This project develops an optimized fuel replenishment strategy for a fleet of continuously patrolling vehicles using genetic algorithms. The goal is to minimize operational costs and vehicle downtime over a 15-day period.
- Genetic algorithm optimization for refueling strategies
- Dynamic dispatch and gather strategies
- Adaptive refuel and emergency thresholds
- Cost comparison between optimized and baseline strategies
- K-means clustering for spatial analysis
- Population initialization
- Fitness evaluation
- Parent selection
- Crossover and mutation
- Elitism
- Solution optimization
- Significant cost reduction compared to baseline strategy
- Improved operational efficiency
- Enhanced vehicle management and resource allocation
- Real-time data processing integration
- Environmental factor consideration
- Multi-modal transportation expansion
- API development for broader application
- Python 3.x
- NumPy
- Pandas
- Matplotlib
- Scikit-learn
Special thanks to the Institute for Systems Studies and Analyses, Defence R&D Organisation, for supporting this research.
MIT License