🚀 Falcon 9 Landing Prediction (Capstone Project)

🧠 Project Overview

This project aims to predict whether the first stage of the Falcon 9 rocket will successfully land or not. The ability to recover and reuse rocket stages significantly reduces launch costs. SpaceX lists Falcon 9 launches at $62 million, while competitors charge upwards of $165 million.
Knowing the likelihood of a successful landing can help estimate the true cost of a launch — a key advantage for competitors or clients in the growing commercial space sector.

As a fictional data scientist at "Space Y" (a competitor to SpaceX, founded by billionaire Allon Musk), you're tasked with using public data and machine learning to predict landing outcomes and estimate launch costs. This project is part of the IBM Data Science Professional Certificate.

✨ Please note: This project and its scenario are fictional and were completed as part of an AI-generated educational exercise. The content is structured to resemble a real-world application, but is intended for learning purposes only.

---

📚 Learning Objectives

• Manipulate and clean real-world launch data using Python and Pandas
• Access and process JSON data from a REST API
• Use web scraping to collect supplementary data from Wikipedia
• Perform exploratory data analysis (EDA) using visual tools
• Apply machine learning models to solve a binary classification problem
• Build an interactive dashboard for predictions and insights

---

🛰 Data Sources

• SpaceX REST API
• Wikipedia Falcon 9 Launch History pages
• Pre-cleaned datasets from Kaggle and provided materials

---

📈 Project Workflow

1. Data Collection

   • Gathered using SpaceX’s public API
   • Web scraping from Wikipedia tables for extra details

2. Data Wrangling & Preparation

   • Flattening nested JSON
   • Handling nulls and missing values
   • Filtering out irrelevant data (e.g., Falcon 1 launches)

3. Exploratory Data Analysis (EDA)

   • Understanding how features like payload, orbit, and site affect landings
   • Visualization using matplotlib, seaborn, and plotly

4. Machine Learning Modeling

   • Models used: Logistic Regression, Random Forest, SVM, XGBoost
   • Target: Predict if the rocket first stage will land (1) or not (0)
   • Evaluated using Accuracy, F1 Score, and Confusion Matrix

5. Cost Estimation Logic

   • If landing = success ➝ estimated cost ≈ $62M
   • If landing = failure ➝ estimated cost ≈ $165M

6. Interactive Dashboard (Optional)

   • Built using Streamlit or Plotly Dash
   • Input parameters, model predictions, and launch cost estimation

---

🧾 Key Features

• 🧠 Machine Learning-based prediction of Falcon 9 first-stage landing
• 💰 Cost estimation based on landing outcome
• 📊 Interactive data visualizations and insights
• 🔍 Real-world data wrangling, modeling, and storytelling

---

🤝 Contributors

This capstone project was developed as part of the IBM Data Science Professional Certificate.
I’m working as a fictional data scientist for the project scenario — this is an AI-generated learning exercise that simulates a real-world data science task.

Special thanks to Joseph Santarcangelo, Yan Luo, and Azim Hirjani, the instructors who guided this learning journey.

---

📂 Project Structure

alcon9-landing-prediction-ds/
├── data/                   # Collected and cleaned datasets
├── notebooks/              # Jupyter notebooks for each project stage
├── models/                 # Trained model files (optional)
├── dashboard/              # Streamlit or Dash app (optional)
├── utils/                  # Helper scripts and functions
└── README.md               # Project documentation

---

💡 Final Thoughts

This project combines real-world datasets, data wrangling, and machine learning to solve a business-relevant problem in the commercial space sector. While fictional, the process reflects practical steps taken in real data science work.