---
title: Evolutionary multi-objective optimization
slug: evolutionary-multi-objective-optimization
qid: 
snapshot: enwiki-001372280614
content_hash: 
license: CC BY-SA 4.0
source: https://en.wikipedia.org/w/index.php?oldid=1372280614
tags: [20181550, 23118434, 23603337]
---

# Evolutionary multi-objective optimization

> [!note] Provenance
> Snapshot `enwiki-001372280614` · content root `…`
> Derived from <https://en.wikipedia.org/w/index.php?oldid=1372280614> · Licensed CC BY-SA 4.0

Evolutionary multi-objective optimization (EMO), also called evolutionary multi-criterion optimization, is the application of evolutionary algorithms to multi-objective optimization problems. Such problems contain two or more objectives that are optimized simultaneously and generally give rise to a set of trade-off solutions rather than a single optimum. Evolutionary methods are particularly suited to generating finite approximations of the Pareto front because a population can contain several candidate solutions representing different trade-offs at the same time. A typical EMO algorithm therefore attempts to improve convergence toward the Pareto front while maintaining diversity among the solutions. Most EMO methods follow an a posteriori approach in which an approximation set is generated before a final solution is selected by a decision maker. Preference information can, however, also be incorporated during the evolutionary search.

Evolutionary multi-objective optimization (EMO), also called evolutionary multi-criterion optimization, is the application of evolutionary algorithms to multi-objective optimization problems. Such problems contain two or more objectives that are optimized simultaneously and generally give rise to a set of trade-off solutions rather than a single optimum. Evolutionary methods are particularly suited to generating finite approximations of the Pareto front because a population can contain several candidate solutions representing different trade-offs at the same time. A typical EMO algorithm therefore attempts to improve convergence toward the Pareto front while maintaining diversity among the solutions. Most EMO methods follow an a posteriori approach in which an approximation set is generated before a final solution is selected by a decision maker. Preference information can, however, also be incorporated during the evolutionary search.
