Improved food sources in Artificial Bee Colony

Foraging behavior has inspired different algorithms to solve real-parameter optimization problems. One of the most popular algorithms within this class is the Artificial Bee Colony (ABC). In the present study the food source is initialized by comparing the food source with worst fitness and the evaluated mean of randomly generated food sources (population). Further the scout bee operator is modified to increase searching capabilities of the algorithm to sample solutions within the range of search defined by the current population. The proposed variant is called IFS-ABC and is tested on six unconstrained benchmark function. Further to test the efficiency of the proposed variant we implemented it on five constrained engineering optimization problems.