Route Optimization with LSTM and Graphs

About the Project

This project uses LSTM Neural Networks and Graphs to predict travel times throughout the day and optimize routes. The goal is to create a model that analyzes traffic patterns and suggests the fastest route between points of interest.

Features

• Simulation of traffic data based on daily variations.
  
• Training of an LSTM model for travel time prediction.
  
• Construction of a Weighted Graph representing the routes.
  
• Application of the Dijkstra Algorithm to find the best route.
  
• Graphical visualization of variations and the route network.

Technologies Used

• Python (3.9.21)
• NumPy, Pandas (Data manipulation)
• Matplotlib (Visualization)
• NetworkX (Graphs and Dijkstra)
• Scikit-learn (Data normalization)
• TensorFlow/Keras (LSTM model)

Run the Code

python routes.py

Generated Visualizations

• Traffic forecast graph
• Optimized route map with travel times