Investigation of starting conditions in generative processes for the design of engineering structures

Edgar Buchanan Berumen; Rahul Dubey; Simon John Hickinbotham; Imelda Friel; Andrew Colligan; Mark Price; Andy Tyrrell

Investigation of starting conditions in generative processes for the design of engineering structures

Edgar Buchanan Berumen, Rahul Dubey, Simon John Hickinbotham, Imelda Friel, Andrew Colligan, Mark Price, Andy Tyrrell

Electronic Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Engineering design has traditionally involved human engineers manually creating and iterating on designs based on their expertise and knowledge. Bio-inspired Evolutionary Development (EvoDevo) generative algorithms aim to explore a much larger design space that may not have ever been considered by human engineers. However, for complex systems, the designer is often required to start the EvoDevo process with an initial design solution (seed) which the development process will optimize. The question is will a relatively good starting seed always yield a good set of design solutions. This paper considers this question and suggests that sub-optimal seeds can provide, up to certain limits, better design solutions than relatively more optimal seeds. In addition, this paper highlights the importance of designing the appropriate seed for the appropriate problem. In this paper, the problem analysed is the structural performance of a Warren Truss (bridge-like structure) under a single load. The main conclusion of this paper is that up to a limit sub-optimal seeds provide in general better sets of solutions than more optimal seeds. After this limit, the performance of sub-optimal seed starts to degrade as parts of the phenotype landscape become inaccessible.

Original language	English
Title of host publication	2023 IEEE Symposium Series on Computational Intelligence
Subtitle of host publication	International Conference on Evolvable Systems (ICES)
Publication status	Accepted/In press - 15 Sept 2023
Event	2023 IEEE Symposium Series on Computational Intelligence: International Conference on Evolvable Systems (ICES) - Mexico City, Mexico City, Mexico Duration: 5 Dec 2023 → 8 Dec 2023 https://attend.ieee.org/ssci-2023/

Conference

Conference	2023 IEEE Symposium Series on Computational Intelligence
Abbreviated title	SSCI
Country/Territory	Mexico
City	Mexico City
Period	5/12/23 → 8/12/23
Internet address	https://attend.ieee.org/ssci-2023/

Keywords

evodevo
generative design
structural engineering
genetic algorithm
Neural network

Access to Document

ebb_ices_2023
This is an author-produced version of the published paper. Uploaded in accordance with the University’s Research Publications and Open Access policy.
Accepted author manuscript, 2.06 MBLicence: CC BY

Re-Imagining Engineering Design (REID): Growing Radical Cyber-Physical-Socio Phenotypes
Tyrrell, A.
EPSRC
1/05/21 → 30/04/26
Project: Research project (funded) › Research

Cite this

@inproceedings{05655bc4d95f492aba3b76b394e4ffbf,

title = "Investigation of starting conditions in generative processes for the design of engineering structures",

abstract = "Engineering design has traditionally involved human engineers manually creating and iterating on designs based on their expertise and knowledge. Bio-inspired Evolutionary Development (EvoDevo) generative algorithms aim to explore a much larger design space that may not have ever been considered by human engineers. However, for complex systems, the designer is often required to start the EvoDevo process with an initial design solution (seed) which the development process will optimize. The question is will a relatively good starting seed always yield a good set of design solutions. This paper considers this question and suggests that sub-optimal seeds can provide, up to certain limits, better design solutions than relatively more optimal seeds. In addition, this paper highlights the importance of designing the appropriate seed for the appropriate problem. In this paper, the problem analysed is the structural performance of a Warren Truss (bridge-like structure) under a single load. The main conclusion of this paper is that up to a limit sub-optimal seeds provide in general better sets of solutions than more optimal seeds. After this limit, the performance of sub-optimal seed starts to degrade as parts of the phenotype landscape become inaccessible. ",

keywords = "evodevo, generative design, structural engineering, genetic algorithm, Neural network",

author = "{Buchanan Berumen}, Edgar and Rahul Dubey and Hickinbotham, {Simon John} and Imelda Friel and Andrew Colligan and Mark Price and Andy Tyrrell",

year = "2023",

month = sep,

day = "15",

language = "English",

booktitle = "2023 IEEE Symposium Series on Computational Intelligence",

note = "<br/>2023 IEEE Symposium Series on Computational Intelligence : International Conference on Evolvable Systems (ICES), SSCI ; Conference date: 05-12-2023 Through 08-12-2023",

url = "https://attend.ieee.org/ssci-2023/",

}

Buchanan Berumen, E , Dubey, R , Hickinbotham, SJ, Friel, I, Colligan, A, Price, M & Tyrrell, A 2023, Investigation of starting conditions in generative processes for the design of engineering structures. in 2023 IEEE Symposium Series on Computational Intelligence: International Conference on Evolvable Systems (ICES).
2023 IEEE Symposium Series on Computational Intelligence, Mexico City, Mexico, 5/12/23.

Investigation of starting conditions in generative processes for the design of engineering structures. / Buchanan Berumen, Edgar ; Dubey, Rahul ; Hickinbotham, Simon John et al.
2023 IEEE Symposium Series on Computational Intelligence: International Conference on Evolvable Systems (ICES). 2023.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Investigation of starting conditions in generative processes for the design of engineering structures

AU - Buchanan Berumen, Edgar

AU - Dubey, Rahul

AU - Hickinbotham, Simon John

AU - Friel, Imelda

AU - Colligan, Andrew

AU - Price, Mark

AU - Tyrrell, Andy

PY - 2023/9/15

Y1 - 2023/9/15

N2 - Engineering design has traditionally involved human engineers manually creating and iterating on designs based on their expertise and knowledge. Bio-inspired Evolutionary Development (EvoDevo) generative algorithms aim to explore a much larger design space that may not have ever been considered by human engineers. However, for complex systems, the designer is often required to start the EvoDevo process with an initial design solution (seed) which the development process will optimize. The question is will a relatively good starting seed always yield a good set of design solutions. This paper considers this question and suggests that sub-optimal seeds can provide, up to certain limits, better design solutions than relatively more optimal seeds. In addition, this paper highlights the importance of designing the appropriate seed for the appropriate problem. In this paper, the problem analysed is the structural performance of a Warren Truss (bridge-like structure) under a single load. The main conclusion of this paper is that up to a limit sub-optimal seeds provide in general better sets of solutions than more optimal seeds. After this limit, the performance of sub-optimal seed starts to degrade as parts of the phenotype landscape become inaccessible.

AB - Engineering design has traditionally involved human engineers manually creating and iterating on designs based on their expertise and knowledge. Bio-inspired Evolutionary Development (EvoDevo) generative algorithms aim to explore a much larger design space that may not have ever been considered by human engineers. However, for complex systems, the designer is often required to start the EvoDevo process with an initial design solution (seed) which the development process will optimize. The question is will a relatively good starting seed always yield a good set of design solutions. This paper considers this question and suggests that sub-optimal seeds can provide, up to certain limits, better design solutions than relatively more optimal seeds. In addition, this paper highlights the importance of designing the appropriate seed for the appropriate problem. In this paper, the problem analysed is the structural performance of a Warren Truss (bridge-like structure) under a single load. The main conclusion of this paper is that up to a limit sub-optimal seeds provide in general better sets of solutions than more optimal seeds. After this limit, the performance of sub-optimal seed starts to degrade as parts of the phenotype landscape become inaccessible.

KW - evodevo

KW - generative design

KW - structural engineering

KW - genetic algorithm

KW - Neural network

M3 - Conference contribution

BT - 2023 IEEE Symposium Series on Computational Intelligence

T2 - <br/>2023 IEEE Symposium Series on Computational Intelligence

Y2 - 5 December 2023 through 8 December 2023

ER -

Investigation of starting conditions in generative processes for the design of engineering structures

Abstract

Conference

Keywords

Access to Document

Projects

Re-Imagining Engineering Design (REID): Growing Radical Cyber-Physical-Socio Phenotypes

Cite this