VRP

In this project, I tackled a Vehicle Routing Problem (VRP), a challenging optimization task in logistics. The scenario involved efficiently routing a fleet of vehicles from a central depot to deliver goods to 10 customers, each with specific demands. What made this problem intriguing was its combination of real-world applicability and algorithmic complexity.

Key technical challenges included designing an edge selection mechanism that considered pheromone levels, distance heuristics, and vehicle capacity constraints. I also implemented a pheromone update system and an early stopping mechanism for efficiency.

The project required a delicate balance of algorithm design, parameter tuning, and problem-specific adaptations. It was particularly satisfying to see the algorithm converge rapidly to a cost-effective solution, demonstrating the power of nature-inspired computing in solving practical business problems.

The project can be found on Github.