Shape optimisation using evolutionary techniques in product design

doi:10.1016/j.cie.2007.06.010

Computers & Industrial Engineering

Volume 53, Issue 2, September 2007, Pages 200-205

https://doi.org/10.1016/j.cie.2007.06.010 Get rights and content

Abstract

Shape or surface optimisation in product design is a very essential and time-consuming process, especially at the conceptual design stage. In this paper, we introduce a research project aiming to develop an evolutionary design system capable of evolving product shape designs that are easy to manufacture and satisfy the given geometric constraints. One of the issues in applying evolutionary techniques to conceptual design is how to represent designs in a way in which genetic algorithms can be used to support the process of generating and optimising innovative and imaginative geometric components and parts. This paper examines two stages of using genetic algorithms in product shape design-the representation of shapes or phenotype and how to encode designs in a manner analogous to genes in nature, which can be manipulated by genetic algorithms. The early research result and directions for future work are also presented in this paper.

Introduction

The competition in the current global markets is commanding rapid improvements in product performance and quality with the reductions in costs and development time-scales. This forces the product-oriented manufacturing enterprises to seek more advance technologies to gain profits by reducing the design development time. As a result, design automation is required to improve the functionality of the designs and reduce the development time and thus reduce the cost, especially when designing complex surfaces are required. Designing the shapes of products is the primary activity of the design process and the CAD/CAM systems used today for the designs provide limited facilities for automatic optimisation of surfaces. They supported an iterative surface optimisation, which is very time consuming and requires an enormous amount of skilled engineering labor.

On the other hand, evolutionary computation techniques have been successfully applied to engineering design optimisation, constraint satisfaction, symbolic equation solving and manufacturing process planning. These adaptive and generative techniques provide more creative and intelligent support to designers than other design support systems that have to rely on geometric representations and explicit deductive inference mechanisms. However, the application of evolutionary computation techniques, such as genetic algorithms (GAs) (Davis, 1991, Goldberg, 1989, Holland, 1975), in product design has been limited (Gen and Cheng, 1997, Hawkes and Abinett, 1984). The difficulty arises from the fact that product design requires a wide variety of knowledge and information at the early creative state of the design process that is difficult to formulate in a computer system (Karl, 1997, Medland, 1986, Roy et al., 1999). One of the important issues in applying evolutionary techniques to conceptual design is how to represent 3D structures in a way in which genetic algorithms can be used to support the process of generating and optimising innovative and imaginative geometric components and parts (Frazer, 1995, Graham, 1995). So this paper examines two stages of using genetic algorithms in design-how to encode designs as chromosome and how to translate certain manufacturing constraints into the evaluation functions which can be manipulated by GAs.

In this paper, an improved simple GA is used to demonstrate the concept that surface optimisation can be effectively supported within an existing CAD system. Work in this paper focuses on the most difficult aspect of design representation, the geometry or shape of design, based on the ergonomic and aesthetic consideration, and factors such as materials and costs are not considered at this stage. Based on the discussion and analysis of the existing shape representation methods, how to choose optimal characteristic parameters for shape design and how to define the combination of the geometry or shapes from simple to complex are presented in this paper. The methods presented in this paper have been tested in the early case studies on the simple ruled and non-ruled surface optimisation problems. The analysis of the initial research results and the discussion of the future work are also presented in this paper.

Section snippets

Shape representation and optimisation

Existing solid modeling systems provide facilities for creating, modifying and inspecting models of 3D solid objects, but there are a large number of different possible methods for representing such models in a computer. However, representation schemes may be divided into six general classes (Rooney & Steadman, 1997): Pure primitive instancing; Generalised sweeps; Spatial occupancy enumeration; Cellular decomposition; Constructive solid geometry (CSG) and Boundary representation. Among them,

Application development

The proposed project in this paper aims to investigate how genetic algorithms can be successfully applied in product design and to develop an evolutionary design system capable of evolving solid object designs which are easy to manufacture and can satisfy all the given constraints. When applying a genetic algorithm to any applications, four main elements must be considered (Bentley, 1999): Firstly, the phenotype must be specified, in other words the allowable solutions to the problem must be

Conclusions and future work

The paper is about the early stage work of the research project – the application of evolutionary computation techniques in design for manufacturability. This research is concerned with the development of a generic computer framework in which genetic algorithms are used to support the process of automatically generating the geometric forms of products using primitive shells that are easy to manufacture (Sun, Frazer, & Tang, 1999). The surface optimisation within a CAD/CAM system is founded to

Acknowledgements

This research project is supported by a UGC Ph.D. project grant from the Hong Kong Polytechnic University. The authors would like to acknowledge the support from academic and research staff in the Design Technology Research Centre of School of Design.

References (19)

Bentley, J. P. ( 1996). Generic evolutionary design of solid objects using a genetic algorithm. Ph.D....
L. Davis
Handbook of genetic algorithms
(1991)
Frazer, J. (1995). An evolutionary architecture, Architecture Association,...
Frazer, J., Tang, M. X., & Sun, J. (1999). Towards a generative system for intelligent design support. In Proceedings...
M. Gen et al.
Genetic algorithms and engineering design
(1997)
D.E. Goldberg
Genetic algorithms in search, optimization, and machine learning
(1989)
Graham, P. (1995). The application of evolutionary and rule-based techniques in computer aided design. Ph.D....
B. Hawkes et al.
The engineering design process
(1984)

There are more references available in the full text version of this article.

Cited by (23)

A 3D shape generative method for aesthetic product design
2020, Design Studies
Citation Excerpt :
Renner and Ekárt (2003) offer a review of their application to computer-aided design. Since some early works such as Bentley (1998); Gero (1996); Hybs and Gero (1992), many studies have tested the use of evolutionary algorithms in product design: Lamp holders (Liu, Tang, & Frazer, 2004); shape optimisation for mobile phones (Sun, Frazer, & Mingxi, 2007); wine glasses profiles (Su & Zhang, 2010). Shieh, Li, and Yang (2018) use a combination of Kansei and evolutionary algorithms for vase designs.
This work describes a generative method for the exploration of product shapes in the conceptual design stage. The method is based on three concepts: the notion of grammars to capture product appearance, the implementation of sketching transformation rules to produce design variations and the use of a parametric modeller to build shapes. We represent product solutions as 3D sketches using combinations of basic shapes arranged in simple and schematic product structures. This procedure allows creating many varied configurations with a minimal number of shapes, and facilitates the adaptation of the generative model to different products. The performance of the method is demonstrated through several examples from the literature.
Generative design in the development of a robotic manipulator
2018, Procedia CIRP
The emergence of cyber physical production systems has brought with it an increased utilization of robotics in collaborative manufacturing environments. An approach to meet this demand is to democratize robotics by making cheaper more customizable robots that can be implemented by small and medium enterprises. To tackle this problem this research looks at using rapid prototyping techniques for the development of customizable robotic manipulators which can be implemented in cyber physical production systems. This research therefore contributes an approach for designing connected and rapid prototyped robotic manipulators. This approach considers both the software and hardware development required for implementing a robotic manipulator. Furthermore generative design, an evolutionary and artificial intelligence based approach, is used to design the link modules between the robot joints. This component has been identified as the ideal to be designed with this approach as it benefits most of the generative design approach coupled with rapid prototyping. This paper also explores a robotic manipulator control structure based on Ethernet control technology for implementation within cyber physical production systems.
G-BIM framework and development process for integrated AEC design automation
2014, Procedia Engineering
The integration and automation of the whole design and implementation process has become a pivotal factor in Architecture, Engineering, and Construction (AEC) projects, especially regarding recent technological developments and emergent drivers in the field. Extant literature has highlighted a series of recurrent problems in process integration, especially at the conceptual design stage. This study presents the adoption of Generative Building Information Modelling (G-BIM) workspaces as an emerging technology. This has the potential to leverage conceptual design innovation in AEC projects. It builds upon the findings of an initial survey, and proffers a framework for using generative BIM workspaces at the conceptual design stage. This framework highlights the links and dependencies between generative/parametric tools and BIM applications to expedite information transition using generative tools primarily based on neutral BIM standards. Limitations of tools and approaches for providing accurate project information models are also captured in this framework. This paper demonstrates an overview of the G-BIM framework and the developed conceptual tool. Moreover, it reports on the challenges and opportunities associated with existing software applications. Findings reveal that the application of Generative Design (GD) can significantly enhance the design experience by assisting designers in the iterative generation of design alternatives and parameterisation processes. This framework purposefully integrates BIM with GD to enhance the design process at the conceptual design stage. This forms the rubrics for a working prototype which actively engages GD methods into a single dynamic BIM environment – the results of which will be presented in later works.
Structural optimization based on CADCAE integration and metamodeling techniques
2010, CAD Computer Aided Design
Citation Excerpt :
Another disadvantage is that CAD systems offer restricted facilities for structural optimization. Moreover, the potential of CAE tools has not been exploited thoroughly in design optimization since these tools are often confined to single simulation [3] while an optimization process always requires a loop of design-evaluate-redesign. To perform structural optimization in particular and with engineering optimization in general, some engineering designers choose the traditional solution to adjust design parameters manually [4,5].
Traditional structural optimization, which identifies the best combination of geometrical parameters to improve the product’s performance and to save the material, is often carried out manually. This paper presents a framework that performs the integration between commercial CAD–CAE software by using common scripting, programming languages and Application Programming Interface. The loop of design-analysis-redesign in optimization process was done automatically and seamlessly without interaction with designer. Along with CAD–CAE computer-aided tools, metamodeling techniques including response surface methodology and radial basis function were applied to structural optimization according to the number of design variables. This approach reduces the time for solving computation-intensive design optimization problems and the designers are free from monotonous repetitive tasks. Three case studies were carried out in order to verify the feasibility and general-purpose characteristics of the proposed method for the structural optimization process of mechanical components. The results show that the proposed method facilitates the structural optimization process and reduces the computing cost compared to other approaches.
The perceived value of human-AI collaboration in early shape exploration: An exploratory assessment
2022, PLoS ONE
As a vital element of early shape exploration, divergence can be time-consuming and challenging, with iterative cycles where idea fixation and creative blocks must be overcome for fuzzy ideas to be fully expanded and understood. Despite interesting tools that have been developed for this purpose, some important challenges remain, as it appears that many designers still prefer simple freehand sketching and tend to defer the use of computational tools to later stages. This work presents an exploratory assessment of the perceived value of a new tool, Shapi, developed to assist early shape exploration by addressing some of the pitfalls reported in the literature. Shapi is envisioned as an autonomous assistant that provides local and global shape variations in the form of rough sketches based on an initial human sketch and interactive cycles. These shape variations are What-If scenarios and cognitive facilitators that may spark new ideas or enable a deeper understanding of the shape and the identification of interesting patterns. Shapi’s capabilities are explored in a diverse set of case studies with different purposes: nine implementations in industrial design, three in graphic design, and five with open-ended artistic purposes. These implementations are then used in a survey about initial perceived value in which the majority gave high ratings in terms of exploration (75.5% ≥ 4 out of 5), interpretation (83.7% ≥ 4), adaptation (77.6% ≥ 4), value (73.5% ≥ 4), creativity (69.4% ≥ 4), and general interest in the tool (79.6% ≥ 4). This work brings insight into promising functionalities, opportunities, and risks in the intersection between artificial intelligence, design, and art.
Beyond Skin Deep: Generative Co-Design for Aesthetic Prosthetics
2023, Conference on Human Factors in Computing Systems - Proceedings

View all citing articles on Scopus

View full text

Shape optimisation using evolutionary techniques in product design

Abstract

Introduction

Section snippets

Shape representation and optimisation

Application development

Conclusions and future work

Acknowledgements

Handbook of genetic algorithms

Genetic algorithms and engineering design

Genetic algorithms in search, optimization, and machine learning

The engineering design process