An overview of manufacturing knowledge sharing in the product development process

Theoretical definitions of knowledge

Knowledge management as a subject of study and practice arose as a consequence of the knowledge economy. The value of the organisation lies not only within the product and service that it provides but also within its knowledge asset. ¹ This has become increasingly important during the birth of the digital age and the Internet.^2,3 The definition of knowledge depends on how it is categorised and can consequently affect the way in which it can be used. ⁴ A typical definition given by Liu and Young ⁵ stated: Data is text or numbers. Information is data with added context to explain the data. Knowledge is the interpretation of information in order to assign meaning. In this definition, knowledge can be organised into blocks which can be coded, shared and used. Consequently, this definition has originated from and is mostly used for managing knowledge by information systems.

Polanyi ⁶ classified Knowledge into tacit and explicit. Explicit knowledge can be articulated and represented in formal languages (codified) and can be shared between different systems without losing its integrity. Tacit knowledge is rooted in people’s personal experience and beliefs, and is more difficult to express, codify and therefore share. Polanyi’s concept was adopted in Nonaka’s research, in which knowledge was seen as residing in and originating from individual employees, which together forms the collective knowledge of the organisation. ⁷ Explicit knowledge may be captured and represented in a database, while the data itself needs to be interpreted and understood within the context of the organisation (tacit knowledge) so that it may be effectively used.

In a more pragmatic approach, knowledge was defined in terms of its source and its application, and is categorised as being declarative, procedural or causal. ⁸ Declarative knowledge (or know-what) is the content of knowledge. Procedural knowledge (or know-how) refers to the processes of using knowledge. Causal knowledge (or know-why) refers to the underlying recognition of where it is appropriate to apply knowledge. Hicks et al. ⁹ pointed out that other components such as behavioural aspects and organisational learning are missing from the data–information–knowledge hierarchy. They presented an alternative definition, referred to as Explicit Islands in a Tacit Sea, in which the sea of tacit knowledge enables the creation of data, information and explicit knowledge and the selection of the tools for best practice.

Bohn ¹⁰ defined technological knowledge as the knowledge associated in realising products and services, and recognised its tacit and explicit elements. A third dimension, that is, knowledge maturity, was introduced. A scale of knowledge maturity was proposed in which knowledge types are used to describe the maturity of a process. Process maturity is defined as the ability to which its attributes can be codified and standardised. Bohn took a pragmatic view of industry, claiming that any organisation will have a mix of processes at varying levels of knowledge maturity, which will in turn affect learning, problem solving, operation and job roles. Mountney and Gao ¹¹ also recognised the need to incorporate knowledge about the development of new manufacturing processes which were categorised as immature (because their capabilities cannot be codified) and was referred to as innovative manufacturing knowledge in their research.

The shared context for knowledge management

As described above, knowledge has been interpreted in several ways, and some commonalities exist between various definitions, for example, all definitions indicated a codifiable perspective to knowledge where it is easy to articulate in a shared context – a common emerging theme of sharing information. Alavi and Leidner ¹ stressed that a knowledge management system must be able to capture individuals’ knowledge yet assign meaning which is relevant for specific applications. The need for the shared context governs the appropriate knowledge type. They supported the use of both tacit and explicit knowledge as they are mutually dependent and reinforce each other.

Zack ⁸ stressed that tacit knowledge should be made explicit because it is only then that knowledge can be a competitive advantage. Fahey and Prusak ¹² saw tacit knowledge as being fundamental to the creation and use of explicit knowledge. Tuomi ¹³ proposed ways in which tacit and explicit knowledge could be exploited for different purposes. The strength in explicit knowledge would be in articulating and codifying routine operations in a shared context.

What emerges from previous research is how the definition of knowledge governs the way in which it can be represented and managed. As described by McMahon et al., ¹⁴ two main approaches have been adopted in knowledge management: The first is concerned with codifiable knowledge and the development of a mainly information system perspective in supporting this. The second is a more social perspective evident, or a community approach, where knowledge is not so easy to articulate and a shared context needs to be established in addition to the knowledge.

Saunders et al. ¹⁵ reported the findings of an exploratory industrial investigation to evaluate the challenges encountered through a case study within a multinational automotive manufacturing company. They commented that there remains a continued focus on further reducing the number of unique new product development programmes, yet simultaneously increasing the number of worldwide product launches and maximising global presence, especially in the emerging growth markets. This has been facilitated by reducing the overall development and test costs by maximising on shared common global programmes, that is, global vehicle platforms, coupled with the realignment of the respective worldwide manufacturing and product development organisations. The authors believed that the socio-technical dimension within the auto-industry has become increasingly more complex and unmanageable. Therefore, globally dispersed engineering teams that rely on the close integration of a network of specialist external suppliers, across multiple time zones, to leverage design and manufacturing expertise will benefit from further research in this domain.

Giddaluru et al. ¹⁶ carried out research on establishing a shared product model for the integration of product lifecycle applications. They proposed a modular product structure methodology and demonstrated its advantages in the implementation of product lifecycle systems in the collaborative company. The proposed methodology gives guarantee for manageability, integrity, consistency, security and traceability of product data in the whole lifecycle, and various dynamic data information can be shared in the product evolution chain. A mapping process of product data between computer-aided design (CAD), product lifecycle management (PLM) and enterprise resource planning (ERP) systems was developed which gives seamless flow of information between them. ¹⁷ Because multinational manufacturing enterprises are evolved for many years and there is a large amount of previous product development and manufacturing information, they are also developing tools for the migration of legacy data to the new structure using a PLM system.

Knowledge management issues in product development

The above theories of knowledge are applicable to the product development process. Hicks et al. ¹⁸ referred to the product development process as an information transformation process, in which the decisions necessary to move a design from one state to the next are driven by knowledge and information. The need for explicit knowledge and the importance of representing it in an information framework was explored. Information was interpreted as being either formal or informal. Formal information is structured and can be represented. Informal information is less structured and usually obtained from discussions. Knowledge is inferred from information, the preference being towards informal information. Because of its less structured and personalised nature, informal knowledge is seen as being possibly unreliable due to bias. An example of the use of the declarative–procedural–causal approach to knowledge can be seen in research by Fu and coworkers ¹⁹ who investigated the knowledge content required for design in the aerospace industry. The results of their research were high-level and knowledge was categorised into market, human, technology and procedural.

Previous research demonstrated that in product development, knowledge content and form are inextricably linked. McMahon et al. ¹⁴ demonstrated how the knowledge form can influence the type of support system used. They discussed the personalisation (tacit) and codification (explicit) of knowledge, linking this with the community and information system approaches to knowledge management. The product type itself also demonstrates a major influence on the selection of the approaches. The information system approach, which promotes the capture, use and reuse of knowledge, is advantageous for standardised, mass-produced products, while the community approach is suited to customised products. However, the need for both approaches is acknowledged in the product development process.

Other researchers have also considered the tacit and explicit components of knowledge, for example, when analysing the product development of automotive products. ²⁰ The tacit nature of manufacturing, in particular, was also highlighted by Grant and Gregory ²¹ because it requires a wide range of knowledge from different people. Wong and Radcliffe ²² investigated tacit and explicit knowledge used in the design process of small hydraulic cylinders from a knowledge retention perspective, developing a knowledge schema to identify and segregate explicit and tacit knowledge. The aim of the schema is to understand the relationship of various knowledge types in an engineering design context. Most of the knowledge identified in the design activities is explicit. The tacit element is concerned with knowing what theory to use and when to apply it.

Fei et al. ²³ reported that design changes and their impacts are difficult to capture and evaluate, which make product design in uncertainty. They argued that design change propagation is caused by deign conflicts arising from the initial and consequent changes which are difficult to predict without knowing their preceding change solutions, since all the following changes are based on the solutions of preceding changes. Many design conflicts arising from change analysis can be tackled by reusing well-formalised and managed knowledge abstracted from previous design cases. They proposed a methodology for design change management to implement these ideas using modelling method, matrix analytical method and knowledge-based engineering. ²⁴ A system engineering modelling method was used to capture dependency relationships in and between functions and components. A matrix-based analytical method helps trace change propagations and identify design conflicts and behavioural and spatial connections between components. With the help of a knowledge repository and the reasoning method, designers are able to find proper solutions for design conflicts occurring in design change propagation.

A further investigation was carried out by Saunders et al. ⁴ with the collaborating multinational automotive manufacturer, and their study confirmed that there is a distinct lack of integration between the numerous standalone web-based databases and information and communication technology (ICT) tools utilised within the company. The study also highlighted the inadequacy in the overall knowledge management infrastructure to effectively store new knowledge created during new vehicle development programmes and make it widely accessible to support ongoing product development later in the vehicle lifecycle during full-scale high-volume manufacture. The incorporation of the extended enterprise architecture, as the critical organisational factor missing in previous frameworks, will support the development of a more suitably optimised web-based knowledge management system. It is proposed that this will better serve the community of geographically dispersed engineering teams on global automotive systems engineering projects throughout the complete vehicle system lifecycle.

Manufacturing knowledge issues in product development

Manufacturing knowledge was defined as knowledge about manufacturing processes, assembly, quality, materials handling and operation planning by Pahl et al. ²⁵ Design engineers were encouraged to reduce risks as far as possible and also to ‘design for manufacturing’, and the manufacturing impact was seen primarily as a cost impact. ²⁶ The Design for Manufacture and Assembly (DFMA) methodology was developed by Boothroyd et al. ²⁷ to enable greater consideration of manufacturing and assembly requirements before the detailed design stage. The aim was to design products which are easier to manufacture and assemble and can consequently be produced at a lower cost. Process also needs to be considered in conjunction with the materials to be selected. ²⁸

A practical application of the DFM methodology is seen in O’Driscoll. ²⁹ With this application, a number of checkpoints were built into the existing design process to ensure that the required knowledge had been supplied.

There were three stages of increasing detail, including definition, development and validation/scale-up. Initial process list and process capabilities were considered in the definition stage. Chen ³⁰ also acknowledged the influence of process, material and shape and looked at manufacturing feasibility and cost. For the earlier stages of product development, this was seen as a ‘screening out’ process for unsuitable processes. A conceptual framework for a manufacturing decision support system was developed, which was in essence about information interaction between designers and manufacturing engineers and its components.

In terms of content, manufacturing knowledge tends to be defined as method, capability and cost. The link between material, shape and process is strongly defined. The desired situation is that the manufacturing process is mature and stable in that it has a known capability that matches the product attributes. However, there is a recognition that such knowledge may be less detailed earlier in the design process, and this has not been defined specifically. ¹¹ Design knowledge and manufacturing knowledge have been noted as separated specialist domains with a need to share knowledge between them. As discussed earlier, the themes of knowledge sharing and the need to support this within a knowledge management environment are apparent. Engineering knowledge is viewed as being, or needing to be, explicit. However, the support through tacit knowledge is also essential. ³¹ The knowledge type and appropriate support systems should also be linked to the nature of product development and manufacturing processes.

Although new manufacturing processes have been acknowledged, the inherent nature of previous methodologies was to select capable processes to reduce risk. Innovative processes that are still in development and are therefore immature have not been considered by most previous researchers. Bohn discussed how processes are not inherently capable, particularly in high-tech industries. Further development work is required to develop or validate such processes to a capable standard. This stance was echoed by Grant and Gregory ²¹ and Mountney et al. ³² in their study of the global transfer of manufacturing processes, noting that manufacturing processes will change over time according to their maturity.

Another important issue is how to manage manufacturing knowledge in the supply chain during collaborative product development. To produce products at the targeted cost, time and quality, the supply chain must be aligned with product development processes and contribute to key decision making. This will allow manufacturing firms to overcome problems such as (partially) failed product launches due to the lack of timely provision of parts and systems caused by insufficient capacities in the supply chain. ³³ With integrated product development and supply chain management, enterprises have the benefit of increased supply chain capability, thus increasing the effectiveness of new product introduction and improving their overall performance. They have tried to link the global product development process of a multinational automotive manufacturer to its global network of suppliers ³⁴ at some selected key integration points through an integrated framework, and provided guidelines to identifying where critical decisions are made in collaboration with the supply chain.

In the manufacturing lifecycle of aerospace products, large amounts of data are generated throughout the activities involved. Much of these data contain useful knowledge for engineers to improve various aspects of design for manufacturing (DFM) of new or existing products that may help reduce defects in future manufacturing activities. An investigation was carried out by El Souri ³⁵ at a large UK-based manufacturer of aerospace products that aim to introduce an improved approach that supports design engineering activities to make use of the ‘data’ generated to drive the number of reoccurring defects down. They collected raw data captured from defects on complex aerospace products, both manufactured in-house and by third party suppliers, and applied a structured approach, in order to evaluate its effectiveness towards the DFM knowledge that can be used to reduce reoccurring defects. ³⁶ The evaluation has been analysed which has shown that knowledge management can have a significant effect on DFM practice when the collected raw data from defects and associated processes are structured, represented and linked to an itemised bill of material (BOM) in a design engineering environment, and the main focus of their work was the capture of manufacturing knowledge related to defects of parts from suppliers.

Information system approach to manufacturing knowledge sharing

The information system approach is concerned with the capture, representation and reuse of knowledge, particularly applicable to situations where there is some degree of standardisation in processes. Successful knowledge sharing is at the heart of this approach from two perspectives. The main perspective is interoperability. In order for knowledge to be shared and successfully interpreted across different platforms, it must be defined and represented in a standardised manner. This requires the definition of knowledge exchange standards, protocols and languages, such as STEP, CORBA, UML and ontologies. The second perspective is concerned with the techniques used to define knowledge in a specific domain and interpret this knowledge from the point of view of other domains. Two techniques that are of particular interest are features and information models, as both have been used extensively to represent design and manufacturing engineering.

Including social perspectives in manufacturing knowledge sharing

The business management sector of research has been critical of the information system approach to knowledge management. Information systems do not consider the social perspectives and the communication between people. ⁷¹ The explicit and tacit elements of knowledge are divorced. Walsham suggested that the information system approach misunderstands and misrepresents tacit knowledge. It is seen as something which can be converted into explicit knowledge and then stored, that is, as a transferable object. Consequently, explicit knowledge has been promoted at the expense of instilling the meaning from tacit knowledge and such systems have ignored the need to manage knowledge simultaneously at both the explicit and tacit levels. Walsham suggested that such systems should be designed to promote the value of the sense-reading and sense-giving activities which would give them value, and suggested two methods – communities of practice within a shared boundary and organisational translators across domain boundaries.

Roberts ⁷² also examined the issue of knowledge transfer through the use of ICTs and issues surrounding explicit and tacit knowledge transfer. Knowledge transfer is defined as the diffusion of knowledge from an individual to others. The knowledge transfer process is achieved by socialisation, education and learning which can occur either deliberately or as a by-product of another activity. Roberts noted that person-to-person (and preferably face-to-face) communication is the most critical element of technology transfer and suggested that even video conferencing is insufficient for tacit knowledge transfer and trust building as the process of digitisation codifies the human element of the contact. Furthermore, initial tacit knowledge transfer is an essential prerequisite for explicit knowledge transfer.

Johannessen et al. ⁷³ noted that increasing investment in IT does not appear to have reaped the benefits, and were also concerned about the focus on explicit knowledge being to the detriment of tacit knowledge transfer. They were specifically concerned with the influence of tacit knowledge on the use of IT systems and vice versa. They saw the organisational challenge as being to transform personal tacit knowledge into organisational explicit knowledge. Processes for building trust and relationships are seen as being pivotal to this. Swan et al. ⁷⁴ also acknowledged that IT-based knowledge management projects which focus primarily on codifying knowledge will not be able to deal with innovative (immature) knowledge. They view innovation as a time-phased multi-function communication process, and therefore knowledge sharing is essential for this process and the development of new innovative approaches. They also share the stance that tacit knowledge cannot be codified because of its personal and context-specific elements. Consequently, networking, a social communication process, was promoted as a major requirement of a successful process-based approach to innovation.

Mountney et al. ³² investigated the nature of manufacturing knowledge required for preliminary design and how this knowledge can be shared effectively between the specialist design and manufacturing domains. Their research was specifically concerned with immature innovative manufacturing knowledge for the preliminary design, when a lot of information about the product is still incomplete. A methodology was developed which supports and consolidates the themes of tacit and explicit knowledge, codification and maturity of knowledge, and with an emphasis on the social knowledge-sharing aspects and requirements needed in an industrial setting. Their methodology includes a process maturity audit to assess the manufacturing processes and to establish the level of knowledge (completeness of information) required including process feasibility, capability and estimated cost.

Some researchers whose prime focus is the information system approach have attempted to recognise and address tacit knowledge requirements. Cheung et al. ⁷⁵ acknowledged the necessity of tacit knowledge for knowledge sharing and seek to codify employee tacit knowledge and experience as part of ‘know-how’ in their latest development of an Internet-based product data management (PDM) system for collaborative product development. Koh and Gunasekaran ⁷⁶ reported work on managing uncertainty in manufacturing operation planning and scheduling was another demonstration of attempts to convert tacit knowledge to explicit knowledge.

Piorkowski et al. ⁷⁷ investigated the importance of capturing knowledge during face-to-face meetings in the new product development process in the high-value manufacturing sector. They described the concept of dynamic knowledge management (KM) which is a combination of cultural and technological factors, including the cultural factors of people and their motivations, technological factors of content and infrastructure and, where these both come together, interface factors. In their paper, ⁷⁸ a dynamic knowledge management framework and tools was described for capturing and indexing face-to-face meetings. It was reported how the framework was discussed during a meeting of the collaborating company project stakeholders. Participants agreed that the framework would have most benefit at the start of the product lifecycle before key decisions were made.

Evans et al. ⁷⁹ reported a 5-year participant observation study at a largest aerospace and defence organisation, and proposed a virtual knowledge-sharing framework in dispersed aerospace product development based on emergent social software platforms. As knowledge and social communications become increasingly important in modern engineering enterprises, emerging and fast-developing social media technologies have shown significant potential in complementing traditional engineering information technologies for managing more dynamic and unstructured engineering information and knowledge. They demonstrated a new paradigm for collaboration and knowledge sharing during the product development process through the utilisation of emergent social software platforms including Web 2.0 technologies which have been increasingly and widely used for social communications and connectivity. ⁸⁰ Compared with traditional engineering information systems, which are often highly complex and require specialised knowledge to maintain and operate, the new paradigm provides much simpler usability and less formal communication methods for employees.

The advent of social media in the last decade has changed the way many employees and global organisations communicate and interact with one another, providing a medium for information, especially tacit knowledge, to be transferred across the World Wide Web in a less formal (unstructured) manner. Zammit et al. ³¹ explored whether storytelling and video sharing functionality, embedded into a corporate social media site, were capable of facilitating the capture and sharing of employee knowledge during the product development cycle. They developed a knowledge capture and sharing framework which is directly driven by the knowledge users, addressing the special nature and application context of integrated design and testing operations. ⁷² The use of social media, video sharing and storytelling technologies is to capture complex engineering knowledge by the knowledge experts themselves, rather than by media professionals whom are paid to develop content.