Hugging Face's logo

Datasets:
foundation-models
/
golden-batch-sentinel-data

Tasks:
Time Series Forecasting
Tabular Classification
Modalities:
Tabular
Formats:
parquet
Size:
10M - 100M
Tags:
bioprocess
manufacturing
anomaly-detection
fault-detection
batch-monitoring
pharma
Libraries:
Datasets
Dask
Polars
License:
Dataset card Data Studio Files
xet
Community
1
golden-batch-sentinel-data
File size: 2,881 Bytes 
a0b6884
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
 
---
license: cc-by-4.0
task_categories:
  - time-series-forecasting
  - tabular-classification
tags:
  - bioprocess
  - manufacturing
  - anomaly-detection
  - fault-detection
  - batch-monitoring
  - pharma
pretty_name: Golden Batch Sentinel Data
size_categories:
  - 1M<n<10M
---

# Golden Batch Sentinel Data

Benchmark datasets for process monitoring and fault detection in batch manufacturing.

## Datasets

### IndPenSim (Industrial Penicillin Simulation)

A 100,000L fermentation simulation with 100 batches and rich multivariate signals.

- **Source**: [Mendeley Data](https://data.mendeley.com/datasets/pdnjz7zz5x/2)
- **Paper**: [Modern day monitoring and control challenges...](https://doi.org/10.1016/j.compchemeng.2018.05.019)
- **Batches**: 100 (90 normal, 10 faulty)
- **Variables**: 37 process variables (Raman spectra excluded for efficiency)
- **Time resolution**: 0.2 hours

**Files:**
- `indpensim/batches.parquet` - Main batch data
- `indpensim/statistics.parquet` - Batch statistics and fault labels

### Tennessee Eastman Process (TEP)

The most common benchmark for fault detection in multivariate industrial processes.

- **Source**: [Harvard Dataverse](https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/6C3JR1)
- **Fault types**: 20 different fault scenarios
- **Variables**: 52 (41 measured + 11 manipulated)

**Files:**
- `tep/fault_free_train.parquet` - Normal operation (training)
- `tep/fault_free_test.parquet` - Normal operation (testing)
- `tep/faulty_train.parquet` - Faulty operation (training, all 20 faults)
- `tep/faulty_test.parquet` - Faulty operation (testing, all 20 faults)

## Usage

```python
from datasets import load_dataset

# Load IndPenSim
indpensim = load_dataset("foundation-models/golden-batch-sentinel-data", data_dir="indpensim")

# Load TEP
tep = load_dataset("foundation-models/golden-batch-sentinel-data", data_dir="tep")

# Or load specific files
import pandas as pd
from huggingface_hub import hf_hub_download

path = hf_hub_download(
    repo_id="foundation-models/golden-batch-sentinel-data",
    filename="indpensim/batches.parquet",
    repo_type="dataset"
)
df = pd.read_parquet(path)
```

## License

The original datasets are provided under their respective licenses:
- IndPenSim: CC BY 4.0
- TEP: Public domain

This compilation is provided under CC BY 4.0.

## Citation

If you use this data, please cite the original papers:

```bibtex
@article{goldrick2019modern,
  title={Modern day monitoring and control challenges outlined on an industrial-scale benchmark fermentation process},
  author={Goldrick, Stephen and others},
  journal={Computers \& Chemical Engineering},
  year={2019}
}

@article{downs1993plant,
  title={A plant-wide industrial process control problem},
  author={Downs, James J and Vogel, Ernest F},
  journal={Computers \& Chemical Engineering},
  year={1993}
}
```