🧭 Project Overview
This project explores how payment type (Card vs Cash) impacts taxi fares and trip characteristics in New York City.
It provides data-driven insights to help drivers and companies maximize revenue by understanding passenger behavior and fare patterns.
🎯 Objective
To determine whether payment method affects the total fare amount and trip distance, and to recommend strategies that improve driver earnings and overall operational efficiency.
⚙ Tools & Technologies
- Programming Language: Python
- Libraries: Pandas, NumPy, Matplotlib, Seaborn, SciPy, Statsmodels
- Environment: Jupyter Notebook
- Reporting Tools: PowerPoint, PDF
📊 Dataset Details
Source: NYC Taxi Trip Records
Columns Used:
passenger_count, payment_type, fare_amount, trip_distance, pickup_datetime, dropoff_datetime
🧩 Project Workflow
- Data Cleaning – Removed missing, duplicate, and invalid records
- Exploratory Data Analysis (EDA) – Analyzed trip and fare distributions by payment type
- Statistical Testing (t-Test) – Checked whether fare differences between payment types are statistically significant
- Regression Analysis – Modeled fare as a function of trip distance and payment type
- Visualization – Created comparative charts for insights
- Reporting – Summarized results with visuals and business recommendations
🔍 Key Insights
Comparison of Card vs Cash payments in NYC Taxi:
| Metric | Card | Cash |
|---|---|---|
| Payment Share | 67.5% | 32.5% |
| Average Fare (USD) | 13.7 | 12.25 |
| Average Distance (mi) | 3.23 | 2.80 |
Insights:
- Card payments are more frequent and yield higher fares.
- t-Test Result: Statistically significant difference → payment type impacts fare.
- Drivers earn ~12% higher revenue per trip with card payments.
💡 Business Recommendations
- Offer rewards or incentives to encourage card payments
- Educate drivers on higher earnings potential via card transactions
- Simplify cashless payment process in the app for better adoption
- Track payment-type trends using dashboards or BI tools
🧠 Learnings
- Applied EDA, hypothesis testing, and regression modeling on real-world data
- Learned to convert raw data into actionable business insights
- Improved data visualization, interpretation, and communication skills
- Gained hands-on experience in end-to-end data analysis workflow
📘 Project Impact
- Discovered that payment type significantly affects fare amount and trip behavior
- Provided actionable recommendations for increasing driver revenue
- Delivered insights in a clean, reproducible, and business-focused format
- Demonstrated practical ability to combine data analytics with business understanding
📂 Supporting Files
For reference and verification
- 📘 Jupyter Notebook – Analysis Code
- 📊 Dataset (CSV File)
- 🧾 Presentation – Summary Report (PPT)
- 📄 Final PDF Report
👩💻 Author
Swati Mirashi
📧 swatimirashi298@gmail.com
🔗 LinkedIn | GitHub
⭐ If you found this project insightful, give it a star!