| dc.description.abstract | ‘Design’ is a means of changing existing situations into preferred ones. The engineering design process has been broadly classified into four stages: task clarification, conceptual design, embodiment design and detail design. ‘Design research’ aims to develop knowledge in various forms, making the design process more efficient and effective. This can be achieved by (a) developing an understanding of the design phenomena in the form of theories and models and validating them; (b) developing and validating support in the form of strategies, techniques, frameworks, guidelines, methods, software tools, etc. founded on the current understanding, to improve the design process.
Conceptual design is an early stage in the design process, which involves generating solution concepts to satisfy the functional requirements of a design problem. The conceptual design stage is considered important because a high percentage of the product cost is committed at this stage. Decisions made during this stage will strongly affect all the subsequent stages of the design process.
A complex mechanical device is often characterised by multiple states whereby it can achieve different functions at different operating states by changing its topological structure and the interaction among its elements. In general, mechanical devices’ input and output elements have a single input-output relationship. However, for specific applications, many mechanical devices include multiple input-output relations, each of which is enabled by a set of behaviours in the device’s relevant operating state. In literature, these devices are termed multi-state mechanical devices (MSMD), mechanisms with variable topology (MVT), multi-modal devices and metamorphic mechanisms. This thesis particularly focuses on the conceptual design of such complex mechanical devices called MSMDs. The broad aim of this thesis work is to support the representation, synthesis, and analysis of MSMDs at the conceptual design phase.
Function representation, also known as function modelling, is important in the design process as it is used to analyse, share, and communicate design concepts across various disciplines and assist designers in concept generation, evaluation, and selection. Function models primarily represent the purpose served by a (desired) system, which means, among others, the transformation of material, energy, and (or) information flows. The first specific objective of this thesis work is to develop a function modelling scheme that provides a comprehensive description of an MSMD design concept. Here, the comprehensiveness of the function modelling scheme is determined by its ability to encapsulate the following essential aspects of an MSMD: (a) structure and changes in structure across different operating states, (b) structural and exogenous behaviours, and (c) intended behaviour and purpose of the MSMD. This has been achieved through adapting and integrating various features of existing function modelling schemes where SAPPhIRE—a model of causality—is used as the basis. The proposed approach is demonstrated through example case studies.
The second specific objective of this thesis is to support the synthesis of MSMD design concepts. Developing radically new and significantly better solutions requires exploring a large solution space. Most of the existing literature on conceptual design synthesis of mechanical devices (a) is on synthesising device concepts for a single relation of ‘converting one input set to another output set’—this is called single-state design synthesis (SSDS), and (b) primarily employs ‘composition of building blocks’ as the approach for the synthesis of concepts. On the other hand, multiple-state design synthesis (MSDS) refers to synthesising device concepts for more than one relation between an input set and an output set. Since not much literature is available on studies of MSDS, it is essential to understand how and how well designers currently carry out MSDS. This knowledge can be used for developing prescriptive support to improve MSDS. Therefore, to fulfil the second specific objective, first, a scheme for representing an MSMD design task is described. Then, empirical studies are conducted with 12 designers, and an MSMD design task is given to them. All the synthesis processes are video recorded and analysed to develop a shared understanding of the process and outcomes. Based on this empirical understanding, a prescriptive model of MSDS is developed. A web-based computational tool – called ‘CoDe SyMM’– is implemented to support the effective and efficient use of the prescriptive model. An initial evaluation of the tool’s usefulness is carried out with the help of design experiments involving external designers. The results obtained from the evaluation study are reported.
The third specific objective of this thesis is to support the analysis of MSMD design concept space. ‘Variety’ is one of the parameters by which one can analyse the creative quality of a concept space explored by the designers. It is useful to assess variety at the conceptual design stage because, at this stage, designers have the freedom to explore different solution principles so as to satisfy a design problem with substantially novel concepts. To achieve this objective, this work elaborates on the existing variety metrics and discusses their limitations. Then, the use of multi-instance SAPPhIRE models to measure the real-valued distance between two design concepts is addressed. A new distance-based variety metric is proposed, and a prescriptive framework for supporting the evaluation process has been developed. A GUI-based computational tool – called ‘VariAnT’– is also developed to automate the variety assessment process. | en_US |
