Skip to content

onurfgg/multi-depot-vrp-optimizer

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 

Repository files navigation

🚚 Multi-Depot VRP Optimizer & Logistics DSS

HTML5 CSS3 JavaScript Operations Research Algorithms

The Multi-Depot VRP Optimizer is an interactive Decision Support System (DSS) designed for supply chain and logistics management. It solves the complex Multi-Depot Vehicle Routing Problem (MDVRP) by clustering customers to their nearest distribution centers and optimizing each route using heuristic algorithms.

🔗 Explore Live Simulation (Live Demo)

📊 Mathematical Model & Algorithms

This tool does not rely on third-party routing APIs; it is built entirely on custom mathematical models in JavaScript. The engine processes the network in three phases:

  1. Clustering (Distance-Based): The algorithm scans the Cartesian plane and assigns every customer node to its nearest depot node using Euclidean distance calculations.
  2. Initial Routing (Nearest Neighbor): For each isolated cluster, the system quickly constructs an initial, feasible route by continuously navigating to the nearest unvisited node.
  3. Local Search Optimization (2-Opt Heuristic): The system detects intersecting (crossed) routes and systematically swaps edges (twoOptSwap) to untangle the graph. This iterative improvement minimizes the total travel distance until a local optimum is reached.

🌟 Highlighted Features

  • Multi-Depot Support: Users can place multiple depots on the map. The system intelligently partitions the workload among available distribution centers.
  • Cartesian Grid & Visualizer: A built-in custom <canvas> engine renders the nodes on a scaled coordinate grid, simulating a real-world geographic distribution.
  • Dynamic Cost Function ($Z$): Users can input custom transportation unit costs. The system calculates the estimated total transportation cost in real-time.
  • Algorithm Savings Metric: The dashboard displays the exact percentage of distance saved by the 2-Opt algorithm compared to the initial Nearest Neighbor route.
  • Live Animation: The pathfinding process is visually animated, allowing users to watch the routes being constructed and optimized dynamically.

⚙️ Installation and Usage

The application is a 100% client-side web tool. No backend configuration is needed.

  1. Clone the repository:
    git clone [https://github.com/onurfgg/BU_KISMA_REPO_ADINI_YAZ.git](https://github.com/onurfgg/multi-depot-vrp-optimizer.git)
    
    

Onur Furkan Gök

Linkedin GitHub

About

An interactive Decision Support System for the Multi-Depot Vehicle Routing Problem (MDVRP). Utilizes nearest-neighbor and 2-Opt heuristic algorithms to cluster, optimize, and animate logistics networks on a Cartesian grid.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages