(Fusion Model)的方式來提升整體準確率，並保留其初始模型可探討各特徵影響程度之特性。本研究進而探討資料收集的程度是否足以使得時間序列模型有良好的運算效果，並能夠在預測性維護模型應用初期透過監督式學習模型與時間序列模型兩者的融合，調整其資訊應用的比重。研究結果發現透過融合方式使得小數據也可能達到最佳的預測效果以利工廠後續排定生產維修計畫以及零件備品採購決策。

Among the topics in industrial data analytics, one of the important topics is predictive maintenance. Predictive maintenance mainly focuses on the traceability of where the machine is broken or needed to maintain based on the improved prediction accuracy on the sensor data from machines or devices. In previous research works, time series data analysis of the health status of the machine under fixed maintenance mode or fixed recession cycle are studied. In this research, the proposed data fusion model is expected to improve the overall accuracy through the combination of the results and various-data driven technologies with the time series prediction method. At the same time, the size of data collection is an important factor to make the time series model have good quality. Moreover, the property of the initial model to explore the influence degree of each feature is used to facilitate the subsequent scheduling of production and maintenance plans. The experimental result shows that the proposed fusion model can provide a better maintenance decision making on the machine/device maintenance plan based on the relative small or not complete data.

Banerjee, T. P. and S. Das (2012). "Multi-sensor data fusion using support vector machine for motor fault detection." Information Sciences 217: 96-107.

Baptista, M., et al. (2018). "Forecasting fault events for predictive maintenance using data-driven techniques and ARMA modeling." Computers & Industrial Engineering 115: 41-53.

Ben-Gal, I. (2007). "Bayesian Networks." Encyclopedia of Statistics in Quality and Reliability.

Biagetti, T. and E. Sciubba (2004). "Automatic diagnostics and prognostics of energy conversion processes via knowledge-based systems." Energy 29(12): 2553-2572.

Breiman, L. (2001). "Random Forests." Machine Learning 45(1): 5-32.

Byington, C. S., et al. (2004). A model-based approach to prognostics and health management for flight control actuators. 2004 IEEE Aerospace Conference Proceedings (IEEE Cat. No.04TH8720).

Cerqueira, V., et al. (2016). Combining Boosted Trees with Metafeature Engineering for Predictive Maintenance.

Cerrada, M., et al. (2016). "Fault diagnosis in spur gears based on genetic algorithm and random forest." Mechanical Systems and Signal Processing 70-71: 87-103.

Chen, T. and C. Guestrin (2016). XGBoost: A Scalable Tree Boosting System. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco, California, USA, ACM: 785-794.

Christ, M., et al. (2018). "Time Series FeatuRe Extraction on basis of Scalable Hypothesis tests (tsfresh – A Python package)." Neurocomputing 307: 72-77.

Christ, M., et al. (2016). "Distributed and parallel time series feature extraction for industrial big data applications." arXiv preprint arXiv:1610.07717.

Cryer, J. D. and N. Kellet (1991). Time series analysis, Springer.

Dietterich, T. G. (2002). Machine Learning for Sequential Data: A Review, Berlin, Heidelberg, Springer Berlin Heidelberg.

Freund, Y. and R. E. Schapire (1997). "A Decision-Theoretic Generalization of On-Line Learning and an Application to Boosting." Journal of Computer and System Sciences 55(1): 119-139.

Fuentes, M., et al. (2002). On line signature verification: Fusion of a Hidden Markov Model and a neural network via a support vector machine. Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition.

Górecki, T. and M. Łuczak (2015). "Multivariate time series classification with parametric derivative dynamic time warping." Expert Systems with Applications 42(5): 2305-2312.

Geurts, P. (2001). Pattern Extraction for Time Series Classification, Berlin, Heidelberg, Springer Berlin Heidelberg.

Geurts, P. and L. Wehenkel (2005). Segment and combine approach for non-parametric time-series classification. European Conference on Principles of Data Mining and Knowledge Discovery, Springer.

Helwig, N., et al. (2015). Condition monitoring of a complex hydraulic system using multivariate statistics. 2015 IEEE International Instrumentation and Measurement Technology Conference (I2MTC) Proceedings.

Helwig, N., et al. (2015). Detecting and Compensating Sensor Faults in a Hydraulic Condition Monitoring System.

Jardine, A. K. S., et al. (2006). "A review on machinery diagnostics and prognostics implementing condition-based maintenance." Mechanical Systems and Signal Processing 20(7): 1483-1510.

Lee, J., et al. (2017). Introduction to Data-Driven Methodologies for Prognostics and Health Management. Probabilistic Prognostics and Health Management of Energy Systems. S. Ekwaro-Osire, A. C. Gonçalves and F. M. Alemayehu. Cham, Springer International Publishing: 9-32.

Lee, J., et al. (2014). "Prognostics and health management design for rotary machinery systems—Reviews, methodology and applications." Mechanical Systems and Signal Processing 42(1): 314-334.

Maharaj, E. A. (1999). "Comparison and classification of stationary multivariate time series." Pattern Recognition 32(7): 1129-1138.

Martin-Diaz, I., et al. (2017). "Early Fault Detection in Induction Motors Using AdaBoost With Imbalanced Small Data and Optimized Sampling." IEEE Transactions on Industry Applications 53(3): 3066-3075.

Mosallam, A., et al. (2016). "Data-driven prognostic method based on Bayesian approaches for direct remaining useful life prediction." Journal of Intelligent Manufacturing 27(5): 1037-1048.

Patel, R. K. and V. K. Giri (2016). "Feature selection and classification of mechanical fault of an induction motor using random forest classifier." Perspectives in Science 8: 334-337.

Patil, S. S. and J. A. Gaikwad (2013). Vibration analysis of electrical rotating machines using FFT: A method of predictive maintenance. 2013 Fourth International Conference on Computing, Communications and Networking Technologies (ICCCNT).

Peng, Y., et al. (2010). "Current status of machine prognostics in condition-based maintenance: a review." The International Journal of Advanced Manufacturing Technology 50(1): 297-313.

Rao, B. (1996). Handbook of condition monitoring, Elsevier.

Saxena, A., et al. (2008). Damage propagation modeling for aircraft engine run-to-failure simulation. 2008 International Conference on Prognostics and Health Management.

Si, X.-S., et al. (2011). "Remaining useful life estimation – A review on the statistical data driven approaches." European Journal of Operational Research 213(1): 1-14.

Wang, T., et al. (2019). "A Self-Learning Fault Diagnosis Strategy Based on Multi-Model Fusion." Information 10(3): 116.

Yan, J. and J. Lee (2007). A Hybrid Method for On-line Performance Assessment and Life Prediction in Drilling Operations. 2007 IEEE International Conference on Automation and Logistics.

Yang, L. and J. Lee (2012). "Bayesian Belief Network-based approach for diagnostics and prognostics of semiconductor manufacturing systems." Robotics and Computer-Integrated Manufacturing 28(1): 66-74.

Yazdanbakhsh, O. and S. Dick (2019). "Multivariate Time Series Classification using Dilated Convolutional Neural Network." arXiv preprint arXiv:1905.01697.

Yu, D., et al. (2009). "Short-Term Solar Flare Prediction Using a Sequential Supervised Learning Method." Solar Physics 255(1): 91-105.

王燕 (2005). 应用时间序列分析, 中国人民大学出版社.

吳鴻材, et al. (2018). "預測維護技術的實施方法, 現況與發展趨勢." 機械工業雜誌(425): 76-82.