🛢️ Well Production Forecasting Dashboard

A Streamlit-based machine learning dashboard for visualizing, cleaning, engineering, training, and forecasting oil & gas well production.

🚀 Overview

The Well Production Forecasting Dashboard is an end-to-end machine learning application designed for petroleum engineers, reservoir analysts, and data scientists. It enables you to:

• Import custom datasets or use included real-world samples
• Visualize production trends with interactive plots
• Prepare and process data for analysis and modeling
• Build CART machine-learning models
• Produce accurate well production forecasts

The app is organized into five interactive Streamlit tabs, offering a smooth and guided workflow from raw data to final prediction.

📁 Project Structure

production-forecasting-dashboard/
│
├── app/
│   ├── main.py               # Main Streamlit app with all tabs
│   ├── upload_tab.py         # Dataset upload + built-in dataset selector
│   ├── data_eng_tab.py       # Visualization + preprocessing tools
│   ├── train_tab.py          # CART model training + evaluation
│   ├── prediction_tab.py     # Prediction using trained model
│
├── data/                     # Included real-world ND datasets
│   ├── ND_cumulative_formation_2020.xlsx
│   ├── ND_gas_1990_to_present.xlsx
│   ├── ND_historical_barrels_of_oil_produced_by_county.xlsx
│   └── ND_historical_MCF_gas_produced_by_county.xlsx
│
├── src/
│   ├── plots.py              # Unified Seaborn + Matplotlib plotting utilities
│   └── model.py              # CART model builder, evaluator, saver
│
├── strings/
│   └── strings.py            # UI messages and text constants
│
├── README.md
├── requirements.txt
└── LICENSE

🧠 Features

🔹 1. Dataset Handling

• Upload custom .xlsx files
• Select from four included North Dakota datasets
• Automatic dataset summary and preview
• Preprocessing support for modeling and visualization

🔹 2. Exploratory Data Analysis

The dashboard includes 15+ interactive plot types, grouped into:

📊 Distribution Plots

• Distribution plot
• Histogram (configurable bins)
• KDE
• ECDF
• Rug plot

🧩 Categorical Plots

• Catplot
• Strip plot
• Swarm plot
• Box plot
• Violin plot
• Point plot
• Bar plot

🔗 Relational Plots

• Scatter plot
• Line plot

All plots use clean Seaborn + Matplotlib visuals optimized for Streamlit.

🤖 Machine Learning (CART)

The app supports CART decision tree models for both regression and classification.

✔ Automatically handles:

• Numeric columns
• Date/time formatting
• Label encoding for non-numeric features

✔ Model evaluation includes:

• Accuracy (classification)
• MSE & R² (regression)

Models can be trained, evaluated, and saved locally for later predictions.

🔮 Prediction

• Generate predictions using trained CART models
• Interactive input forms
• Downloadable prediction results

▶️ How to Run the App

1️⃣ Clone the repository

git clone https://github.com/sobhankohanpour/production-forecasting-dashboard.git
cd production-forecasting-dashboard

2️⃣ Install dependencies

pip install -r requirements.txt

3️⃣ Launch the Streamlit app

streamlit run app/main.py

📦 Dependencies

Requires Python 3.8+

Core libraries:

• streamlit
• pandas
• matplotlib
• seaborn
• scikit-learn
• joblib

📘 Included Datasets

The data/ directory includes curated North Dakota production datasets:

• Cumulative Oil Production by Formation (2020)
• Gas Production (1990–Present)
• Historical Monthly Oil Production by County
• Historical Monthly Gas Production by County

All datasets are directly accessible from within the dashboard.

🧩 Plot Utilities (src/plots.py)

Reusable plotting functions include:

• scatterplot()
• lineplot()
• distplot(), histplot(), kdeplot()
• ecdfplot(), rugplot()
• catplot(), stripplot(), swarmplot()
• boxplot(), violinplot()
• pointplot(), barplot()

Easy to extend for custom visualizations.

📄 License

MIT License — free for personal and commercial use.

🤝 Contributing

Contributions, enhancements, and feature requests are welcome! Feel free to open an issue or submit a pull request.

⭐ Support

If you find this project useful, please consider giving it a star ⭐ on GitHub.