The Travelling Salesman Problem has no optimal solution, but here we proposed a solution that takes advantage from the Google Maps Api to find one of the best routes, which is not necessary the best. The algorithm generates a connected graph and test some of the posible ways to travel arround.

This algorithm was developed to work inside real life cities, so it takes in count the traffic. It tries to make the travel in different schedules, which adds a grade of difficulty to the problem. Our solution finds some of the best solutions without traffic and after it test the different solutions between a given space of time, the 'schedule'.

• Using the distance with no traffic between the nodes, a connected graph is built.
• To allow circuits around the graph the 'Divided groups' and 'Bow tie' problems are solved creating additional connections.
• Then all the posible ways to travel around the graph are generated.
• Finally routes are tested at the given starting time to discover how much time it takes to pass thru all.

First you need to get a Google Maps Api Authentication Key: https://developers.google.com/maps/documentation/javascript/get-api-key

Check the included test cases. For more information in how to use the code, go to this repository wiki.