Optimization with the Nature-Inspired Intelligent Water Drops Algorithm

Abstract

Scientists are beginning to realize more and more that nature is a great source for inspiration in order to develop intelligent systems and algorithms. In the field of Computational Intelligence, especially Evolutionary Computation and Swarm-based systems, the degree of imitation from nature is surprisingly high and we are at the edge of developing and proposing new algorithms and/or systems, which partially or fully follow nature and the actions and reactions that happen in a specific natural system or species. Among the most recent nature-inspired swarm-based optimization algorithms is the Intelligent Water Drops (IWD) algorithm. IWD algorithms imitate some of the processes that happen in nature between the water drops of a river and the soil of the river bed. The IWD algorithm was first introduced in (Shah-Hosseini, 2007) in which the IWDs are used to solve the Travelling Salesman Problem (TSP). The IWD algorithm has also been successfully applied to the Multidimensional Knapsack Problem (MKP) (Shah-Hosseini, 2008a), n-queen puzzle (Shah-Hosseini, 2008b), and Robot Path Planning (Duan et al., 2008). Here, the IWD algorithm and its versions are specified for the TSP, the n-queen puzzle, the MKP, and for the first time, the AMT (Automatic Multilevel Thresholding). Some theoretical findings have also been reviewed for the IWD algorithm. Next section reviews briefly the related works. Section 3 examines natural water drops. Section 4 states about Intelligent Water Drops (IWDs). Section 5 specifies the Intelligent Water Drops (IWD) algorithm. Next section, reviews the convergence properties of the IWD algorithm. Section 7 includes experiments with the IWD and its versions for the four mentioned problems. Final section includes the concluding remarks.