A Levy Flight-based Decomposition Multi-objective Optimization Based on Grey Wolf Optimizer

The goal of an optimization technique is to find the best solution to an optimization problem. In a single-objective problem, the best solution is the optimal value for the objective function, while in a multi-objective problem, the selection of solutions is not a straightforward task because there are several objective functions which are in conflict. There are many diverse applications such as image processing and data mining, which can be formulated as a multi-objective problem. This paper presents a new decomposition-based multi-objective optimization method using the grey wolf optimizer, which transforms the problem into several sub-problems and examines all the sub-problems simultaneously. Our proposed algorithm obtains the Pareto front using a neighborhood relation among the sub-problems. The levy flight distribution has also been used which increases the exploration and exploitation features in the algorithm in order to improve the search ability. The performance of our proposed algorithm is evaluated on UF family of benchmark functions in terms of different metric such as inverted generational distance (IGD), generational distance (GD), hyper-volume (HV), and spacing (SP). The experimental results indicate the superior performance of the proposed method.