Summary of Time-series Forecasting in Smart Manufacturing Systems: An Experimental Evaluation Of the State-of-the-art Algorithms, by Mojtaba A. Farahani et al.
Time-Series Forecasting in Smart Manufacturing Systems: An Experimental Evaluation of the State-of-the-art Algorithms
by Mojtaba A. Farahani, Fadi El Kalach, Austin Harper, M. R. McCormick, Ramy Harik, Thorsten Wuest
First submitted to arxiv on: 26 Nov 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A recent study aims to bridge a crucial gap in Time Series Forecasting (TSF) by evaluating top-performing TSF algorithms on 13 manufacturing datasets. The evaluation focuses on the applicability of these algorithms in manufacturing, considering two problem categories and two forecasting horizons. The performance is measured using the Weighted Absolute Percentage Error (WAPE), with post-hoc analyses to assess significance. To ensure usability, only open-source libraries were used, and no hyperparameter tuning was performed. The results show that transformer and MLP-based architectures excel in univariate TSF, while N-HITS and TiDE perform well in multivariate problems. Interestingly, simpler algorithms like XGBoost outperform complex ones in certain tasks, challenging the assumption that more sophisticated models produce better results. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Time Series Forecasting is important for manufacturing because it helps predict what will happen in the future. Researchers have developed many new algorithms to do this, but nobody has checked which one works best. This study did just that by testing 13 different algorithms on 13 real-world datasets from manufacturing. The results show that some algorithms are better than others at predicting certain types of data. For example, some algorithms are great at predicting what will happen in the short-term, while others do better with longer-term predictions. The study also found that simpler algorithms can sometimes be just as good as more complicated ones. This research is important because it helps us understand which algorithms are best to use in different situations. |
Keywords
» Artificial intelligence » Hyperparameter » Time series » Transformer » Xgboost
