Developing Four Metaheuristic Algorithms for Multiple-Objective Management of Groundwater

Abstract

Groundwater is one of the important sources of freshwater and accordingly, there is a need for optimizing its usage. In this paper, four multi-objective metaheuristic algorithms with new evolution strategy are introduced and compared for the optimal management of groundwater namely: Multi-objective genetic algorithms (MOGA), multi-objective memetic algorithms (MOMA), multi-objective particle swarm optimization (MOPSO), and multi-objective shuffled frog leaping algorithm (MOSFLA). The suggested evolution process is based on determining a unique solution of the Pareto solutions called the Pareto-compromise (PC) solution. The advantages of the current development stem from: 1) The new multiple objectives evolution strategy is inspired from the single objective optimization, where fitness calculations depend on tracking the PC solution only through the search history; 2) a comparison among the performance of the four algorithms is introduced. The development of each algorithm is briefly presented. A comparison study is carried out among the formulation and the results of the four algorithms. The developed four algorithms are tested on two multiple-objective optimization benchmark problems. The four algorithms are then used to optimize two-objective groundwater management problem. The results prove the ability of the developed algorithms to accurately find the Pareto-optimal solutions and thus the potential application on real-life groundwater management problems.