Date of Award
Master of Science
Multi-objective optimization problem formulations reflect pragmatic modeling of several real-life complex optimization problems. In many of them the considered objectives are competitive with each other; emphasizing only one of them during solution generation and evolution incurs high probability of producing a one-sided solution, which is unacceptable with respect to other objectives. An appropriate solution to the multi-objective optimization problem is to investigate a set of solutions that satisfy all of the competing objectives to an acceptable extent, where no solution in the solution set is dominated by others in terms of objective optimization. In this work, we investigate well known Non-dominated Sorting Genetic Algorithm (NSGA-II), and Strength Pareto Evolutionary Algorithm (SPEA-II), to find Pareto optimal solutions for two real-life problems: Task-based Sailor Assignment Problem (TSAP) and Coverage and Lifetime Optimization Problem in Wireless Sensor Networks (CLOP). Both of these problems are multi-objective problems. TSAP constitutes five multi-directional objectives, whereas CLOP is composed of two competing objectives. To validate the special operators developed, these two test bed problems have been used. Finally, traditional NSGA-II and SPEA-II have been blended with these special operators to generate refined solutions of these multi-objective optimization problems.
dissertation or thesis originally submitted to the local University of Memphis Electronic Theses & dissertation (ETD) Repository.
Chaudhuri, Koyel, "A Multi-objective Evolutionary Algorithm to solve Complex Optimization Problems" (2010). Electronic Theses and Dissertations. 58.