**3.3 Genetic algorithms**

Genetic algorithms are based on natural selection and they create a population of solutions. Based on this population of solutions, a new population is created using some operators: cross-over and mutation [38]. Genetic algorithm has a strong exploratory feature. Currently, it can be used with integer and continuous parameters. However, it does not have any convergence criteria. After a certain number of iterations, the best so far solution is defined as the *converged solution*. This weakness motivates one chapter in this set of chapters.

The chapter "Mixed-discrete nonlinear programming engineering problems" combines genetic algorithms with sequential quadratic programming. The sequential quadratic programming is a well-known gradient-based search algorithm.
