Evolutionary computing: try, select, improve
Some algorithms search for good solutions by imitating the processes of evolution.
What is it?
Evolutionary algorithms search for solutions the way nature does: they create many variants, test them, keep the better ones and change them slightly. After many rounds, the solutions improve. They are useful when there is no direct route to a solution.
An everyday example
Engineers used this to design a satellite antenna: the computer tried thousands of shapes and kept those that received the signal best.
Why does it matter?
The method often finds unexpected solutions in design, scheduling and optimisation, where there are too many options to try them all.
A common misconception
That this is biological evolution, or that the algorithm always finds the best solution. It finds a good solution according to a measure we chose ourselves.
In more detail
How does it work?
The system creates many candidates, scores them, and gradually builds new options from the better ones.
Is this biological evolution?
No. It is a computational analogy based on variation and selection.
Where is it useful?
In optimising designs, planning and tuning parameters.
Does it always find the best solution?
No. It usually looks for a good-enough solution in a huge space of possibilities.
Remember: Being inspired by nature is no proof that a program works like a living thing.
Reliable sources
Last reviewed 1 November 2026. An explainer, not professional, legal or medical advice. For rules and figures that change, check the official version.