A user-centered approach to evolutionary algorithms and their use in industry

The key contribution of this article to the domain of engineering optimisation is establishing best practices with respect to the development of evolutionary algorithms in the context of design engineering. Despite their various uses, the uptake of evolutionary algorithms in industry remains limited...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Cogent engineering Ročník 8; číslo 1
Hlavní autori: Vincalek, Jakub, Walton, Sean, Evans, Ben
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Abingdon Cogent 01.01.2021
Taylor & Francis Ltd
Taylor & Francis Group
Predmet:
Best practice
Design engineering
design optimisation
Evolutionary algorithms
Genetic algorithms
human in the loop
optimisation algorithms
Optimization
Questions
reflexive thematic analysis
ISSN:2331-1916, 2331-1916
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:The key contribution of this article to the domain of engineering optimisation is establishing best practices with respect to the development of evolutionary algorithms in the context of design engineering. Despite their various uses, the uptake of evolutionary algorithms in industry remains limited. In order to understand why uptake is low a survey of engineers was undertaken, the results of which are presented here. A total of 23 participants (N = 23) took part in the 3-section mixed methods survey. Reflexive thematic analysis was conducted on the open-ended questions. A common thread throughout participants responses is that there is a question of trust towards evolutionary algorithms within industry. Perhaps surprising is that the key to gaining this trust is not producing good results, but creating algorithms which explain the process they take in reaching a result. Based on this, recommendations have been made to increase their use in industry.
Bibliografia:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2331-1916
2331-1916
DOI:10.1080/23311916.2021.1995110