https://github.com/nurulashraf/predictive-maintenance-analysis-for-machine-failure-prevention
Predictive maintenance analysis for machine failure prevention using sensor data and ML. Built a Random Forest model and Gradio dashboard to identify high-risk machines for proactive maintenance.
https://github.com/nurulashraf/predictive-maintenance-analysis-for-machine-failure-prevention
data-science failure-prediction gradio industrial-iot machine-learning power-bi predictive-maintenance python scikit-learn
Last synced: about 2 months ago
JSON representation
Predictive maintenance analysis for machine failure prevention using sensor data and ML. Built a Random Forest model and Gradio dashboard to identify high-risk machines for proactive maintenance.
- Host: GitHub
- URL: https://github.com/nurulashraf/predictive-maintenance-analysis-for-machine-failure-prevention
- Owner: nurulashraf
- License: mit
- Created: 2025-10-04T16:55:54.000Z (8 months ago)
- Default Branch: main
- Last Pushed: 2025-10-15T11:17:58.000Z (8 months ago)
- Last Synced: 2025-10-16T04:20:03.984Z (8 months ago)
- Topics: data-science, failure-prediction, gradio, industrial-iot, machine-learning, power-bi, predictive-maintenance, python, scikit-learn
- Language: Jupyter Notebook
- Homepage:
- Size: 331 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Predictive Maintenance Analysis for Machine Failure Prevention
Predictive maintenance analysis using sensor data and machine learning. This project builds a Random Forest model and a Gradio dashboard to identify high-risk machines for proactive maintenance.
---
## Project Structure
- `data/` — contains datasets used in the analysis (raw, cleaned, processed)
- `notebooks/` — Jupyter notebooks for data exploration, model building, and dashboard prototyping
- `requirements.txt` — list of Python dependencies
- `LICENSE` — the licensing file
- `README.md` — this documentation
---
## Features
- Exploratory data analysis and visualization of sensor data
- Data cleaning, feature engineering, outlier handling
- Training of a **Random Forest** classifier to detect machines at risk of failure
- Model evaluation metrics: accuracy, precision, recall, F1 score, ROC curve
- Gradio-based dashboard for interactive inference / risk scoring
- Ability to flag high-risk machines to inform maintenance decisions
---
## Tools & Libraries
- `pandas`, `numpy` - Data handling
- `matplotlib`, `seaborn` - Machine learning
- `scikit-learn` - Machine learning
- `gradio` - Interactive dashboard
- `jupyter` - Notebook environment
---
## How to Use
### Setup
1. Clone the repository:
```bash
git clone https://github.com/nurulashraf/predictive-maintenance-analysis-for-machine-failure-prevention.git
cd predictive-maintenance-analysis-for-machine-failure-prevention
````
2. Create a virtual environment (recommended):
```bash
python3 -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
```
3. Install dependencies:
```bash
pip install -r requirements.txt
```
4. **Open notebook**
Launch Jupyter:
```bash
jupyter notebook
```
or
```bash
jupyter lab
```
Then open the notebook in the `notebooks/` folder.
The dashboard allows users to input sensor features and get a risk prediction (e.g. “high risk” or probability) for machine failure.
---
## License
This project is licensed under the **MIT License**. See the [LICENSE](LICENSE) file for full details.