Traffic Crash Analytics & Safety Intelligence Platform

Project Description

This project analyzes traffic crash data using SQL and Python.
The system uses SQLite as the database and Streamlit to build an interactive dashboard for traffic crash analysis.

The project helps identify:

• Crash trends
• Injury patterns
• High-risk streets
• Weather impact on crashes
• Dangerous time slots
• Hotspot zones

Interactive visualizations and advanced SQL queries are used to generate meaningful business insights.

---

Features

• Traffic crash trend analysis
• Injury analysis dashboard
• Weather impact analysis
• High-risk street identification
• Crash hotspot zone analysis
• Interactive Streamlit dashboard

Advanced SQL Concepts Used

• CTEs
• Window Functions
• Aggregations
• Ranking Functions

Dynamic Filters

• Year
• Month
• Weather condition
• Street name

---

Technologies Used

Technology	Purpose
Python	Data Analysis
SQL	Data Querying
SQLite	Database
Streamlit	Dashboard Development
Pandas	Data Manipulation
Matplotlib	Visualization

---

Project Structure

Project Structure: TRAFFIC_CRASH_PROJECT/ │ ├── app.py ├── README.md ├── traffic_crash.ipynb ├── traffic_crash.sqlite ├── Traffic_CrashesData.csv ├── traffic_crash.jpg │ ├── reports/ │ └── Traffic_Crash_Project_Report.pdf │ └── screenshots/ ├── Home Page 1.png ├── Home Page 2.png ├── Query Page.png ├── Visualisation 1.png └── Visualisation 2.png

---

Dataset Notice

The original CSV dataset and SQLite database files have been excluded from the GitHub repository due to large file size limitations.

How to Run the Project

Step 1: Clone the Repository

git clone https://github.com/your-username/TRAFFIC_CRASH PROJECT.git

Step 2: Open the Project Folder

cd TRAFFIC_CRASH PROJECT

Step 3: Run the Streamlit Application

streamlit run app.py

Step 4: Open in Browser

After running the command, Streamlit will automatically open in your browser.

---

Dashboard Screenshots

Home Page 1

Home Page 2

Visualisation 1

Visualisation 2

Query Page

---

Business Insights Generated

• Identified dangerous weather and crash type combinations
• Found high-risk streets with maximum injuries
• Analyzed peak crash hours
• Detected dangerous time buckets
• Compared daylight vs darkness injury severity
• Identified hotspot crash zones using latitude and longitude

---

Future Enhancements

• Machine Learning crash prediction
• Real-time traffic monitoring
• GIS map integration
• Cloud deployment
• Advanced forecasting models

---

Author

A. Harini

Skills

• Python
• SQL
• Streamlit
• Data Analysis
• Pandas