Optimization is the heart of many natural processes, to name a few: the natural selection phenomenon in the biological evolution theory that is based on the survival-of-the-fittest principle, the social swarming behavior of birds or the foraging strategies of ants. Hence, simulating such natural phenomena into computational mechanics in form of computer algorithms may offer very promising approaches in the computer-aided structural design.

This is a Pharaonic-Grecian work that aims to frame the art of the evolutionary and swarm based structural optimization. That is through five chapters. At the first chapter, the structural optimization problem is properly defined as well to elaborating its basic components. Then, an intensive review of the literature is introduced, through screening 28 of the most recent and most cited scientific articles in the topic during the first two decades of the 21st century. Afterwards, a set of 21 of the most promising and trending state-of-the-art algorithms is introduced through descriptive paragraphs that elaborate how each algorithm works. Following such elaboration of the literature and the state of the art, numerical tests of 6 benchmark structural optimization problems are implemented in order to assess the performance of 4 of these most promising state-of-the-art algorithms. Eventually, a proper estimation of the future trends in Structural Optimization is presented.

This work concluded the following points:

• In the current era of structural optimization, neural networks are broadly utilized instead of the accurate, but complex, FE methods in evaluating and predicting different aspects (e.g., structural time history responses).

• The combination between MOAs and NN is a trending direction in the modern literature that shows high capability in reducing the computational time and efforts.

• Equipping AI features (e.g., Fuzzy Logic) within an optimization technique, is observed through the literature review as a trending approach as well, in Structural Optimization. As that gets the process done faster and gets it more open to handle general families of optimization problems.

• Penalty function is the most used approach for handling the constraints.

• It is trending recently to use a random-search optimization algorithm during the exploration phase to spot just the promising areas within the search space, then to employ a deterministic or derivative-based optimization algorithm in order to exploit that promising areas, in seek of finding the exact optimal solution.