An application of fruit fly optimization algorithm for traveling salesman problem

In this study, an application of fruit fly optimization algorithm (FOA) is presented. FOA is one of the recently proposed swarm intelligence optimization algorithms used to solve continuous complex optimization problems. FOA has been invented by Pan in 2011 and it is based on the food search behavior of fruit flies. The FOA has a simple framework and it is easy to implement for solving optimization problem with different characteristics. The FOA is also a robust and fast algorithm and some researchers used FOA to solve discrete optimization problems. In this study, a new modified FOA is proposed for solving the well-known traveling salesman problem (TSP) which is one of the most studied discrete optimization problems. In basic FOA, there are two basic phases, one of them is osphresis phase and the other is vision phase. In the modified version of FOA the ospherisis phases kept as it is and for vision phase two different methods developed. In vision phase, the first half of the city arrangement matrix is updated according to first %30 part of best solutions of the ospheresis phase. The other half of the city arrangement matrix is randomly reproduced because of the possibility that initial solutions are far from the optimum. According to the results, travelling salesman problem can be solved with FOA as an alternative method. For big scale problems, it needs some improvements.