A Genetic Algorithm to solve the Traveling Salesman Problem with Web Workers by Francis Shanahan


Details: Solving the Traveling Salesman Problem with Genetic Algorithms and HTML5 Web Workers
Usage: This demo requires a browser capable of serializing JSON to a Web Worker. At this time (10/9/2010) only Chrome 7 beta, Firefox 4+ and Opera 10+ work.
Adjust the number of cities and click "Randomize Cities". 50 is a pretty good number.
Click "Evolve" to improve the solution. The number of Generations controls how many new generations are created. Each generation improves on the solution. Unfortunately the algorithm never knows when no further improvement is possible.
Keep clicking "Evolve" until you're happy with the solution.
Because GAs are not guaranteed to arrive at the best solution, you can try "Start a NEW Solution w/same Cities" to start fresh with the current set of cities and see if it arrives at an even better solution.

When you're done playing, go back and read the details HERE.
Num Cities:


Num Generations: