Project Overview
This project addresses the critical issue of road accidents in Morocco, where incidents have increased significantly in recent years. By leveraging historical data and machine learning, we've developed a predictive model that helps identify high-risk areas and factors contributing to accidents, ultimately aiming to reduce fatalities and improve road safety.
Key Achievements
30%
Reduction in predicted accidents in high-risk areas
10,000+
Accident records analyzed for model training
85%
Prediction accuracy for accident hotspots
Methodology
The predictive model analyzes comprehensive historical data with a focus on rural areas, employing advanced machine learning algorithms including Random Forest and K-Nearest Neighbors to identify patterns and risk factors contributing to accidents.
Our approach utilizes the CLASSE_ISR factor from the EURORAP program to classify road danger levels based on five years of accident statistics. The model incorporates multiple variables including driver behavior, vehicle conditions, and environmental factors, trained on 2022 accident data for optimal accuracy.
Pyramid of accident risk classification
ISR classification system for road safety
Data Visualizations
Interactive Power BI dashboards provide comprehensive insights into accident patterns, risk factors, and predictive model performance across different regions of Morocco.
Geospatial analysis of accident hotspots
Temporal trends in accident occurrences
Risk factor correlation analysis
Full Project Report
For complete details on methodology, algorithms, and findings, download the comprehensive project report.
Download Full ReportTools & Technologies
R
Power BI
Python
Excel
Anaconda
Contact Me
If you’d like to get in touch, feel free to send me a message.