Integrating design attributes, knowledge and uncertainty in aerospace sector

CIRP Journal of Manufacturing Science and Technology 7 (2014) 83–96

Contents lists available at ScienceDirect

CIRP Journal of Manufacturing Science and Technology journal homepage: www.elsevier.com/locate/cirpj

Integrating design attributes, knowledge and uncertainty in aerospace sector Tariq Masood a,b,c,*, John Ahmet Erkoyuncu a,c,d, Rajkumar Roy a,c,d, Andrew Harrison b a

Manufacturing and Materials Department, Cranfield University, Cranfield, MK43 0AL, UK Life Cycle Engineering, Rolls-Royce plc, P.O. Box 31, Derby DE24 8BJ, UK c EPSRC Centre for Innovative Manufacturing in Through-life Engineering Services, Cranfield University, Cranfield, MK43 0AL, UK d Operations Excellence Institute, Cranfield University, Cranfield, MK43 0AL, UK b

A R T I C L E I N F O

A B S T R A C T

Article history: Available online 28 March 2014

The delivery of integrated product and service solutions is growing in the aerospace industry, driven by the potential of increasing profits. Such solutions require a life cycle view at the design phase in order to support the delivery of the equipment. The influence of uncertainty associated with design for services is increasingly a challenge due to information and knowledge constraints. There is a lack of frameworks that aim to define and quantify relationship between information and knowledge with uncertainty. Driven by this gap, the paper presents a framework to illustrate the link between uncertainty and knowledge within the design context for services in the aerospace industry. The paper combines industrial interaction and literature review to initially define the design attributes, the associated knowledge requirements and the uncertainties experienced. The framework is then applied in three cases through development of causal loop models (CLMs), which are validated by industrial and academic experts. The concepts and inter-linkages are developed with the intention of developing a software prototype. Future recommendations are also included. ß 2014 CIRP.

Keywords: Knowledge Design Uncertainty Digital feedback Life cycle

1. Introduction The aerospace industry is experiencing a shift from ad-hoc service provision to integrated product and service solutions that enable the delivery of the availability and capability required from an engine [1]. This has promoted an emphasis of the life cycle implications of engine design due to the shift in the business model, which incentivises reduced maintenance cost whilst enhancing equipment operability/functionality [2]. The need to predict service requirements much earlier than the traditional model (e.g. spares sales) and the bundled nature of service delivery has increased the uncertainties experienced by the Original Equipment Manufacturer (OEM) [3,29]. As a result, the OEMs are facing challenges associated with the boundaries of their knowledge in delivering services within the emerging business model [27].

* Corresponding author at: Department of Engineering, University of Cambridge, Cambridge CB3 0FS, UK and Centre for Process Excellence and Innovation, University of Cambridge, Cambridge CB2 1AG, UK. E-mail address: [email protected] (T. Masood). http://dx.doi.org/10.1016/j.cirpj.2014.02.001 1755-5817/ß 2014 CIRP.

Knowledge can be defined in terms of a justified true belief [4]. It involves personalised information, which is processed in the minds of individuals [5]. In an industrial setting, knowledge is considered as an ‘actionable understanding’. Knowledge has typically been classified into tacit and explicit knowledge and the associated contents depend on the context. Tacit knowledge refers to the personal and experience based nature of knowledge [6]. On the other hand, explicit knowledge involves formally documented, systematic, and well-structured language [4]. Knowledge in context of life cycle design includes a number of aspects associated to different phases of an aero-engine [7]. The existence of knowledge enhances the confidence in events that have been predicted. Uncertainty refers to things that are not known or known imprecisely [8,15]. The sources of uncertainty have often been classified into two bases, including epistemic and aleatory [9]. Aleatory uncertainty refers to the uncertainty that arises from natural, unpredictable variation in the performance of the system under study [10]. On the other hand, epistemic uncertainty arises from lack of knowledge about the behaviour of the system that is conceptually resolvable [11]. It is worth recognising that uncertainty does not have to hold negative consequences, it may also lead to

84

T. Masood et al. / CIRP Journal of Manufacturing Science and Technology 7 (2014) 83–96

positive outcomes. Though, it may have a constraining role from a decision-making perspective when designing an engine. The link between knowledge and uncertainty has often been highlighted (particularly in the case of epistemic uncertainty). Ackoff [12] presents that with increased knowledge the level of uncertainty diminishes, whilst emphasising a close association. Understanding the relationship (e.g. root causes) between uncertainty and knowledge can enhance decision making during the design process, whilst influencing the life cycle [28]. For instance, it will be possible to conduct cost-benefit analysis to understand the value of changing the level of knowledge. In light of the challenge of achieving optimised engine design, this paper aims to develop a framework/methodology to demonstrate the influence of knowledge on uncertainty and the implications of changing the level of knowledge on the level of uncertainty experienced in life cycle design. The objectives include:  Capture uncertainties;  Capture design attributes; and  Build a mechanism that links the level of knowledge and the level of uncertainty. Uncertainty in design, design attributes, and knowledge are discussed from academic and industrial contexts in the following sections. A digital decision making framework based upon these is also presented along with its application through CLMs. This is followed by validation, conclusions and future work.

2. Methodology An iterative process was followed to accomplish the objectives of this paper. Close industrial interaction was achieved with four major organisations. A number of suitable research strategies have been considered in formulating the research design to this study. Throughout the research a range of research strategies

were applied including, workshops and interviews. The selection of these approaches has been driven by the industrial context of the study and the research focus, which has necessitated an indepth interaction to understand the current practice and experienced challenges and to validate the developed framework. Fig. 1 demonstrates the steps that were followed as part of the overall methodology for this paper, integrating design attributes, knowledge and uncertainty (DKU). The first phase focused on understanding the context, where extensive literature analysis and outcomes from attended conferences supported in understanding the types of uncertainties, knowledge and attributes that are commonly considered during the design stage. A rigorous keyword search using service, engine design attributes, uncertainty, cost, design and risk register guided the study. During this stage, industrial interaction was also achieved through collaboration with four major defence and aerospace organisations in the UK. This involved conducting semi-structured interviews. Initially, the focus was on the outcomes of the literature review and the aim was to assess the types of uncertainties, knowledge and engine design attributes that were realised from literature. A total of over 40 h of semi-structured interviews were conducted with designers, attribute owners, cost engineers, project managers, support managers, engineering managers, and functional experts (e.g. in risk and uncertainty). The triangulation approach was adopted to analyse outcomes from the interactions. This involved transcription of the interviews, developing mind maps and writing reports to illustrate the learning to collaborating organisations. Samples of the key questions used in the interviews included:  What are the attributes considered during engine design?  What are the types of uncertainties experienced across design attributes?  How does knowledge affect uncertainty?

Fig. 1. Methodology – DKU.

T. Masood et al. / CIRP Journal of Manufacturing Science and Technology 7 (2014) 83–96

The second phase involved organising a workshop to derive relationships between uncertainties, knowledge and design attributes. The principal cost lead, the attribute owners (unit cost, specific fuel cost, weight), the life cycle engineering lead and the research and technology lead attended this three-hour long workshop. In the third phase of the research, initially the concepts were assessed in terms of how comprehensive they were. For instance, the list of uncertainties was assessed through two semistructured interviews, which lasted for an hour each. The first participant was a software tool consultant with expertise across the defence domains associated to cost estimation with over 20 years of experience. The second interview was conducted with an Integrated Logistics Support Manager with over 12 years of experience. The interviews enabled to confirm that the uncertainty findings were applicable across the defence domains and different organisations. The framework and conceptual models were also validated through a workshop of industrial and academic experts. The following section will present state of the art in life cycle design from academic and industrial perspectives. 3. The state of the art in life cycle design A number of challenges have been illustrated in literature for the design phase of a service oriented business model [13,14]. Along these lines, uncertainty plays a critical role, where the dynamic behaviour experienced in service delivery (e.g. maintenance, repair, and capability upgrade) including manipulation of requirements causes challenges. Thus, at the design stage there is a need to be able to adopt a life cycle view of the equipment through the individual design attributes from a systems perspective. The role of knowledge has often been recognised in literature in association with one’s perception of uncertainty. Along these lines this section aims to highlight literature associated to uncertainty, design attributes and knowledge. 3.1. Uncertainty in design – a challenge Uncertainty is an issue of confidence in decision-making, which is caused by the difference between the amount of information or knowledge required to perform a task and the amount of information or knowledge already possessed by the organisation [15,18]. Decision-making is influenced by uncertainty and choices are being made between alternatives based on expected gain and the likelihood of their outcomes. From this perspective, uncertainty may concern [16]:  Probability of an event (state uncertainty),  Lack of information or knowledge about the outcomes of an event, which is a central theme of this paper, and  Underlying cause - effect relationships (effect uncertainty), or a lack of information or knowledge about response options and their likely consequences (response uncertainty).

85

Ackoff [12] described a knowledge continuum to better define perceptions of the environment. This was developed to represent the deviation between no knowledge or ignorance and the scenario of complete knowledge, where the system is capable of providing required data to support analysis. Risk is considered to be a subcategory of uncertainty, where it is possible to assign probabilistic values in terms of likelihood of occurrence and impact of an event. This is shown in Fig. 2. The move to services has created challenges in collecting and generating information and knowledge at the design stage [13]. The highlighted list of uncertainties below was determined to be unsuitable for the purpose of this research due to the lack of emphasis on service delivery, while schedule is considered to be outside the scope of the research. Studies by the Ministry of Defence, Department of Defence, NATO, and NASA specify a range of categories of uncertainties [17], including:  Business or economic: variation from change in business or economic assumptions.  Cost estimating: variations in the cost estimate despite a fixed configuration baseline.  Programme: risks outside the programme office control.  Requirements: variation in the cost estimate caused by change in the configuration baseline from unforeseen design shifts.  Schedule: any event that changes the schedule-stretching it out may increase funding requirements, delay delivery, and reduce mission benefits.  Software: cost growth from overly optimistic assumptions about software development.  Technology: variations from problems associated with technology maturity or availability. The identification of uncertainty is part of the higher-level challenge of managing uncertainties. ‘‘Uncertainty management is not just about managing perceived threats, opportunities and their implications. It is also about identifying and managing all the many sources of uncertainty that give rise to and shape our perceptions of threats and opportunities. It implies exploring and understanding the origins of project uncertainty before seeking to manage it, with no preconceptions about what is desirable or undesirable’’ [18]. Thus, understanding the role of knowledge in managing uncertainty is a major element of the design stage, whilst no direct associations between these concepts within the engine design context in a comprehensive manner have been realised. 3.2. Design attributes – as enablers There are three major stages of the product design process. Concept Design Stage is concerned with the product function. During this stage, the intended functions of the product and potential solutions to achieve them are explored. Preliminary Design Stage is concerned with the relationship between function and form. The requirements and functions finalised during the

Fig. 2. Knowledge continuum: varying levels of knowledge and uncertainty.

86

T. Masood et al. / CIRP Journal of Manufacturing Science and Technology 7 (2014) 83–96

concept stage are transformed into an initial engineering general arrangement (i.e. a physical representation) during this stage. Detailed Design Stage is concerned with the detail form of every component in the product. The arrangement from the preliminary stage is optimised and finalised, each part is fully defined (including geometric dimensions and tolerances), the final material selection takes place and the product is assessed for technical and economic viability. The necessary documentation is also created to enable the product to be produced and maintained. Product design has also been categorised in the literature as original or adaptive [32,33]. An original design is a completely new solution and product. An adaptive design will satisfy an existing requirement by providing solution in a new way, therefore requiring new or substantial changes to existing components and possibly assemblies. Adaptive design may also be applied to incremental improvement to an existing solution to meet a new requirement. Design for service (DfS) is a design process which aims to reduce maintenance costs at design stages by supporting product design with service information. The knowledge pyramid is important to designers in this context. In literature, several authors discussed certain aspects of service information that are beneficial to designers [31]. Norman [34,35] mentioned that past operating experience could contribute towards forecast reliability/availability that also depends on the sample size. Jones and Hayes [36,37] discussed the value of collecting field failure information and during a product’s life, and the analysis of this data to assess the product’s reliability. Petkova [38] described the flow of in-service information back to the manufacturer (within the context of consumer electronics industry) and stressed the significance of the failure root causes for the improvement of product quality. In order to benefit DfS process, identification of information types is important. It is important to see the industrial perspective of uncertainty challenges and design attributes and their knowledge as enablers. 3.3. Knowledge in design – as enabler Knowledge is defined in many different ways depending upon specific domain. Hey [39] defined knowledge as subjective, personal and shaped by an individual’s perceptions and experiences [39]. Young et al. [40] defined knowledge as interpretation of information in order to assign meaning [40]. Hence, it is information with understanding either within human head, manually documented or computerised/automated e.g. pilot or autopilot. Knowledge can be typified broadly in terms of tacit or explicit. Tacit knowledge is the one that comes typically from experience. It is quite unstructured and hard to communicate [4][6]. Tacit knowledge can be of three types: (i) individuals’ education, abilities, know-how; (ii) individuals’ ideas found in publications or patents; and (iii) individuals’ consciousness of others’ knowledge within and out of organisational boundaries [41]. Explicit knowledge is the one that can be transmitted in formal, systematic and well-structured language [4,6]. It should be noted that this paper takes into account the research that attempts to capture tacit knowledge and makes it explicit by externalising through knowledge capture and re-use. Knowledge capture (or knowledge elicitation) is important as loss of knowledge commonly occur when employees spend part of their time to acquire knowledge for analysis and resolution of problems. They would possibly look for similar issues that have occurred in the past, but the knowledge used in the provided solutions might not have captured the required knowledge. Hence, knowledge capture (elicitation) helps the organisation in retention of knowledge for future use [6]. Knowledge re-use is defined as sharing of best practice for people to resolve common technical

issues [42]. Weise [19] defined knowledge re-use as sharing of information and documentation. The theory of knowledge re-use, presented by Markus [42], emphasises on the role of knowledge management systems and the repositories. The knowledge is systematically processed and stored, then re-used repeatedly when any similar situation arises. 4. Industrial perspective 4.1. Uncertainty in design – a challenge There are many types of uncertainties from a service perspective that can be experienced during the product design process. The sources vary driven by a number of factors and their degree of influence evolves over time. Major categories of uncertainties experienced in service delivery include:  Engineering uncertainty considers factors that affect strategic decisions with regards to the future service and support requirements (i.e. how will the service be delivered? Offshore, obsolescence management, rate of system integration issues);  Operation uncertainty considers factors that affect service and support delivery involved on a daily basis. It focuses on equipment level activities (i.e. how much service need will there be? Onshore, maintenance, quality of components and manufacturing, operating parameters);  Affordability uncertainty considers the predictability in the customers ability to fund a project throughout its contractual duration (e.g. customer’s ability to spend, customer willingness to spend);  Commercial uncertainty considers factors that affect the contractual agreement, (e.g. exchange rates, interest rates, commodity and energy prices);  Performance uncertainty considers factors that affect reaching the performance goals (e.g. key performance indicators); and  Training uncertainty considers factors that affect the delivery of training to the customer. The specified categories of uncertainties may have a strategic or operational influence over the design considerations. Along these lines, the affordability and commercial categories guide how the contract should be agreed at the outset from a financial perspective, whilst also taking account of relationships across the supply network. Industry and the customer jointly contribute the level of uncertainty experienced in these categories. On the other hand, the influence of the operation, engineering and training categories tend to be at an operational level on how service and support is to be delivered. It is also interesting to note the inter-linkages between each of these categories. For instance, with the delivery of training the uncertainty in the performance of the equipment reduces. This is mainly associated to the enhanced skill level to operate the equipment. 4.2. Design attributes – as enablers Within the context of this study, design attributes represent key features of customer requirements regarding aerospace-engine design architecture. Some of the key attributes include specific fuel consumption, weight, maintenance cost, and unit cost. Each design attribute should be considered as a source of value to the customer (increasing their revenue potential or reducing their costs). Whilst there are many design attribute level options to achieve product level requirements, analysing different options in a systematic and rapid manner is essential. Variation in options is driven by the performance against targets for each of the attributes, which may necessitate improving some design attributes and downgrading

T. Masood et al. / CIRP Journal of Manufacturing Science and Technology 7 (2014) 83–96

others. Thus, the manufacturer needs to devise measures (e.g. choose a design attribute value to change) to account for any difference between the current design attribute state and the customerrequired level. Following are the key engine design attributes:  Specific Fuel Consumption (SFC): The weight flow rate of fuel required to produce a unit of power or thrust, for example, pounds per horsepower-hour;  Weight: Whilst the granularity may vary (e.g. engine, component) it focuses on the weight in the product e.g. pounds;  Noise: The total noise from all sources other than a particular one of interest (usually measured in decibels);  Unit cost: The cost of a given unit of a product;  Life Cycle Cost (LCC): A measurement of the total cost of using equipment over the entire time of service of the equipment; includes initial, operating, and maintenance costs;  Emission: The substance discharged into the air, e.g. by internal combustion engine;  Development and testing cost: Costs incurred during development and testing;  Thrust: A propulsive force produced by fluid pressure or change of momentum of the fluid in a jet engine, rocket engine, etc.; and  Reliability: Consistent and productive engines, parts, etc. Each design attribute will typically be assigned a minimum and maximum (or additional threshold) value agreed with the customer that guides the solution provider throughout the equipment life cycle. In achieving the requirements for each design attribute the solution provider may face a number of factors that influences its performance in achieving these targets. Additionally, the targets may change throughout the life cycle. The performance of the solution provider in reacting to and/or driving design attribute requirements throughout the equipment life cycle partly determines the satisfaction level of the customer and hence influences competitive positioning. 4.3. Knowledge in design – as enabler For the purpose of this paper, knowledge is defined in the industrial setting as ‘actionable understanding’. A knowledge hierarchy (data-information-knowledge-wisdom) is defined, in which simple data could be enhanced up to information, knowledge and then wisdom level by increasing understanding and context independence. The authors contend that industrial value is only released when this hierarchy generates sufficient understanding to enable more effective or efficient decisions and actions to be taken. For example, customer value from an aero-engine is released during the service phase of the product life cycle. Whilst functioning in service the engine contributes to the customer revenue generation (by supplying the motive power). In stark contrast, whilst out of operation for servicing the engine contributes only to costs. It is therefore a key requirement to understand the drivers of loss of function and maintenance requirements in order to achieve the maximum functional availability of the product. A knowledge of the maintenance drivers with availability of mitigation guidance for future designs or redesigns is clearly important and of significant value in this context. Digital feedback of through-life engineering service knowledge to product design and manufacture is challenging. There is a lack of available structured methodologies for capturing and structuring service knowledge in order to map service knowledge onto design requirements. The challenge here is to devise an effective methodology to capture service knowledge gained from previous learning, possibly in a structured way, and then feedback to conceptual and detailed product design stages so that new/revised product designs incorporate the new learning.

87

Service knowledge is also important for the service/repair engineering functions of an organisation, especially for its uses in root cause analysis, problem solving, mitigation of operational risks, improving repair policies, recommendations of repair margins, etc. The knowledge of previous service experience could help reduce product LCC by giving priority to mitigation of risks on those product commodities, which exhibit high costs. Keeping product LCC at minimum is challenging. The following feedback loops of through-life engineering service knowledge are considered in this paper: (1) service to design; and (2) manufacture/assembly to design. The design function has to understand both manufacture, assembly and operation and the challenge is to achieve a balanced design that minimises the design cost impact in all three stages (with appropriate weighting to their impact on customer value). In both cases, establishing effective feedback loop is also challenging. The through-life engineering service knowledge and its impact on product design and manufacture are presented as a CLM in Fig. 3. The CLM is mapped across design, development and service stages of product life cycle. The design stage includes conceptual, preliminary and detailed product design. The development stage includes product engineering, manufacturing, assembly and testing. The service stage includes product service, repair and maintenance. Here links between causes and effects are represented across product life cycle stages that have positive or negative links representing increasing or decreasing effects of related causes. The CLM revolves around enhancing the service knowledge backbone (SKB) of an aerospace organisation, which could be partly achieved by improving service knowledge capture. The enhanced SKB could increase knowledge levels in conceptual and detailed product design stages, which could lead to optimise design characteristics. This could lead to increase confidence in previous design, design robustness, and decrease design costs and hence LCC. Improved design characteristics could lead to improve design of fixtures, tooling and inspection and in effect the actual equipment. This could result in minimising maintenance burden, frequency of occurrence and operational disruption. As a result, the number of maintenance and repair events could be reduced with a commensurate reduction in cost of maintenance and repair leading to a reduction in LCC. Quality could be improved by achieving an increased level of design robustness and together with decreased operational disruption it could result in improved customer service. On other side, an increase in service knowledge capture will also result in higher costs of capturing and maintaining knowledge, hence increasing the LCC. Variability in customer requirements is another factor that could lead to increase design costs, hence increasing the LCC. A robust design may increase requirements of capabilities/skills, while higher confidence in design may reduce these requirements. Improved design of fixtures, tooling and inspection also increases these requirements, on provision of which the state of fixtures, tooling and inspection improves as well as quality. Provision of these requirements will result in higher LCC in both cases. However, there is an optimum point at which the maximum value versus cost of enhancing service knowledge is achieved. A digital decision making framework is presented in the following section. 5. Design attributes-knowledge-uncertainty (DKU) framework A digital decision making framework (DKU framework) is proposed that links the role of design attributes, knowledge and uncertainty. The overall framework is presented in Fig. 4. The DKU framework visualises specific relationships between the design attributes, knowledge and uncertainty in a map form. Industrial product-service system delivery is linked with design attributes, knowledge and uncertainty through outgoing knowledge adaptation and incoming service prediction capability

88

T. Masood et al. / CIRP Journal of Manufacturing Science and Technology 7 (2014) 83–96

Fig. 3. CLM: through-life engineering service knowledge feedback to product design and manufacture.

Fig. 4. The DKU framework.

(as shown in Fig. 4). It is also linked with confidence level, knowledge adaptation, service prediction and capability/ experience as shown in Fig. 4. CLMs are created and presented in the following to further elaborate on the DKU framework. 6. Application of the DKU framework in aerospace sector through CLMs The application of the DKU framework is exercised through development of an overall DKU CLM and detailed CLMs for key design attributes. System dynamics has established itself as a

powerful methodology [20]. A CLM can represent causal effects of activities [10,21]. This type of modelling helps identify aspects of complexities and dynamics can be modelled through this technique [22]. Here, the CLM consists of various elements linked to each other through one or more arrows along with positive or negative signs. The primary focus is on the examination of the effect that one element has on another. The system dynamics models are used primarily to study systems that display feedback characteristics [23]. Therefore, it is considered appropriate that the study of causal relations should be conceptualised as a series of influence and CLMs. The positive (or negative) sign on an arrow line shows a positive (or

T. Masood et al. / CIRP Journal of Manufacturing Science and Technology 7 (2014) 83–96

negative) influence of starting element on the final element. If there is a positive (or negative) sign placed within the element, it would mean the element is already influenced positively (or negatively) from some other source. Implementation of CLMs has been reported in several case studies either standalone or as part of integrated approaches [22,24–26,30]. The CLM of the DKU framework is presented in Fig. 5, which looks into causes and effects related to an enhanced SKB. Here knowledge of nine (9) important design attributes are considered, which includes (as shown in columns): weight, SFC, noise, unit cost, LCC, emission, development and testing cost, thrust and reliability/operational disruption. Desired design attribute trends are taken as initial conditions for this CLM i.e. lower weight, lower SFC, lower noise, lower unit cost, lower LCC, lower emission, lower development and testing cost, higher thrust and lower operational disruption (higher reliability). Uncertainties are categorised into engineering, operation, affordability and commercial. Engineering uncertainties include rate of system integration issues, level of obsolescence, rate of rework, rate of capability upgrade, failure rate of software, maintaining design rights, cost estimating data reliability and quality, efficiency of engineering efforts, and cost of licensing and certification. Operation uncertainties include quality of components and manufacturing, component stress and load, operating parameters, maintainer performance, availability of maintenance support resources, effectiveness of maintenance policy part level, complexity of equipment, equipment utilisation rate, performance of internal logistics, supply chain logistics, rate of materials, sufficiency of spare parts, performance of suppliers’ logistics, failure rate of hardware, location of maintenance, rate of

89

beyond economical repair, turn around (repair) time, choice of fuel, mean time between failure data, no fault found rate, and rate of emergent work. Affordability uncertainties include customer ability to spend, customer willingness to spend, and project life cost. Commercial uncertainties include exchange rates, interest rates, commodity and energy prices, material cost, environmental impact, customer equipment usage, suitability of requirements, labour hour, labour rate, labour efficiency, clarity of customer requirements, and experience in other engine service provision. The DKU – CLM presented in Fig. 5 revolves around causes and effects of an enhanced SKB, which are mapped onto design attributes (in columns) and uncertainties (in rows). The CLM presents positive or negative effects of an enhanced SKB onto uncertainty types resulting in positive or negative effect on design attribute. Taking the reliability design attribute, it proposes that the effect of an enhanced SKB would be negative onto engineering uncertainty for reliability, which further results in higher reliability. It affects similarly on other uncertainties for reliability (operational, affordability and commercial) that are considered in this paper. It should be noted here that the paper discusses uncertainty types and their resultant effects; it does not go into detailed uncertainties, for which increasing or decreasing effect may be different. Thrust (another design attribute) has similar effects to reliability, which ends up in an increase with enhanced SKB while reducing respective uncertainties. The enhanced SKB affects respective uncertainties (engineering, operational, affordability and commercial) negatively for other design attributes (noise, weight, SFC, unit cost, LCC, emission and development and training cost) resulting in negative effect on these design attributes.

Fig. 5. DKU CLM – SKB, design attributes and uncertainties.

T. Masood et al. / CIRP Journal of Manufacturing Science and Technology 7 (2014) 83–96

90

In the following, three cases related to ‘‘reliability’’, ‘‘SFC’’ and ‘‘LCC’’ would be discussed. 6.1. Case 1: reliability Fig. 6 demonstrates the link between knowledge and uncertainty in context of ‘‘reliability’’. The following would explain the CLM shown in Fig. 6. Engineering uncertainties: ‘‘Rate of system integration’’ refers to the combination of individual systems whether developed in

house, outsourced or both. It typically forms a major responsibility of OEMs, whilst uncertainties drive the performance of individual systems and the integrated architecture. Any negative issues that may be experienced result in diminishing reliability and increasing operational disruption. ‘‘Level of obsolescence’’ defines the uncertainty in not being able to find replacement parts. As obsolescence increases, with the arising need to source alternative parts, the reliability diminishes due to the new parts that are introduced to the system. ‘‘Rate of capability upgrade’’ involves technological advancements that are made along the equipment life cycle to

Fig. 6. DKU CLM – SKB, reliability and uncertainties.

T. Masood et al. / CIRP Journal of Manufacturing Science and Technology 7 (2014) 83–96

enhance equipment capability. It creates the uncertainty of how the system will respond to changes. Furthermore, ‘‘Capability upgrade’’ can be made with the ambition of reducing uncertainty in ‘‘reliability’’. Operation uncertainties: ‘‘Quality of components and manufacturing’’ is associated to the reliability of parts that have been developed either internally or externally, which involves

91

uncertainty in the quality. There is a correlation between the quality and reliability. ‘‘Sufficiency of spare parts’’, when considered at the integrated system level, influences the operation of other integrated parts which affect the reliability. Rate of rework’’ largely originates from errors in maintenance, which causes rework in the service provision. As a source of uncertainty it has an influence over reliability. ‘‘Failure rate of

Fig. 7. DKU CLM – SKB, SFC and uncertainties.

92

T. Masood et al. / CIRP Journal of Manufacturing Science and Technology 7 (2014) 83–96

software’’ and ‘‘Failure rate for hardware’’ have a direct influence over the reliability of equipment. Decrease in ‘‘complexity of equipment’’ and ‘‘rate of emergent work’’ would enhance ‘‘maintainer performance’’ that would enhance ‘‘reliability’’. The uncertainty is associated to the when, where and how significant the failure is. ‘‘Maintainer performance’’ considers service delivery from a resource dimension. The uncertainty originates from human centred drivers such as skill and motivation, which influence how the reliability evolves. Affordability uncertainties: ‘‘Reliability’’ forms a central focus of the SKB. Any shift in ‘‘reliability’’ directly influences ‘‘unit cost’’, ‘‘development and training cost’’ and ‘‘LCC’’. Such changes fundamentally affect ‘‘customer’s ability to spend’’ and ‘‘sales and revenues’’. The cost of purchase of new equipment and services will vary driven by the proposed reliability level. The spend will be in purchase of new equipment and services rather than on the maintenance costs that have now been reduced – i.e. it encourages a long term increase in OEM revenue not short term. Commercial uncertainties: Increasing ‘‘scope of warranty’’ and enhanced ‘‘SKB’’ would enhance ‘‘customer relationship’’, which would affect ‘‘reliability’’ positively. Reduced ‘‘material cost’’ and enhanced ‘‘SKB’’ would enhance ‘‘supplier relationship’’, which would affect ‘‘reliability’’ positively. Increase in ‘‘exchange rate’’ would positively affect ‘‘customer’s ability to spend’’, while decrease in ‘‘interest rate’’ would have a positive effect on ‘‘customer’s ability to spend’’.

6.2. Case 2: SFC Fig. 7 demonstrates the link between knowledge and uncertainty in context of ‘‘SFC’’. The following would explain the CLM shown in Fig. 7. Engineering uncertainties: Increase in SKB level would affect negatively on ‘‘rate of system integration issues’’, ‘‘rate of rework’’ and ‘‘level of obsolescence’’ that would reduce ‘‘SFC’’. Such an increase in SKB level also increases ‘‘rate of capability upgrade’’ which would mean increase in ‘‘SFC’’. Operation uncertainties: Increase in ‘‘component stress and load’’ would result in an increase of ‘‘equipment utilisation rate’’ that would mean reduced ‘‘SFC’’. Increased SKB level would reduce ‘‘failure rate of hardware’’ and enhance ‘‘quality of components and manufacturing’’ that would reduce ‘‘SFC’’. Decrease in ‘‘complexity of equipment’’ and ‘‘rate of emergent work’’ would enhance ‘‘maintainer performance’’ that would reduce ‘‘SFC’’. Increase in SKB level would directly influence ‘‘availability of maintenance support resources’’ and ‘‘availability of OEM logistics’’, hence increasing ‘‘maintainer performance’’ that would reduce ‘‘SFC’’. Affordability uncertainties: ‘‘Customer’s ability to spend’’ and ‘‘customer’s willingness to spend’’ have positive impact on ‘‘sales and revenues’’ that would enhance ‘‘organisation’s ability to spend’’ and would contribute in enhancing SKB. In addition, reduced ‘‘unit cost’’ would reduce ‘‘LCC’’ resulting in enhanced ‘‘customer’s ability to spend’’. Commercial uncertainties: Increasing ‘‘scope of warranty’’ and enhanced ‘‘SKB’’ would enhance ‘‘customer relationship’’, which would reduce ‘‘SFC’’. Reduced ‘‘material cost’’ and enhanced ‘‘SKB’’ would enhance ‘‘supplier relationship’’, which would reduce ‘‘SFC’’. Increase in ‘‘exchange rate’’ would positively affect ‘‘customer’s ability to spend’’, while decrease in ‘‘interest rate’’ would have a positive effect on ‘‘customer’s ability to spend’’. ‘‘Rate of environmental impact’’ is reduced with an increase in SKB level resulting in reduced ‘‘SFC’’.

6.3. Case 3: LCC Fig. 8 demonstrates the link between knowledge and uncertainty in context of ‘‘LCC’’. The following would explain the CLM shown in Fig. 8. Engineering uncertainties: Increase in SKB level would affect negatively on ‘‘rate of system integration issues’’, ‘‘rate of rework’’ and ‘‘level of obsolescence’’ that would reduce ‘‘LCC’’. Such an increase in SKB level also increases ‘‘rate of capability upgrade’’, which would increase ‘‘LCC’’. Minimising ‘‘cost of licensing and certification’’ would reduce LCC. Increase in ‘‘maintaining design rights’’ would also reduce LCC in longer terms. Reduction in ‘‘failure rate of software’’ would lead to reduced LCC. ‘‘Efficiency of engineering effort’’ would be increased with enhanced level of SKB, resulting in reduced LCC. Operation uncertainties: Increase in ‘‘component stress and load’’ would result in increased ‘‘equipment utilisation rate’’ that would reduce ‘‘LCC’’. Increased SKB level would reduce ‘‘failure rate of hardware’’ and enhance ‘‘quality of components and manufacturing’’ that would reduce ‘‘LCC’’. Increase in ‘‘complexity of equipment’’ and ‘‘rate of emergent work’’ would reduce ‘‘maintainer performance’’ that would further reduce ‘‘LCC’’. Increase in SKB level would directly influence ‘‘availability of maintenance support resources’’ and ‘‘availability of OEM logistics’’, hence increasing ‘‘maintainer performance’’ that would reduce ‘‘LCC’’. Affordability uncertainties: ‘‘Customer’s ability to spend’’ and ‘‘customer’s willingness to spend’’ have positive impact on ‘‘sales and revenues’’ that would enhance ‘‘organisation’s ability to spend’’ and would contribute in enhancing SKB. In addition, reduced ‘‘unit cost’’ would reduce ‘‘LCC’’ resulting in enhanced ‘‘customer’s ability to spend’’. Commercial uncertainties: ‘‘Exchange rate’’ positively affects while ‘‘interest rate’’, ‘‘inflation rate’’ and ‘‘commodity and energy price’’ negatively affects ‘‘customer’s ability to spend’’, which is also affected positively by increase in SKB level and results in reduced ‘‘LCC’’. Increasing ‘‘scope of warranty’’ and enhanced ‘‘SKB’’ would enhance ‘‘customer relationship’’, which would reduce ‘‘LCC’’. Reduced ‘‘material cost’’ and enhanced ‘‘SKB’’ would enhance ‘‘supplier relationship’’, which would reduce ‘‘LCC’’. Reduced labour cost affected by labour rate, labour hours, labour efficiency and labour availability would reduce ‘‘LCC’’. Increase in ‘‘customer equipment usage’’ and ‘‘stability of requirements’’ would reduce LCC in longer terms. The digital framework is aimed to be developed in MS Excel1 and aims to be used as a decision support tool. The main advantages of using MS Excel1 are associated to its: (1) wide use and availability, and (2) flexibility to make changes. The step-wise input process will involve two sets of input requirements to facilitate qualitative and quantitative analysis. Firstly, the user will be offered to choose from a pre-defined set of attributes, uncertainties and service knowledge types based on their relevance to the project/research at hand. This will assist the qualitative analysis, mainly based on a tick box type approach. Secondly, the quantitative analysis aims to illustrate the degree of dependency between uncertainty and attributes as well as between knowledge and uncertainty. Various approaches such as the analytic hierarchy process, which facilitates pair-wise comparisons, will be implemented to reflect the significance of each element. As an output the tool will show the link between uncertainties and knowledge. The tool will offer further analysis to reflect the benefit in reducing/increasing uncertainty by making a change in the knowledge level. The limitations of the DKU framework may include a fewer industrial testing, which needs to be applied widely in the

T. Masood et al. / CIRP Journal of Manufacturing Science and Technology 7 (2014) 83–96

93

Fig. 8. DKU CLM – SKB, LCC and uncertainties.

industry. The framework could also be compared and tested on platforms other than MS Excel1.

Table 1 Attendees of the validation workshop. Job role

Experience (years)

Unit cost attribute owner Specific fuel cost attribute owner Weight attribute owner Research and technology lead Company associate fellow – life cycle engineering lead

28 24 18 16 26

7. Validation of the DKU framework The validation of the DKU framework involved a workshop, which was attended by five participants from an aerospace engine manufacturer in the UK. An overview of the workshop attendees is provided in Table 1, whilst the workshop lasted for 3 h.

94

T. Masood et al. / CIRP Journal of Manufacturing Science and Technology 7 (2014) 83–96

At the workshop one of the main strengths of framework was suggested as the realistic link between uncertainties and knowledge for the design attributes: LCC, reliability, and SFC. It was highlighted that this will facilitate quantitative management of uncertainties through knowledge. An additional strength was suggested to be use of the causal loops to illustrate the linkages. In particular the visualisation ability that supports decision-making was highlighted. Also, the detailed representation of uncertainties for each of the attributes was considered to be a useful feature of the framework. One of the main weaknesses of the framework, considered at the workshop, was the laborious effort and the time consumed when building the causal loop models for each attribute. Although, this effort is a one-off requirement in the current version of the framework, with the integration of quantitative analysis the burden may increase. It was also suggested that causal loop models could prove complex to understand for those who do not have a background in this area. The main threat of the framework, considered at the workshop, was the multiple interpretations of the concepts considered in the framework. Thus, a level of training would be required for the users of the framework in order to eliminate this issue. Also, from a mechanistic view, the framework would need to facilitate the incorporation of new concepts and to be able to draw connections with other concepts of the framework. At the workshop, more industrial testing was considered as main opportunity of the framework. Additionally, causal loop models for the remaining attributes can also be incorporated to the framework.

8. Discussion The undertaken research has made significant contributions to knowledge in both academic and industrial domains. Main contributions to knowledge are presented in the following. 8.1. Development of the DKU framework The development of the DKU framework is significant in terms of its benefits. The examples follow. It provides a structured way of modelling static and dynamic aspects of manufacturing enterprises with induced change capability and facilitation for improved decision making. It also provides a mechanism for structured data capturing during EM stage where the enterprise models provide a ‘knowledge repository’ in the form of a graphical ‘process oriented structure’. The CLM stage is introduced providing rightful definition of objectives, broader situational analysis and more focussed way of looking into global objectives while considering side effects. It provides advancement in the integrated modelling of manufacturing enterprises in terms of its integrated approach and reusable components leading to practical implementation of the approach. It enables extending reusable integrated models in a stepwise, flexible, and extendable way. This also results in reduced time and effort needed to reconfigure manufacturing enterprises in the event of changing requirements and conditions due to availability of knowledge in the form of reusable integrated models. The system complexity is reduced through improved system decomposition. The reengineering process becomes easier due to availability and reuse of existing models. Optimisation of design attributes is also possible. It provides enhanced, cost effective product and service design as well as better targeting of knowledge requirements and improved understanding of the implications of uncertainty on life cycle design.

8.2. Application of the DKU framework The DKU framework is applied in three industrial cases i.e. ‘‘reliability’’, ‘‘SFC’’ and ‘‘LCC’’. The case company operates in the aero engine provider sector, which powers a range of civil aircraft types ranging from small executive jets through to large passenger aircrafts. From a defence industry perspective the company covers all major sectors including transport, helicopters, combat, trainers and tactical aircrafts. The company has made an emphasis to provide services and aims to build long-term relationships. Furthermore, support contracts are considered to add value to customers by using technology, skills, and data management expertise. The aerospace industry has very high entry barriers and is considered to offer the opportunity for organic growth; the products have long-term lives, and can only be delivered through high investments in technology, infrastructure and capability. Moreover, strong growth is forecasted in the industry, whilst factors such as GDP growth, aircraft productivity, operating costs, environmental issues and number of old aircraft retirements determine the speed of growth. The industrial applications helped in capturing the uncertainties, defining and capturing design attributes and building a mechanism that linked the level of knowledge and the level of uncertainty. The application of DKU framework in three cases helped understand influences of the case related knowledge of uncertainties and design attributes. The CLMs created during application of the DKU framework in these cases proved useful contribution to knowledge in: (a) Providing useful case specific insights and helping in decision making during engine design processes by analytically setting the engine design attributes keeping in mind the knowledge of their influence on uncertainties; (b) Classifying uncertainties according to their commonalities in associated processes; (c) Suggesting that the case specific CLMs held possible potential benefits to identify and understand reasons for changing case specific objectives and has a potential to optimise these in future. The DKU framework is potentially relevant to sectors including but not limited to aerospace, defence, marine, energy, rail and automotive–the processes remain the same but uncertainties and types of knowledge might vary from sector to sector. The maintenance strategies might vary from proactive to reactive. The production scales might vary from mass production (automotive) to customised less numbers (aerospace). The dynamic nature of knowledge availability/volumes/types might vary from fast moving to slow paced. The complexity might be higher in aerospace rather than the automotive or other sectors. 9. Conclusions The paper aimed to propose a framework to integrate the design attributes, knowledge and uncertainties in the aerospace sector. The paper focuses on addressing the challenge of achieving optimised engine design, found from extensive literature analysis and industrial interactions. The paper aimed to develop a framework to demonstrate the influence of knowledge on uncertainty for design attributes with following objectives: (i) how could the influence of knowledge on uncertainty be captured and demonstrated; (ii) how could the design attributes be defined and captured; and (iii) how could the implications of changing the level of knowledge or uncertainty be illustrated. The DKU framework is presented that supports decisionmaking in understanding the link between the level of knowledge

T. Masood et al. / CIRP Journal of Manufacturing Science and Technology 7 (2014) 83–96 Table 2 Planned versus achieved research objectives. No.

Planned research objectives

Achieved research objectives

1

Capture the uncertainties

The research presented in this paper has captured and classified key uncertainties in engineering, operation, affordability, commercial, performance and training The following nine design attributes have been defined and captured: reliability, SFC, LCC, weight, noise, unit cost, emission, development and testing cost, and thrust A mechanism is developed that links the level of knowledge (of design attribute) with the level of uncertainty. The mechanism is applied in three cases of ‘‘reliability’’, ‘‘SFC’’ and ‘‘LCC’’ for which detailed CLMs are developed

2

Define and capture the design attributes

3.

Build a mechanism that links the level of knowledge and the level of uncertainty

and uncertainty in life cycle design within aerospace sector. Based on initial feedback, through industrial interaction (including semi-structured interviews and workshops), there are a number of implications of the proposed framework for decision-making. The framework supports efficient and effective product design through identification of the service impact of design decisions at earlier life cycle stages, where greater design freedom exists. The framework also supports demonstrating the link between achievements of the design attributes and the associated knowledge and uncertainty. This enables us to build an understanding of how uncertainties can influence achievement of attributes and how knowledge could be used to reduce the influence. The DKU framework is then applied through development of CLMs for ‘‘reliability’’ case, ‘‘SFC’’ case and ‘‘LCC’’ case in the aerospace sector. It still needs to be tested in more cases to fully mature. Table 2 critically reviews and concludes the undertaken research showing the research achievements against the planned research objectives. Here case specific assumptions apply. The DKU framework is validated through a workshop of industrial experts as well as semi structured interviews with industrial and academic experts in the field. Despite all the benefits, there are few limitations of the DKU framework due to its limited testing in different industrial sectors. It is also limited to MS Excel1 based software platform, which may be further expanded to other platforms as required by other industries in future. The future research work might also include building and further enhancing relationships between knowledge and uncertainty. Building and further enhancing ‘dynamic’ relationships between knowledge based uncertainty and the design implications is another topic of future research. Developing mechanisms to illustrate the value of changing the level of knowledge in relation to the degree of uncertainty and the implications of this on the life cycle design is also important. Costbenefit analysis of the enhanced knowledge is also required. There is a need for frameworks to assess the knowledge level for design attributes as well as mathematical optimisation of attribute set (over time) given the influence of uncertainty and assessment of the value of knowledge that might include comparison of existing versus required capability and seeking benefits (if any) in changing the knowledge value. It is important to further explore the role and methods for through-life engineering service knowledge feedback to product design and manufacture in life cycle engineering. Application of the DKU framework through quantified simulation models (e.g. design attributes models and cost models) would also give more meaning to the future research agenda. The balanced or comparable data gathering through sensors etc. and value in terms

95

of knowledge levels and reduction of uncertainties might help in quantification of the models. Acknowledgements Rolls-Royce plc, EPSRC and Technology Strategy Board are acknowledged for providing funds to Service Knowledge Backbone Project (Knowledge Transfer Partnership Programme No. 7767). KT-Box Project partners are also acknowledged for providing partial funds. Authors are also grateful for kind support from the Cranfield IMRC for funding this research and the industrial partners for their support.

References [1] Alonso-Rasgado, T., Thompson, G., 2006, A Rapid Design Process for Total Care Production, Journal of Engineering Design, 17/6: 509–531. [2] Datta, P.P., Roy, R., 2009, Cost Modelling Techniques for Availability Type Service Support Contracts: A Literature Review and Empirical Study, in: Proceedings of the 1st CIRP IPS2 Conference (Cranfield), pp.216–223. [3] Erkoyuncu, J.A., Roy, R., Shehab, E., Wardle, P., 2009, Uncertainty Challenges in Service Cost Estimation for Product-Service Systems in the Aerospace and Defence Industries, in: Proceedings of the 1st CIRP IPS2 Conference (Cranfield), pp.200–207. [4] Nonaka, I., 1994, A Dynamic Theory of Organizational Knowledge Creation, Organization Science, 5/1: 14–37. [5] Alavi, M., Leidner, D.E., 2001, Review: Knowledge Management and Knowledge Management Systems: Conceptual Foundations and Research Issues, MIS Quarterly, 25/1: 107–136. [6] Sobodu, O., 2002, Knowledge Capture and Representation in Design and Manufacture,(M.Sc. thesis) School of Industrial and Manufacturing Science, Cranfield University, Cranfield. [7] Doultsinou, A., 2010, Service Knowledge Capture and Re-use to Support Product Design, (PhD Thesis) School of Applied Sciences, Cranfield University, Cranfield: 377. [8] Walker, W.E., Harremoes, P., Rotmans, J., Sluijs, J.P.V., Asselt, M.B.A.V., Janssen, P., Krauss, K.V., 2003, Defining Uncertainty: A Conceptual Basis for Uncertainty Management in Model Based Decision Support, Integrated Assessment, 4/1: 5–17. [9] Erkoyuncu, J.A., Roy, R., Shehab, E., Cheruvu, K., 2010, Understanding Service Uncertainties in Industrial Product-Service System Cost Estimation, International Journal of Advanced Manufacturing Technology, 52/9–12: 1223–1238. [10] Daneshkhah, A.R., 2004, Uncertainty in Probabilistic Risk Assessment, University of Sheffield: 45–87. [11] Thunnissen, D., 2005, Propagating and Mitigating Uncertainties in the Design of Complex Multidisciplinary Systems, (PhD thesis) California Institute of Technology, Pasadena, CA: 56–73. [12] Ackoff, R.L., 1989, From Data to Wisdom, Journal of Applied Systems Analysis, 16/1: 3–9. [13] Martinez, V., Bastl, M., Kingston, J., Evans, S., 2010, Challenges in Transforming Manufacturing Organisations into Product-Service Systems, Journal of Manufacturing Technology Management, 21/4: 449–469. [14] Shehab, E.M., Roy, R., 2006, Product-Service Systems: Issues and Challenges, 4th International Conference on Manufacturing Research (ICMR 2006) (Liverpool John Moores University, 5th–7th September), pp.17–22. [15] Galbraith, J.R., 1974, Organisation Design – An Information Processing View, Interfaces, 4:118–131. [16] Milliken, F.J., 1987, Three Types of Perceived Uncertainty About the Environment: State, Effect, and Response Uncertainty, Academy of Management Review, 12/1: 133–143. [17] United States Government Accountability Office (GAO). 2009, Cost Estimating and Assessment Guide, Best Practices for Developing and Managing Capital Program Costs, . Available at: http://www.gao.gov/products/GAO-09-3SP (accessed 02.09.11). [18] Ward, S.C., Chapman, C.B., 2003, Transforming Project Risk Management into Project Uncertainty Management, International Journal of Project Management, 21:97–105. [19] Weise, R.H., 1996, Representing the corporation: strategies for legal counsel, USA, Aspen Law & Business Publishers. [20] Mohapatra, P.K.J., Mandal, P., 1989, The System Dynamics Paradigm, Decision, 16/4: 251–256. [21] Masood, T., Roy, R., Harrison, A., Xu, Y., Gregson, S., Reeve, C., 2011, Challenges in Digital Feedback of Through-Life Service Knowledge to Product Design and Manufacture, 7th International Conference on Digital Enterprise Technology (DET) (Athens, Greece), pp.1–11. [22] Masood, T., 2009, Enhanced Integrated Modelling Approach to Reconfiguring Manufacturing Enterprises, (PhD thesis) Wolfson School of Mechanical and Manufacturing Engineering, Loughborough University, Loughborough: 314. [23] Mohapatra, P.K.J., Mandal, P., et al, 1994, Introduction to System Dynamics Modelling, Universities Press, New Delhi.

96

T. Masood et al. / CIRP Journal of Manufacturing Science and Technology 7 (2014) 83–96

[24] Rashid, S., Masood, T., Weston, R.H., 2009, Unified Modelling Approach in Support of Organization Design and Change, Proceedings of the Institution of Mechanical Engineers Part B: Journal of Engineering Manufacture, 223/B8: 1055–1079. [25] Zhen, M., Masood, T., Rahimifard, A., Weston, R., 2009, A Structured Modelling Approach to Simulating Dynamic Behaviours in Complex Organisations, Production Planning and Control, 20/6: 496–509. [26] Masood, T., Weston, R., Rahimifard, A., 2010, A Computer Integrated Unified Modelling Approach to Responsive Manufacturing, International Journal of Industrial and Systems Engineering, 5/3: 287–312. [27] Cohen, M.A., Agrawal, N., Agrawal, V., 2006, Winning in the Aftermarket, Harvard Business Review, 84/5: 129–138. [28] Garvey, P.R., 2000, Probability Methods for Cost Uncertainty Analysis, Marcel Dekker, New York: 88–102. [29] Goffin, K., New, C., 2001, Customer Support and New Product Development – An Exploratory Study, International Journal of Operations and Production Management, 21/3: 275–301. [30] Masood, T., Weston, R., 2011, An Integrated Modelling Approach in Support of Next Generation Reconfigurable Manufacturing Systems, International Journal of Computer Aided Engineering and Technology, 3/3/4: 372–398. [31] Ng, I.C.L., Williams, J., Neely, A., 2009, Service Transformation and the New Landscape of Outcome Based Contracting: An Executive Briefing, Advanced Institute of Management (AIM) Research Executive Briefing Series, [Online]. Available at: http://www.aimresearch.org/index.php?page=alias-3 (accessed 09.03.10). [32] Mountney, S.L., 2009, Acquisition and sharing of innovative manufacturing knowledge for preliminary design, School of Applied Sciences, Cranfield University, Cranfield, PhD Thesis.

[33] Metters, R., Marucheck, A., 2007, Service management-academic issues and scholarly reflections from operations management researchers, Decision Sciences, 38/2: 195–214. [34] Norman, D., 1988, Incorporating operational experience and design changes in availability forecasts, Reliability Engineering and System Safety, 20: 245–261. [35] Mey Goh, Y., Newnes, L., Mileham, A., McMahon, C., Saravi, M., 2010, Uncertainty in Through Life Costing - Review and Perspectives, IEEE Transactions Engineering Management, 57/4: 689–701. [36] Jones, J., Hayes, J., 1997, Use of a field failure database for improvement of product reliability, Reliability Engineering & System Safety, 55/2: 131–134. [37] Saccani, N., Johansson, P., Perona, M., 2007, Configuring the after-sales service supply chain: A multiple case study, International Journal of Production Economics, 110/1–2: 52–69. [38] Petkova, V.T., 2003, An analysis of field feedback in consumer electronics industry, Technische Universiteit Eindhoven. [39] Hey, J., 2004, The data, information, knowledge, wisdom chain: The metaphorical link, Relato´rio te´cnico Intergovernmental Oceanographic Commission (UNESCO). [40] Young, B., Cutting-Decelle, A.F., et al, 2005, Sharing Manufacturing Information and Knowledge in Design Decision and Support, Advances in Integrated Design and Manufacturing in Mechanical Engineering, 173–185. [41] Garcia-Perez, A., Mitra, A., 2007, Tacit knowledge elicitation and measurement in research organisations: a methodological approach, Electronic Journal of Knowledge Management, 5/4: 373–386. [42] Markus, M.L., 2001, Toward a theory of knowledge reuse: types of knowledge reuse situations and factors in reuse success, Journal of Management Information Systems, 18/1: 57–93.