Exploratory Data Analysis – Optimising NYC Taxi Operations

Project Overview

This project performs an Exploratory Data Analysis (EDA) on the NYC Taxi dataset to identify operational inefficiencies, understand trip patterns, and generate actionable insights for optimizing taxi services.

Using Python and data visualization techniques, the analysis uncovers trends in trip demand, trip duration, passenger behavior, fare distribution, and peak operating hours to support data-driven decision-making.

---

Business Objectives

• Analyze taxi trip demand and usage patterns.
• Identify peak hours and high-demand periods.
• Examine trip duration and distance distributions.
• Analyze fare and payment trends.
• Generate recommendations to improve operational efficiency.

---

Tools & Technologies

• Python
• Jupyter Notebook
• Pandas
• NumPy
• Matplotlib
• Seaborn

---

Project Workflow

1. Data Cleaning

• Removed missing and duplicate records.
• Handled inconsistent values.
• Converted data into appropriate formats.

2. Exploratory Data Analysis

• Univariate and bivariate analysis.
• Distribution analysis.
• Correlation analysis.
• Outlier detection.

3. Data Visualization

• Trip demand trends
• Fare distribution
• Passenger count analysis
• Pickup and drop-off patterns
• Time-based analysis

4. Business Recommendations

• Identified opportunities to optimize taxi operations.
• Suggested improvements based on customer demand and operational trends.

---

Key Analysis

• Trip Frequency Analysis
• Trip Duration Analysis
• Fare Distribution
• Passenger Count Distribution
• Peak Hour Analysis
• Correlation Between Variables
• Operational Performance Assessment

---

Business Insights

• Peak demand occurs during specific hours of the day, indicating opportunities for better driver allocation.
• Fare distribution highlights the majority of trips fall within a predictable price range.
• Passenger count analysis shows most trips involve one or two passengers.
• Understanding demand patterns can improve fleet utilization and reduce passenger wait times.

---

Project Deliverables

This repository contains:

• Jupyter Notebook (.ipynb) – Complete exploratory data analysis and visualizations.
• Project Report (PDF) – Summary of methodology, findings, and business recommendations.
• Dataset Archive (.zip) – NYC Taxi dataset used for analysis.

---

Repository Structure

EDA_Optimising_NYC_Taxis
│
├── EDA_Assg_NYC_Taxi.ipynb
├── EDA_Optimising_NYC_Taxis.zip
├── Report_NYC_Taxi_Operations_Starter.pdf
└── README.md

---

Skills Demonstrated

• Exploratory Data Analysis (EDA)
• Data Cleaning
• Data Wrangling
• Statistical Analysis
• Data Visualization
• Business Analysis
• Python Programming
• Pandas
• NumPy
• Matplotlib
• Seaborn
• Data Storytelling

---

Future Enhancements

• Build an interactive Power BI dashboard.
• Perform predictive demand forecasting.
• Analyze seasonal and weather-based demand patterns.
• Develop machine learning models for trip demand prediction.

---

Author

Kartikey Singh

Data Analyst | Power BI | Python | SQL | Excel

LinkedIn: Kartikey_Singh

GitHub: Kartikey_Singh

Portfolio : Kartikey_Singh

Complete WriteUp: Coming Soon....

LinkedIn: Kartikey_Singh

GitHub: Kartikey_Singh