The dataset is released as part of the academic research conducted for the master’s thesis entitled:
Design and Value Realization of a Digital Solution for Tightening Process Quality Management.
Author: YANG Zhenye (yang.zhenye@foxmail.com or zhen.ye.yang@atlascopco.com)
Supervisor: TIAN Yingjie
Institution: University of Chinese Academy of Sciences(UCAS)
Date: June 2026
DOI: 10.27611/d.cnki.gzkyy.2026.000001
This is a real-world industrial dataset designed for anomaly analysis in threaded fastener tightening processes. The dataset is constructed from actual tightening operation data collected at industrial production sites, where synchronized torque–angle time-series signals are recorded during each tightening process. It aims to support research on anomaly detection, quality assessment, and time-series modeling in industrial tightening applications.
To ensure consistency in process conditions, each sub-dataset is collected from the same workstation, the same tightening tool, and the same tightening program, while different sub-datasets correspond to different tightening joints or operating conditions. For representative anomaly types within each dataset, a SMALL number of samples are manually labeled based on engineering expertise, providing reference labels for algorithm validation and analysis. Since multiple abnormal phenomena may occur simultaneously during a single tightening operation, the dataset adopts a multi-label annotation scheme to better reflect the complexity of real industrial scenarios.
This dataset is intended for research on anomaly detection, weakly supervised learning, and industrial time-series analysis, and can be used for both academic research and engineering-oriented method evaluation.
| Dataset | Method | Total | A1 | A2 | A3 | A4 | A5 | A6 | A7 |
|---|---|---|---|---|---|---|---|---|---|
| D1 | T | 22413 | 24 | / | 4 | 46 | 4 | 24 | / |
| D2 | T | 10000 | 12 | / | / | / | 12 | 6 | / |
| D3 | T | 4996 | / | 11 | 5 | 36 | 2 | 8 | / |
| D4 | TA | 10000 | 15 | / | 34 | / | / | / | / |
| D5 | T | 51 | / | / | / | / | / | / | 1 |
| D6 | TA | 4933 | / | 8 | / | / | / | / | / |
| D7 | T | 10000 | 21 | / | / | / | 2 | 14 | / |
| D8 | T | 40000 | 24 | / | 50 | / | 12 | / | / |
Note:
- Dataset D6 corresponds to yield tightening. Therefore, part of the necking-related samples are labeled as A2.
- T: Torque only method; TA: Torque+Angle method.
- Co-rotation
- Thread stripping / Head rounding
- Thread crossing / Jamming / Floating / Seizing
- Stiffness variation / Gap closure
Due to GitHub size limits, raw data is not stored in this repository, raw data is available at:
[https://zenodo.org/records/18007039]
This dataset is licensed under the
Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) license.
You are free to use, share, and adapt the data for non-commercial purposes, provided that appropriate credit is given.
For commercial use, please contact the author.