ABSTRACT
This chapter develops a constrained multi-objective optimization method in the form of a robust, practical, problem-independent algorithm, and investigates the effect of multi-objective optimization on structural design. It presents the properties and characteristics of multi-objective optimization and solutions for various design objectives and constraints. The chapter proposes two new genetic algorithm (GA) operators namely: the multi-objective fitness function and the niche method. Using GAs to optimize a multi-objective optimization (MOP) problem is totally different from a single-objective optimization one. The chapter outlines the optimizer and decision-maker with the Pareto GA search algorithm, fuzzy penalty function, and game theory decision-making. A multi-objective optimization problem can be cast as a cooperative game problem which assumes that each player is associated with an objective. An important property of a multi-objective optimization problem is that it allows the designer to participate in the design selection process even after formulation of the mathematical optimization model.
