Skip to content

Latest commit

 

History

History
53 lines (45 loc) · 2.24 KB

README.md

File metadata and controls

53 lines (45 loc) · 2.24 KB

TSP

Metaheuristic Algorithms for Solving Symmetric Traveling Salesperson Problem (TSP). This project is an evaluation project for several metaheuristic algorithms on swarm intelligence. The selected algorithims are:

  • Hydrological Cycle Algorithm (HCA).
  • Water Flow-Like Algorithm.
  • Intelligent Water Drops Algorithm (IWD).
  • Rain Water Algorithm.
  • River Formation Dynamics.

How to use?

The following code demonstrates an example of using the HCA algorithm on RL1304 dataset. To know more about the Graph and the dataset details, you should read this page first.

package jo.ju.edu.tsp;  
  
import jo.ju.edu.tsp.algorithms.HCA;  
import jo.ju.edu.tsp.algorithms.TSP;  
import jo.ju.edu.tsp.set.SetInstance;  
import org.xml.sax.SAXException;  
  
import javax.xml.parsers.ParserConfigurationException;  
import java.io.IOException;  
  
public class Main {  
    public static void main(String[] args) {  
        TSP tsp = new HCA();  
        try {  
            tsp.solve(SetInstance.RL1304);  
        } catch (ParserConfigurationException e) {  
            e.printStackTrace();  
        } catch (SAXException e) {  
            e.printStackTrace();  
        } catch (IOException e) {  
            e.printStackTrace();  
        }  
    }  
}

Systems Prerequisites

  • JDK 1.8+
  • Maven
  • Git - Optional in case of download the project

Run the project

  • Download or clone the project
  • Open the source code in an IDE; for example, IntelliJ
  • Go to any algorithm; for example, to run the HCA, you want /tsp/src/main/java/jo/ju/edu/tsp/algorithms/HCA.java
  • Right click in the HCA class -> Click on "Run'HCA.main()'" (Same thing for IWD and WFA java files)

The Team

  • Ahmad Al Mughrabi.
  • Raneem abuzayed.
  • Ali Al Sarhan
  • Nancy