Important terms and concepts

To better understand the structure and working of GA, it is required to first discuss a few important terms and concepts.

Chromosome

A chromosome is a candidate/potential solution to the problem being solved. The solution may not necessarily be the optimal solution. A chromosome may be considered as an array of binary/integer variables where each variable may be considered as a gene and the value of a variable is called an allele. For better understanding see the following diagram.

Population

A set of solutions participating in the process of optimization is called population. Unlike trajectory-based algorithms, GA is a population-based algorithm in which several solutions interact with each other to find the global optimum.

Objective function

It is also called a fitness function. Whenever an optimization problem is solved it is first formulated as a mathematical function that evaluates the quality/fitness of the candidate solution. We usually pass a solution to this function and this function returns the fitness of that solution. Once the fittest/optimal solution is found the process is stopped.