The literature suggests that organisations which have adopted Six Sigma have realised that upon achieving a Five Sigma level the only way to surpass this is to redesign the process(es) by means of Design for Six Sigma (DFSS). For others, the selection of Six Sigma over the DFSS approach is not a definitive question and just a guideline can be provided. A major objective of this research was to extend the selection of the requisite Six Sigma approach beyond the sigma level case and the general guidelines, towards a multi-criteria decision using established techniques. Thus, two research questions were defined: what influences the selection of the requisite Six Sigma approach, i. e. Six Sigma versus DFSS? and, how effective is the use of Multi-Criteria Decision Analysis (MCDA) techniques in the selection of the requisite Six Sigma approach? An action research methodology was applied where one Six Sigma project, one DFSS project and one Six Sigma project applied in a non-manufacturing process were implemented and analysed in collaboration with 3M Corporation, General Domestic Appliances (GDA) and Land Rover. From the action research spiral it was concluded that the sigma level has a positive association with the selection of redesign or improvement efforts within Six Sigma, however the Five Sigma level cannot necessarily dictate the use of one approach over the other. Besides the sigma level the selection of the requisite Six Sigma approach is influenced by multiple and conflicting criteria. In addition, the selection can occur at different stages of the methodologies. To assist decision-makers in organising, synthesising and optimising the criteria affecting this decision, the Stochastic Analytic Hierarchy Process (SAHP) was developed and applied to the problem at hand. The SAHP was developed on the basis of Analytic Hierarchy Process (AHP) and disparate sources of relevant literature. SAHP provides a mechanism for achieving a more effective selection of the requisite Six Sigma approach in the form of considering multiple and conflicting criteria using quantitative and qualitative information under uncertainty. In contrast to the traditional AHP, SAHP incorporates probabilistic distributions to incorporate uncertainty that people have in converging into a Likert scale their judgments of preferences. The vector of priorities is calculated using Monte Carlo simulation and the final rankings analysed for rank reversal using statistical analysis with managerial aspects introduced systematically. The concept and implementation of SAHP is new to the selection of the requisite Six Sigma approach and as such it constitutes the main innovation to result from this research. It extends the selection of the requisite Six Sigma approach towards a systematic multi-criteria decision considering multiple and conflicting criteria under uncertainty. Furthermore, while SAHP was originally conceived as a specific aid to the improve or redesign issue within Six Sigma, this research indicates that it is potentially much more widely applicable. This research also provides evidence of how different factors affecting the selection of requisite Six Sigma approach were considered. Further areas of research include the use of a positivist method in order to increase the sample size of the research and identify different factors affecting the decision improve or redesign. The development of SAHP software and extending the SAHP practice to different multi-criteria decisions are also potential areas for further research.

This work investigates and identifies the mechanisms that are at work in the creation of instabilities during co-injection moulding. Two aspects are investigated, neither of which has been previously reported. One seeks to eliminate the instabilities, the other to control them to produce mechanical interlocking of incompatible polymeric materials. Complex rheological and thermodynamic interactions take place during the co-injection moulding of materials of different generic families, which need to be understood before successful multi material mouldings can be achieved. Moulding trials on miscible, compatibilised and immiscible polymer systems were carried out to determine processing parameter effects. Analysis of tensile behaviour identified differences between injection moulding and co-injection moulding samples which are indicative of different heating and cooling regimes in the systems. Scanning electron microscopy analysis also assisted explanation of these effects. A previously unobserved bulk weakness in compatibilised systems was found. Surface profilometry was used to measure the size of disturbances at the wave fronts. The extent to which the interfacial instability occurs and to the material systems to which it applies was found. Instabilities were found or induced in all material systems investigated, including those where skin and core materials are the same. Mechanisms of instability at the melt front interface were determined and were found to be the result of stratification of elastic properties. Processing conditions were found to minimize instability by minimizing differences in elasticity at the interface. A novel process route using controlled instabilities was also proposed and investigated for the use of immiscible material systems. By controlling the moulding parameters, the potential problem of instability was used to provide a solution to bonding immiscible materials in co-injection mouldings without the use of compatibiliser.

This document is an Executive Summary of individual submissions of work that the author has submitted towards the degree of Engineering Doctorate. The work comprises three main themes, which can be demonstrated in a broader sense as contributing towards improved productivity in fusion welding: i) The use of active fluxes for Tungsten Inert Gas (TIG) welding. ii) An investigation into the reduction of porosity when Metal Active Gas (MAG) welding galvanneal coated steel sheet used in the automotive industry. iii) The use of high power Nd:YAG laser welding for the production of large diameter, long distance land pipelines. Active fluxes give improved productivity by increasing the penetration depth of the TIG welding process by the simple addition of a flux applied to the surface. Although the productivity benefits of the process had been proven through a joint TWIIindustry project, the mechanism by which the fluxes produced this improvement was not fully understood. The first theme investigated the mechanisms at work in providing increased penetration and concluded that the primary mechanism responsible for the action of the fluxes was not due to a change in the flow of the molten pool but, as others had suggested, due to arc constriction. This work contributed to a greater understanding of the welding process and, furthermore, a greater understanding of the potential opportunities and limitations of the process when designing new fluxes for other alloy systems. MAG welds in coated steel sheet used in the automotive industry are prone to porosity leading to high reject rates. The second phase of work reported here determined welding procedures capable of delivering low porosity welds developed through statistical experimental design. These procedures demonstrated how low porosity welds could be made using conventional MAG welding techniques on steels that had been galvanneal coated to provide corrosion resistance. The procedures developed could be easily implemented at high production rates in an industrial manufacturing environment to reduce defect levels, and thus costly repairs or high scrap rates. The third theme of the work demonstrated how Nd:YAG laser welding could potentially be used to replace conventional arc welding techniques for land lay of gas transmission pipelines. The application of a single laser fill pass, made at high production rates, could replace the use of multiple MAG welding stations greatly reducing the costs associated with pipeline fabrication. BP has claimed that half pipeline cost savings of up to $300 million dollars are achievable through the implementation of such a technique. The justification for the use of lasers in pipelines is discussed in terms of both technical and economic suitability. Preliminary experimental work showed that high power Nd:YAG laser welds could achieve productivity targets, although in order to reduce defects and achieve the necessary structural performance it would be necessary to combine laser welding with a MAG welding process.

The quality of a composite material produced using a textile reinforcement depends largely on the way the textile deforms during processing. To ensure the production of high quality parts and minimise costs in designing such parts it is necessary to develop methods to predict the deformations of textiles. This thesis employs a multi scale modelling approach in predicting mechanical properties of textile fabrics. The three scales involved are the microscopic, mesoscopic and macroscopic. This thesis concentrates on the micro and mesoscopic scales leading to results applicable to the macroscopic scale. At the microscopic scale fibres are modelled as individual entities and the interactions between these entities are modelled. In compaction of yarns, the contact between fibres and bending resulting from these contacts governs the force response. A numerical model is developed to simulate this behaviour and results are validated against experimental studies found in the literature. The numerical model is extended to the mesoscopic scale where the shear of a plain woven fabric consisting of low filament count yarns is modelled. At the mesoscopic scale a large part of the work consists of characterising the geometry of textile fabrics. New and existing algorithms are combined together to form a consistent modelling approach. This work was performed in conjunction with the development of a software package named TexGen where these algorithms are implemented. The geometric models created by TexGen are then used to predict mechanical properties of textile unit cells using a finite element method which takes yarn properties as an input. Validation is performed for a series of woven fabrics subjected to compression and in-plane shear.

This thesis describes a technology selection framework for manufacturing technology selection. Technology selection has been identified as a major area of decision making in a company's manufacturing strategy and is highlighted as a manufacturing objective that can provide competitive advantage to a company. The research identifies the emergence of global supply chains as a major phenomenon that has revolutionised the global business environment. Considering the presence of global supply chains and their ever increasing importance the research has proposed a process for manufacturing technology selection keeping in view the supply chain perspective. The technology selection framework introduces the concept of risk evaluation of available technology alternatives for strategic technology selection. The risk associated with technology alternatives is evaluated in the shape of opportunities and threats. The decision making environment for technology selection is divided in a way to consider intra as well as inter-organisational factors. The classification of the decision making environment, inclusion of risk calculations and consideration of a supply chain perspective enables the developed technology selection framework to thoroughly evaluate a technology alternative before its strategic selection. The research presented in this thesis is composed of two main sections. The first section deals with the development of the technology selection framework, whereas the second section describes the application of the developed framework in an aerospace manufacturing company in detail. The application of the framework in industry helped in understanding the issues surrounding the technology selection process and provided an insight into how the existing technology selection processes can be improved and why it is necessary to address the supply chain factors functionally as well as holistically in manufacturing technology selection. The major contribution of this research is a technology selection framework integrating manufacturing and the supply chain. Academically the research establishes a link between manufacturing technology selection and the supply chain and emphasises the importance of alignment between manufacturing and supply chain objectives.

Agglomerated and sintered 25Ni(Cr)-75Cr3C2 powder was deposited onto 304 stainless steel substrates by HVOF thermal spraying. Post spraying thermal treatments were carried out on as-sprayed samples at different temperatures (700°C and 800°C) in air (oxidising). Samples were heat treated from 1hour up to 16 days and furnace cooled. The powder, coatings and substrate were characterised by a variety of techniques including optical microscopy, X-ray diffraction (XRD), scanning electron microscopy (SEM), microhardness testing and thermogravimetric analysis (TGA). The results of agglomerated and sintered powder coating microstructural characterisation show that the as-sprayed coating comprised the carbide phase Cr3C2, a small amount of amorphous phase and a metallic phase which was Ni-rich. The spray process also led to the formation a small amount of Cr2O3 in the coatings. In the heat treated coatings, the carbides Cr7C3 and Cr23C6 were found at the coating substrate interface, the amorphous structure phase was eliminated. With heat treatment, matrix phase supersaturation was reduced, while widespread carbide nucleation and growth generated an expansive carbide skeletal network especially near the coating/substrate interface. An initial softening of the coatings occurred but subsequently the hardness increased again after 2 days was a function of carbide development. The oxidation kinetics of the 25Ni(Cr)-75Cr3C2 powder and coatings during elevated temperature oxidation in air at 650°C to 1000°C have also been studied. The oxidation behaviour was found to be governed by a parabolic rate law up to 800°C. The activation energy of the coating oxidation reaction, Q, was found to be 164 kJ/mol. The mechanism of high temperature oxidation of 25Ni (Cr)-75Cr3C2 coating on top surface has been discussed. Inter-diffusion between the coating and the 304 stainless steel substrate has been investigated. A distribution of the precipitate phase Cr23C6 within the 304 steel was found from etched substrate samples. The decrease of the microhardness measured with distance away from the interface also revealed that diffusion occurred between coating and substrate during the annealing period. Carbon diffusion distances at different annealing conditions in this work have been measured according to the appearance of the Cr23C6 precipitates and variation in microhardness away from the interface. A simple mathematical model has been used to calculate the theoretical carbon diffusion distance in substrate.

The future of manufacturing in high-cost economies is to maximise responsiveness to change whilst simultaneously minimising the financial implications. The concept of Reconfigurable Assembly Systems (RAS) has been proposed as a potential route to achieving this ideal. RASs offer the potential to rapidly change the configuration of a system in response to predicted or unforeseen events through standardised mechanical, electrical and software interfaces within a modular environment. This greatly reduces the design and integration effort for a single configuration, which, in combination with the concept of equipment leasing, enables the potential for reduction in system cost, reconfiguration cost, lead time and down time. This work was motivated by the slow implementation of the RAS concept in industry due, in part, to the limited research into the planning of multiple system reconfigurations. The challenge is to enable consideration of, and planning for, the production of numerous different products within a single modular, reconfigurable assembly environment. The developed methodology is to be structured and traceable, but also adaptable to specific and varying circumstances. This thesis presents an approach that aims towards providing a framework for the configuration of modular assembly systems. The approach consists of a capability model, a reconfiguration methodology and auxiliary functions. As a result, the approach facilitates the complete process of requirement elicitation, capability identification, definition and comparison, configuration analysis and optimisation and the generation of a system configuration lifecycle. The developed framework is demonstrated through a number of test case applications, which were used during the research, as well as the development of some specific technological applications needed to support the approach and application.

Three HVOF and a VPS CoNiCrAlY coating were sprayed using Praxair CO-210-24 and Sulzer Metco Amdry 9951 powders. All coatings were subject to vacuum annealing treatments at different temperatures for different exposure times in order to achieve equilibrium microstructures in all test samples. Both HVOF and VPS, as sprayed and heat treated coatings were oxidised at 850°C and 1100°C for periods of up to 250 hours. Scanning electron microscopy (SEM), energy dispersive X-ray analysis (EDX), transmission electron microscopy (TEM), X-ray diffraction (XRD) and laser diffraction powder size analyses were used in order to analyse the powder particle size distribution and characterise the microstructure of the as-received powders, as-sprayed coatings and annealed and oxidised coatings. Thermogravimetric analysis (TGA) was carried out on as-sprayed and heat treated HVOF and VPS coatings at 1100°C for 96 hours in order to compare the oxidation rate of HVOF and VPS coatings and investigate the effect of microstructure and heat treatment on the oxidation rate of both types of coatings. Furthermore, micro hardness, nano hardness and dynamic mechanical analysis (DMA) with a 3-point bend clamp were used in order to study the hardness and Young’s modulus of as-sprayed and annealed HVOF and VPS coatings. The Eshelby inclusion model and image analyses were used in order to further investigate the effect of microstructure on the mechanical properties of these coatings. It was observed that heat treatment prior to oxidation will change the dual oxide layer observed for the as-sprayed oxidised coatings to a single alumina layer for the heat treated and oxidised coatings. Furthermore, it was observed that microstructural features such as porosity and oxide stringers have an effect on the oxidation rate of coatings. A model was presented indicating the role of oxide stringers and porosity on the resulting heat treated microstructure and oxidation rate. It was also understood that the beginning stage of oxidation can play an important role on the oxidation behaviour (oxidation rate and different oxides present) of both HVOF and VPS coatings. It was observed that annealing will change the microstructure of the coatings and affect their mechanical properties. It was seen that annealing reduced the hardness of the thermally sprayed coatings while it had increased the Young’s modulus of the coatings. It was also observed that features such as porosity levels and β volume fraction have an effect on the mechanical properties of the coatings.

With the rapid proliferation of new products into the marketplace, understanding emotional responses may offer a differential advantage beyond traditional hedonic measures. Thomson et al. (2010) argued that consumers also associate other functional connotations (e.g. refreshing) and abstract feelings (e.g. sophisticated) to a product, referring to these associations (emotional, abstract and functional) as 'conceptualisations'. The aim of this project was to investigate the effect of the sensory attributes and packaging cues of commercial blackcurrant squashes on consumers' liking and conceptualisations. Initially, the sensory attributes of the squashes were characterised using a sequential approach of quantitative descriptive analysis (QDA) and temporal dominance of sensations (TDS). Using QDA and TDS in tandem was revealed to be more beneficial than each on its own, providing a fuller sensory profile. Next, emotional response and liking within the squash category was measured using the EsSense Profile TM, in which consumers rated a predefined emotion lexicon (n=100) under three conditions: (1) blind, (2) pack and (3) informed (product and packaging). The project also measured how emotional, abstract and functional responses changed across blind, pack and informed conditions. A conceptual lexicon was defined by consumers (n=29), after which a different group of subjects (n=100) rated the squashes using the lexicon and a check-all-that-apply (CATA) approach (CD-CATA). The findings of both EsSense Profile and CD-CATA experiments revealed that intrinsic sensory attributes had more association with emotions and liking, than the packaging. Interestingly, the CD-CATA experiment suggested that extrinsic packaging cues had more association with abstract/functional conceptual responses. The relationship between liking and emotional responses to debranded squash (sensory attributes) was investigated comparing EsSense Profile and CD-CATA approaches. Both approaches yielded emotional data that clearly discriminated across the products more effectively than the hedonic scores. In addition, both approaches produced similar emotional spaces and product configurations. A two dimensional structure (pleasantness vs. engagement/activation) corresponding to published circumplex emotion models was observed in each method. The final phase of the PhD was to determine the relationship between sensory attributes of the squashes (as measured by QDA and TDS) and consumer responses (EsSense Profile and CD-CATA approaches). Sensory attributes in squashes that were found to drive liking and positive conceptual responses in consumers were 'natural processed blackcurrant' and 'natural sweetness'. The study also shows how some temporally dominant sensory attributes (e.g. 'minty') evoked positive conceptual responses in consumers. Throughout the thesis, recommendations regarding practical implications for emotion measurement and general ideas for future research are discussed.

Many items of medical, aeronautical, electronic and military equipment exposed to corrosive conditions, or required to have extreme performance characteristics, are sealed hermetically into micro packages. Laser beam welded (LBW) Ti6AI4V alloy has been adopted in anti-corrosion micro packages for the impeller of a left ventricular assistance device (LVAD). Thin and narrow welds were required for such medical equipment. A wide variety of laser types can be applied in sealing micro packages, which include traditional lasers (CO2, Diode and Nd: YAG) to the newest laser types (fibre). Compared with other LBW types, continuous wave fibre laser welds are well known for exhibiting narrow weld zones, low distortion, lower heat input and high efficiency. However, in this work significant porosity was found in the continuous wave fibre laser welds due to the high traverse speeds and high associated solidification rates. The largest distortion and melting area was found in the continuous wave diode laser welds due to the high heat generation. A pulsed Nd: YAG welding was suggested as the hermetic laser welding technique for sealing the micro packages, since it is a good-balance between low porosity, less distortion and a narrow weld zone. The microstructures of Ti6AI4V were complex and strongly affected the mechanical properties. These structures include: a' martensite, metastable ß, Widmanstätten, bimodal, lamellar and equiaxed microstructure. Bimodal and Widmanstätten structures exhibit a good-balance between strength and ductility. The microstructure of laser beam welded Ti6AI4V was primarily a' martensite, which showed the lowest ductility but not significantly high strength. A heat treatment at 950°C followed by furnace cooling can transform the microstructure in the weld from a' martensite structure into Widmanstätten structure. The fatigue fracture behaviour of laser beam welded thin sheet Ti6AI4V was examined in this project as the lifetime of the LVAD impeller has been seen to be limited by fatigue cracking. Grain size, phase content, stress ratio and frequency are all affected fatigue fracture behaviour. A transformed Widmanstätten structure in the weld gave the highest fatigue fracture life. A time of 8 hours heat treatment gave the highest fatigue fracture life due to a good-balance between the grain size and phase content; the existence of metastable (3 limits the crack propagation as well. The geometries, distortion and porosity of the welds were variable and depended on the types of the laser used for the welding process. These factors significantly affected the local stress levels during the tensile testing and fatigue testing. FEA was used to understand and evaluate these influences. By simulation in FEA, the maximum stress locations were observed to be strongly dependent on the specifics of the crosssectional geometries of the welds. For the pulsed Nd: YAG weld in the fatigue testing, the FEA-derived modified local stress amplitudes exhibited a increase of about 40% compared to the nominal applied stress.