New methods and tools for product design
Product design is a hybrid socio-technological process by which engineering specifications are formulated and various product models are created and analysed to ensure that a product will possess the desired functions, behaviour, properties, quality, and cost. From a methodological point of view, the process of designing products is often interpreted as a kind of combination of industrial design activities and engineering design activities. According to its classical interpretation, industrial design is essentially applied art whereby the aesthetic values and usability features of products are defined. Industrial designers deal with aspects such as overall form of the artefact, structure of shape features, colours, textures, sounds, ergonomics, user interfaces, comfort of use, appeal of artefacts, and human impressions. Engineering design focuses on the artefact as a system, and interprets the technical requirements, determines the functionality, defines the working principles, structure, materialization, and behaviour of artefacts. Our understanding is that product design has actually a broader scope than these two together, since it extends to market placement, product innovation, product technologies, service development, sustainability, manufacturing, maintenance, recycling, and other concerns in addition to those that are in the scope of either industrial design or engineering design. The aforementioned duality and the complexity coming from the large number of aspects to be taken into consideration make computer support of product design a challenging task.
What we try to show in this special issue of Journal of Engineering Design is that design researchers have managed to develop novel computer-oriented methods and tools that are not only significant from an academic point of view, but also valuable solutions for the industrial practice. Of course, the presented new methods and tools are just examples of the best solutions. Our endeavour was to select papers that report on matured results, whose applicability has been demonstrated and tested. The papers were initially presented at the Fifth International Symposium on Tools and Methods of Competitive Engineering (TMCE 2004). The TMCE 2004 Symposium was jointly organized by the Swiss Federal Institute of Technology, Lausanne, and the Delft University of Technology, The Netherlands, and was held in Lausanne, Switzerland, 13–17 April 2004. Thanks to the collaboration of the authors and the reviewers, each paper chosen for this special edition has been extended or reworked, then reviewed again according to the expectations of this journal, and finally revised once more based on the comments of the reviewers and the guest editors. The papers have been thematically arranged in this special issue. The first group of papers is dealing with issues typical for industrial design, in particular, for form giving and shape design. The second group of papers is dealing with structural and system issues of engineering design.
The first paper in the industrial design oriented group, titled Fully free form deformation features for aesthetic shape design, written by J.-P. Pernot, B. Falcidieno, F. Gianninit and J.-C. Léon, contributes to the computer support of the industrial design part of product development. It focuses on free form shape design and proposes deformation features as the means of defining and modifying the shape of objects. The authors propose to generate a bar structure on the control points of the initial surface and take each bar as a spring. They define the intended deformation of the surface by applying external forces and constraints on the bar network. They use target and limit lines to control the region of deformation. Actually this region is considered a free form deformation feature. The authors show that the properties of the surface can be controlled alongside the target and limit lines by certain control parameters. They also propose a classification of the free form deformation features and a method for handling compound shape features.
The second paper, titled Instantiation of shapes for products by shape formation rules based on the vague discrete interval model, is co-authored by Z. Rusák and I. Horváth. This presents the results of a research and development work in the field of computer support of conceptual shape design. The proposed method starts out from an interval shape and allows the designer to derive specific instances by a composition of shape formation rules. It utilizes the opportunities offered by vague discrete interval modelling (VDIM), which represents a family or a cluster of shapes in one single model. The shape formation rules are converted to a bi-cubic bi-parametric effect function whose coefficients define the requested changes in a specified region of shape. Operators for simple, compound and constrained shape instantiation have been defined. The first results show that this approach gives larger freedom and more opportunities for designers to express their ideas than the conventional shape modelling techniques.
In the third paper, titled A NURBS finite element method for product shape design, K. Inoue, Y. Kikuchi and T. Masuyama integrate NURBS-based representation with a physically based generation of shapes. Their method can be effectively applied to a class of shape design problems. They introduce a finite shell element that takes its shape from the NURBS representation of the product shape. The NURBS surface is divided into bounded regions by orthogonal isoparametric curves. Stiffness and mass matrices are formulated in the conventional way, and used in the deformation calculations. The element shape is explicitly represented in the formula for the calculation of nodal positions and the shape function is used to interpolate the nodal displacements. The presented method can be applied in behavioural (deformation, vibration, etc.) investigation of free-formed parts of products. However, it does not support designing multiple-connected surfaces or surfaces with sharp singularities.
The paper titled Effectiveness and efficiency of three-dimensional shape retrieval, by K. Lou, N. Iyer, S. Jayanti, Y. Kalyanaraman, K. Ramani and S. Prabhakar, addresses the issue of reusing past engineering designs. The stored designs, that is, 3D shapes, are retrieved based on feature vectors. The proposed system has three main components, the client interface for submitting queries and presenting the search results, the application server that extracts the feature vectors based on voxelization and employs intelligent agents to select the relevant shapes based on indexing, and the database server that stores the shapes together with the multi-dimensional indices. The degree of similarity is computed, which is not necessarily the same that the user perceives. The authors completed several experiences to explore the nature of differences and to optimize the efficiency of the system. They found that multi-step refinement and multi-dimensional indexing together improve search effectiveness.
The last paper in the industrial design group, Virtual humans and prototypes to evaluate ergonomics and safety, by G. Colombo and U. Cugini, presents an approach to evaluation of ergonomics and safety of products. They use virtual human models to interact with the product model and to analyse human sensing and comfort. The virtual human model is composed of a number of rigid links and joints in between, and is implemented as a human-controlled avatar, equipped with visual senses. They can replicate the motion of the human body in the region defined by the motion constraints and use the avatar to simulate actions of humans. The authors discuss two cases, where the comfort of using the product has been enhanced through the application of an avatar. The proposed methods are general enough, that is, they can also be used in case of other products and design problems.
The first paper in the engineering design group, Configurable product design using multiple fuzzy models, submitted by E. R. Deciu, E. Ostrosi, M. Ferney and M. Gheorghe, interprets the product configuration task as a fuzzy combinatorial composition. The main idea is that the transition from customer specification to the physical architecture of the product can be supported by a sequence of four fuzzy models. They propose a computer-oriented method for fuzzy coupling of the customer requirements and the functions, the functions and the solution elements, and the solutions and the constraints. This fuzzy functional network model supports the development of configurable product families. The alternatives are sorted by a fuzzy ranking method. The authors applied the method in the design of a family of chairs.
A. P. Hofer and J. I. M. Halman deal with layout design of products in the paper titled The potential of layout platforms for modular complex products and systems. They propose a method for effective sharing of modules and parts between complex systems. The basis is the concept of a system layout platform that standardizes the arrangement of components within a family of products. They investigate three case studies and came to the conclusion that standard system layout platforms reduce the complexity of the product as well as the engineering risk. This approach offers the designers hierarchic layering of product architecture and a range of architectural choices. It allows a market segment oriented mass customization of products.
The last paper in the engineering design group, Statistical robust design of a complex product through a sequential approach, by H. Mizuyama, reports on a method which is based on the investigation of the investments under uncertainty. The goal is to enhance the economic performance of complex products, and the means is experiment-based analysis of robustness and noise source control. Rather than considering the product a simple input-output system, the author models it as a complex system and focuses on the optimization of its quality characteristics. He proposes a five-step procedure for robust design that is supported by computation. His conclusion is that the experiment-based robustness enhancement improves the economic quality of complex products, but cannot substitute the reasonable decisions of designers.
We hope that with this composition of papers we managed to provide good examples of the computer methods and tools that can be introduced in the product design processes in the industry. We also hope that by these papers we could demonstrate the best research practices for fellow researchers. We are grateful to all authors for their contribution to this special issue, and do appreciate their efforts and nice collaboration. We are also indebted to the reviewers who helped us increase the quality of the papers and the value of this special issue.
Dr. Imre Horváth
Dr. Zoltán Rusák
Faculty of Industrial Design Engineering
Delft University of Technology