A comparison of four types of evolution strategies for beef cattle feed optimization

Tirana Noor Fatyanosa; Wayan Firdaus Mahmudy

A comparison of four types of evolution strategies for beef cattle feed optimization

Tirana Noor Fatyanosa^*
, Wayan Firdaus Mahmudy
, Marjuki

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

5 Citations (Scopus)

Abstract

Beef cattle feed optimization is a multi-objective problem. For different weight of beef cattle, the required nutrition is also different. The feed also requires a balance of nutrients with a low price. This paper presents a comparison of four types of Evolution Strategies (ES) for beef cattle feed optimization. The results of our experiments suggest that the performance and robustness of ES (μ,Λ), ES (μ/ρ,Λ), and ES (μ+Λ) are comparable, while ES (μ/ρ+Λ) performs slightly worse. This fact together promotes ES (μ/ρ,Λ) as the most robust for practical use. The experimental results show that the feed price obtained from ES (μ/ρ,Λ) is 5524.465 with fitness value of 1.809861462.

Original language	English
Pages (from-to)	165-171
Number of pages	7
Journal	Journal of Telecommunication, Electronic and Computer Engineering
Volume	10
Issue number	1-8
Publication status	Published - 2018

Keywords

Beef Cattle Feed Optimization
Evolution Strategies

Cite this

@article{2314b50a4e03429095d2c814fe18bf41,

title = "A comparison of four types of evolution strategies for beef cattle feed optimization",

abstract = "Beef cattle feed optimization is a multi-objective problem. For different weight of beef cattle, the required nutrition is also different. The feed also requires a balance of nutrients with a low price. This paper presents a comparison of four types of Evolution Strategies (ES) for beef cattle feed optimization. The results of our experiments suggest that the performance and robustness of ES (μ,Λ), ES (μ/ρ,Λ), and ES (μ+Λ) are comparable, while ES (μ/ρ+Λ) performs slightly worse. This fact together promotes ES (μ/ρ,Λ) as the most robust for practical use. The experimental results show that the feed price obtained from ES (μ/ρ,Λ) is 5524.465 with fitness value of 1.809861462.",

keywords = "Beef Cattle Feed Optimization, Evolution Strategies",

author = "Fatyanosa, \{Tirana Noor\} and Mahmudy, \{Wayan Firdaus\} and Marjuki",

note = "Publisher Copyright: {\textcopyright} 2018 Universiti Teknikal Malaysia Melaka. All rights rerserved.",

year = "2018",

language = "English",

volume = "10",

pages = "165--171",

journal = "Journal of Telecommunication, Electronic and Computer Engineering",

issn = "2180-1843",

publisher = "Universiti Teknikal Malaysia Melaka",

number = "1-8",

}

TY - JOUR

T1 - A comparison of four types of evolution strategies for beef cattle feed optimization

AU - Fatyanosa, Tirana Noor

AU - Mahmudy, Wayan Firdaus

AU - Marjuki,

PY - 2018

Y1 - 2018

N2 - Beef cattle feed optimization is a multi-objective problem. For different weight of beef cattle, the required nutrition is also different. The feed also requires a balance of nutrients with a low price. This paper presents a comparison of four types of Evolution Strategies (ES) for beef cattle feed optimization. The results of our experiments suggest that the performance and robustness of ES (μ,Λ), ES (μ/ρ,Λ), and ES (μ+Λ) are comparable, while ES (μ/ρ+Λ) performs slightly worse. This fact together promotes ES (μ/ρ,Λ) as the most robust for practical use. The experimental results show that the feed price obtained from ES (μ/ρ,Λ) is 5524.465 with fitness value of 1.809861462.

AB - Beef cattle feed optimization is a multi-objective problem. For different weight of beef cattle, the required nutrition is also different. The feed also requires a balance of nutrients with a low price. This paper presents a comparison of four types of Evolution Strategies (ES) for beef cattle feed optimization. The results of our experiments suggest that the performance and robustness of ES (μ,Λ), ES (μ/ρ,Λ), and ES (μ+Λ) are comparable, while ES (μ/ρ+Λ) performs slightly worse. This fact together promotes ES (μ/ρ,Λ) as the most robust for practical use. The experimental results show that the feed price obtained from ES (μ/ρ,Λ) is 5524.465 with fitness value of 1.809861462.

KW - Beef Cattle Feed Optimization

KW - Evolution Strategies

UR - https://www.scopus.com/pages/publications/85045194859

M3 - Article

AN - SCOPUS:85045194859

SN - 2180-1843

VL - 10

SP - 165

EP - 171

JO - Journal of Telecommunication, Electronic and Computer Engineering

JF - Journal of Telecommunication, Electronic and Computer Engineering

IS - 1-8

ER -

A comparison of four types of evolution strategies for beef cattle feed optimization

Abstract

Keywords

Other files and links

Fingerprint

Cite this