MicroForest: Lightweight Bottleneck Prediction for Manufacturing Processes on Edge Devices
Source abstract
Abstract
As digital transformation in manufacturing accelerates, process bottleneck prediction has emerged as a central task in industrial automation. To streamline manufacturing processes, where diverse tasks interact in complex ways, it is essential to identify in advance both the location and timing of bottleneck occurrences. However, manufacturing environments often lack high-performance computing resources and must rely on cost-effective, resource-constrained embedded devices, making fast and accurate prediction challenging. We present MicroForest, a lightweight decision tree-based model designed to predict multiple process bottlenecks simultaneously under such resource-constrained environments. MicroForest reassembles the high-information-gain nodes from dozens of large random forests into compact forests. Evaluated on a simulation containing up to 150 production tasks, MicroForest achieves 34%p higher recall scores compared to original random forests while shrinking model size by two orders of magnitude and accelerating inference latency by up to 7.2×. Compared with other recent work, MicroForest outperforms them with the highest prediction accuracy (F1 = 0.74) and shows a much gentler increase in latency as process complexity grows.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Scheduling and Optimization Algorithms · Industrial Vision Systems and Defect Detection · Digital Transformation in Industry
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Computer Engineering, Kongju National University, Cheonan-si 31080, Chungcheongnam-do, Republic of Korea
Department of Software, Kongju National University, Cheonan-si 31080, Chungcheongnam-do, Republic of Korea
Funding
MSIT (Ministry of Science, ICT), Korea
References (24)
Show source-provided references
1. Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q., and Liu, T.Y. (2017, January 4–9). LightGBM: A Highly Efficient Gradient Boosting Decision Tree. Proceedings of the Advances in Neural Information Processing Systems, Long Beach, CA, USA.
2. Chen, T., and Guestrin, C. (2016, January 13–17). XGBoost: A Scalable Tree Boosting System. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA.10.1145/2939672.2939785 ↗
3. Rehill, P. (2024). Distilling interpretable causal trees from causal forests. arXiv.
4. Farhadi, Z., Bevrani, H., Feizi-Derakhshi, M.R., Kim, W., and Ijaz, M.F. (2022). An Ensemble Framework to Improve the Accuracy of Prediction Using Clustered Random-Forest and Shrinkage Methods. Appl. Sci., 12.10.3390/app122010608 ↗
5. Zhu, N., Zhu, C., Zhou, L., Zhu, Y., and Zhang, X. (2022). Optimization of the Random Forest Hyperparameters for Power Industrial Control Systems Intrusion Detection Using an Improved Grid Search Algorithm. Appl. Sci., 12.10.3390/app122010456 ↗
6. Liu, C.Y., Ku, C.Y., Wu, T.Y., and Ku, Y.C. (2024). An Advanced Soil Classification Method Employing the Random Forest Technique in Machine Learning. Appl. Sci., 14.10.3390/app14167202 ↗
7. Cao, Y., Zhao, P., Xu, B., and Liang, J. (2024). An Improved Random Forest Approach on GAN-Based Dataset Augmentation for Fog Observation. Appl. Sci., 14.10.20944/preprints202410.0725.v1 ↗
8. Nan, F., Wang, J., and Saligrama, V. (2016, January 5–10). Pruning random forests for prediction on a budget. Proceedings of the Advances in Neural Information Processing Systems, Barcelona, Spain.
9. Ren, S., Zhao, W., Zhou, X., and Sun, J. (2015, January 7–12). Global Refinement of Random Forest. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Boston, MA, USA.
10. Ensemble of optimal trees, random forest and random projection ensemble classification10.1007/s11634-019-00364-9 ↗
11. Forest pruning based on branch importance10.1155/2017/3162571 ↗
12. Improving Land Cover Classification in an Urbanized Coastal Area by Random Forests: The Role of Variable Selection
13. Node harvest10.1214/10-AOAS367 ↗
14. PnT: Born-again tree-based model via fused decision path encoding10.1016/j.inffus.2024.102545 ↗
15. Lossless compression of random forests10.1007/s11390-019-1921-0 ↗
16. Buciluǎ, C., Caruana, R., and Niculescu-Mizil, A. (2006, January 20–23). Model Compression. Proceedings of the 12th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Philadelphia, PA, USA.10.1145/1150402.1150464 ↗
17. Hinton, G., Vinyals, O., and Dean, J. (2015). Distilling the Knowledge in a Neural Network. arXiv.
18. Vidal, T., Schiffer, M., and Pacheco, T. (2020, January 13–18). Born-Again Tree Ensembles. Proceedings of the 37th International Conference on Machine Learning, Online.
19. Zhou, Y., and Hooker, G. (2016). Interpreting Models via Single Tree Approximation. arXiv.
20. Bénard, C., Biau, G., Da Veiga, S., and Scornet, E. (2021, January 13–15). Interpretable Random Forests via Rule Extraction. Proceedings of the 24th International Conference on Artificial Intelligence and Statistics, Virtual.10.1214/20-EJS1792 ↗
21. Alkhoury, F., and Welke, P. (2024). Splitting Stump Forests: Tree Ensemble Compression for Edge Devices. International Conference on Discovery Science, Springer.10.1007/978-3-031-78980-9_1 ↗
22. Explainable Decision Forest: Transforming a Decision Forest into an Interpretable Tree10.1016/j.inffus.2020.03.013 ↗
23. Tree-based ensembles for multi-output regression: Comparing multivariate approaches with separate univariate ones10.1016/j.csda.2022.107628 ↗
24. IDA2016Challenge. UCI Machine Learning Repository.