Business process modelling is the process of visualising business processes in order to understand, improve and/or enact these processes in a particular computing environment. Furthermore, it is envisaged that environments that adopt the Service Oriented Architecture (SOA) model of computing are increasingly becoming the de facto environment for executing business processes represented by their visual models. However, it is not possible to use application domain specific web or grid services to enact or execute business processes without translating them into executable programmes. This suggests the need for a generic approach to perform these translations in a structured manner. This research is an attempt to bridge the gap between business process models and service-oriented environments with reference to RAD (Role Activity Diagramming) business process models and grid-aware service-oriented environments.

The decisions made at the conceptual design stage are crucial to the overall success of the product as they affect all the downstream phases of the product life cycle, the user satisfaction of the product and the environment that the product is used and disposed of. The consequences due to these design decisions could therefore be good or problematic. Due to the lack of availability of knowledge and understanding about the complexity of such knowledge spanning these different areas, designers find it difficult to know the implications of their decisions made at the conceptual stage on the product's life cycle, the user of the product and the environment in which the product operates. Reviews of existing methodologies reveal that there is a, need for a holistic view of knowledge in terms of the total context of the design problem under consideration to aid designers in their decision making at the conceptual design stage. This thesis addresses this problem by proposing, implementing and evaluating a computational framework for supporting decision making at the conceptual design stage. The need for considering the implications of design decisions on other life cycle stages of the product and using the whole context of the design problem lead to the characterization and formalization of the Design Context Knowledge into different groups and context knowledge categories. This structuring facilitates the creation of feasible design solutions composed of what is called Product Design Elements (PDEs) i.e. basic elements as a functional means to constitute a conceptual product design solution. The proposed Function to POE mapping model uses the aforesaid design context knowledge structured in different categories for reasoning and eliciting consequences, associated with selecting a particular design solution and determining its implications on the product's subsequent life cycle stages, user of the product and on the product itself. After developing a system architecture model based on the system requirements, the PROCONDES prototype system has been implemented for a sheet metal component design domain. An evaluation of PROCONDES performed by conducting a case study indicates the importance of design context knowledge in proactively supporting effective decision making during function to POE mapping process (i.e. conceptual design stage) by generating timely potential (good and problematic) consequences. However, further work is required to improve the model and its implementation to fully explore the approach and use of PROCONDES for real-time design scenarios.

The research establishes a methodology for developing Web-based CAD/CAM software systems to industrial quality standards in a time and cost e ective manner. The methodology de nes the scope of applicability, outlines major considerations and key principles to follow when developing this kind of software, describes an approach to requirements elicitation, resource allocation and collaboration, establishes strategies for overcoming uncertainty and describes the design concerns for industrial Web-based CAD/CAM systems. The crucial parts of the methodology are a novel project development model facilitating architecture optimisation early in the project to minimise total development e orts, create future-proof solutions and ensure system maintainability; and a novel approach for planning based on time reserve management and task prioritisation, which provides the exibility required for exploratory development while maintaining the main focus on project objectives. The e ectiveness of the Web-based CAD/CAM software development methodology has been examined using two real software development case studies: a Web-based CAD/CAM system for involute spur gear shaper cutters and a Web-based CNC code editor for online modi cation of the pro le for manufacturing gear shaper cutters. The development of case studies using the established methodology resulted in on-time delivery of two industrial browser-based CAD/CAM systems, that produce valid results, embrace all business processes associated with the application area, ensure all functional and non-functional requirements and are used in production now. The developed software products demonstrate robustness, performance, reliability, security and usability comparable with the standards of modern commercial software, utilise advantages of Web-based applications to the highest extent and con rm advantages of Web-based CAD/CAM software compared to similar desktop applications. E ectiveness of the proposed methodology for Web-based CAD/CAM software development was checked through validation, evaluation and analysis of case study results.

The research described in this thesis is concerned with human performance modelling as an aid in the process of manufacturing systems design and re-design. Most manufacturing systems are highly complex constructs and their behaviour is of a dynamic and stochastic nature. They have to be constantly designed and re-designed as organisations are continually being pressured to change their manufacturing facilities, technologies, methods, people and products. All design methods have some form of evaluation where discrete event simulation models are usually used to undertake a comparative analysis of different system designs. Within these discrete event simulation models it is common practice to represent workers as simple resources, often using deterministic performance values. Conversely, the work measurement literature indicates that worker task performance varies between different workers carrying out the same task and moreover for the same worker when repeating a task. The current approach of representing workers within discrete event simulation models ignores the potentially large effect that human performance Variation can have on system performance. This omission affects in particular simulation models of labour intensive manufacturing systems like manual assembly flow lines. It appears that this adds to the inaccuracy of the simulation model output and that consequently the simulation model does not react the behaviour of a real system in an appropriate way. A research programme has been designed to investigate these issues. First, a long term data collection exercise has been conducted to quantify the performance Variation of workers in a typical automotive manual assembly flow line. The data have then been used in form of frequency distributions to represent worker performance Variation at individual Workstations within manual assembly line simulation models. Through designed simulation experiments the impact that this form of worker performance Variation representation has on the accuracy of manual assembly line model behaviour has been investigated. Overall this research has found that adding worker performance Variation models into manual assembly flow line models has an impact on the accuracy of these simulation models. The magnitude of the impact depends very much on the type of Variation to be represented as well as on the system to be modelled. This evidence is an important result to support justification for further research in this area. For a more sophisticated approach of modelling worker performance Computational Organisation Theory using the multi-agent paradigm has been identified as the most suitable way forward.

Computational design is the study of how programmable computers can be integrated into the process of design. It is not simply the use of pre-compiled computer aided design software that aims to replicate the drawing board, but rather the development of computer algorithms as an integral part of the design process. Programmable machines have begun to challenge traditional modes of thinking in architecture and engineering, placing further emphasis on process ahead of the final result. Just as Darwin and Wallace had to think beyond form and inquire into the development of biological organisms to understand evolution, so computational methods enable us to rethink how we approach the design process itself. The subject is broad and multidisciplinary, with influences from design, computer science, mathematics, biology and engineering. This thesis begins similarly wide in its scope, addressing both the technological aspects of computational design and its application on several case study projects in professional practice. By learning through participant observation in combination with secondary research, it is found that design teams can be most effective at the early stage of projects by engaging with the additional complexity this entails. At this concept stage, computational tools such as parametric models are found to have insufficient flexibility for wide design exploration. In response, an approach called Meta-Parametric Design is proposed, inspired by developments in genetic programming (GP). By moving to a higher level of abstraction as computational designers, a Meta-Parametric approach is able to adapt to changing constraints and requirements whilst maintaining an explicit record of process for collaborative working.

This research focussed on the automatic generation of the optimal method of manufacture, and the cost thereof, at an early stage in design. It used geometric and tolerance data, combined with a database of centrally stored manufacturing knowledge. This allowed the construction of potential manufacturing routes, and the evaluation of their cost. In order to find the least costly method of manufacture for a given component, with defined geometry and tolerance, it is necessary to take a holistic view of the manufacturing process. This thesis shows that the typical sequential approach of choosing the manufacturing process, followed by sequencing, will not necessarily find the global optimum. Dynamic process seletion is required to consider both the choice of manufacturing processes and the sequence together to avoid sub-optimisation. Backward and forward propagation from a final manufacturing process, using a random mutation hill climber, was used to construct potential manufacturing process sets. For each process set constructed, the inner optimisation loop was used to find the optimal sequence of manufacturing operations, with the aid of a repair operator to ensure feasibility. Prior to finding the optimal manufacturing sequence, the forming shape for all potential forming processes was required; this research encompasses a methodology to automatically deduce this shape. This methodology demonstrated convergence to the global optimum on a contrived test case, and has demonstrated accurate results on a real commercial test case. Major areas for improvement could be a combination of the developed methodology with feature recognition, to enable maximum usefulness as a fully automated process planning and cost evaluation tool. In summary this research demonstrates that the lack of a holistic view can cause sub-optimal results, and proposes a methodology which provides a holistic perspective. The methodology has been proved to be accurate, generic and expandable.

This thesis documents the rules that were written to define the functionality of the selector tool, and the rules that constrain the pseudo random design programmes. To define the selector tool rules, a comprehensive survey of the current AM technology production systems was conducted. This documented the capabilities of each system in terms of build volume capacity, and minimum achievable geometric feature dimension. As AM materials are often specific to AM production systems, the survey also matched the AM materials to the corresponding AM production systems. The AM production system and material data gathered for this research was also used to define rues for the pseudo random design programme. The research concluded that the varying capabilities of AM systems can be used as a system of criteria for a rule based system, which could automate the process of analysing CAD parts for their suitability to be produced by certain AM systems. The research also finds that AM systems capabilities can also be used to constrain generative design programmes to certain AM systems, allowing the creation of pseudo random designs of object that are specific to certain AM systems. The key findings of the research have shown that gaps in empirical data regarding AM material characteristics prevent a material selection system based on comparative analysis. With current levels of available material knowledge, a more suitable system uses previous examples of AM applications in a case based structure as a metric for material suitability

This thesis reports the research undertaken to aid the system designers for recovery industry with tools and techniques to better understand the recovery business and analyse it under various circumstances. The principle objective of the research is to develop modelling and analysis tools and techniques for the small and medium enterprise operating in the recovery industry taking into consideration their information requirements.

In the computer aided engineering environment, exceptionally smooth but irregular surfaces are often required, such as car bonnets. It is often a lengthy process to design these surfaces to the degree of smoothness and aesthetic beauty that is required by the designer. Smoothing these surfaces is known as fairing and a variety of techniques exist to tackle the problem in different ways. A new method of surface fairing is proposed and demonstrated in this thesis. Many conventional fairing methods use an agreeable curvature plot across the surface as proof of fairness, the method documented here takes the more holistic approach of constructing the improved surface from an agreeable curvature plot.

This study was motivated by the need to understand how 3D CAD systems might impact on the designer and the fashion design process. Current descriptions of the industry indicated that new strategies were necessary to survive globalisation, technological change and the unprecedented demand for fashion products; and emphasised the significance of fashion design education to supply designers with new design skills. The main issue for UK mass-produced fashion appeared to be how to achieve faster design-to-market products within a global industry. Accounts acknowledged the impact of 3D CAD for garment virtual modelling, masscustomisation, garment fit and virtual try-on for retailing, but little has been written about how fashion designers are reacting to, or using new technology.