🚗 Getaround — Delay Analysis & Pricing Prediction
Certification CDSD — Data Science & Deployment Project
📌 Project Overview
Getaround is a peer-to-peer car rental platform where late vehicle returns can create strong friction for subsequent rentals. This project addresses two strategic challenges:
Operational optimization — evaluating the impact of introducing a minimum delay between rentals to reduce late checkout conflicts.
Pricing optimization — exposing a Machine Learning model through an online API to help owners set optimal rental prices.
The project combines data analysis, interactive dashboards, machine learning, and API deployment.
🎯 Business Objectives Delay Management
Measure how often drivers return cars late.
Quantify the impact on subsequent rentals.
Simulate different minimum delay thresholds.
Help Product Management choose:
an optimal delay threshold
an appropriate scope (all cars vs Connect only).
Pricing Optimization
Serve a trained Machine Learning model via an online API.
Allow real-time price prediction through a /predict endpoint.
Provide a simple UI to interact with the model.
🧱 Project Structure
Project_GetAround/
│
├── dashboards/
│ ├── app_delay.py # Streamlit dashboard — Delay analysis
│ └── app_pricing.py # Streamlit dashboard — Pricing prediction
│
├── api/
│ ├── main.py # FastAPI app
│ ├── model.pkl # Trained ML model
│ └── requirements.txt
│
├── notebooks/
│ ├── 01_delay_analysis.ipynb
│ └── 02_pricing_model.ipynb
│
├── README.md
└── requirements.txt
📊 Dashboard 1 — Delay Analysis Purpose
Help Product Managers evaluate the trade-off between:
reducing late checkout conflicts
preserving rental revenue
Features
Threshold selection (0–180 minutes)
Scope selection (check-in types)
Key KPIs:
% of late checkouts
number of impacted rentals
Delay distribution visualization
Business impact summary
Technology
Streamlit
Pandas
Plotly
Run locally streamlit run dashboards/app_delay.py
💰 Dashboard 2 — Pricing Prediction Purpose
Provide an interface to interact with the pricing prediction API.
Features
Manual input of 11 numerical features
API call to /predict
Real-time price prediction display
Input validation and error handling
Run locally streamlit run dashboards/app_pricing.py
🤖 Machine Learning API Endpoint: /predict
Method: POST
Input:
{ "input": [[7.0, 0.27, 0.36, 20.7, 0.045, 45.0, 170.0, 1.001, 3.0, 0.45, 8.8]] }
Output:
{ "prediction": [6] }
Documentation
A full API documentation is available at:
/docs
🌐 Deployment
API: Hosted on Hugging Face Spaces
Dashboards: Run locally or deployable via Streamlit Cloud / Hugging Face
Example API URL https://-.hf.space/predict
🛠️ Installation
git clone https://github.com//Project_GetAround.git
cd Project_GetAround
python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
🧠 Key Takeaways
Data-driven product decision support
Clear separation between analysis, ML, and deployment
Robust handling of user inputs and API failures
Production-oriented mindset aligned with CDSD expectations
👤 Author
Frédéric
CDSD Candidate — Data Scientist
Project completed as part of the Jedha CDSD certification.