A concept c is implied under a concept set C if c can be inferred by other concepts in C. Based on a domain ontology, the concept c is mapped to an ontology node k that has n child ontology nodes ki (i=1 … n). The concept c is an implied concept if each ki’s corresponding concept is either in concept set C or can be implied by other concepts in concept set C.
4.5 Case illustration and analysis
This section uses a base knowledge flow of a mobile phone company, named Smart-Tech Company, to illustrate the application of the KFV model and conduct
36
preliminary analysis. The base KF represents the knowledge-needs that a project team requires when conducting a mobile phone development process in the company. According to the process, KF designers consult domain experts and team participants to acquire important knowledge-needs and identify corresponding knowledge concepts for the purpose of representing knowledge-needs in a base KF.
In this company, the mobile phone development team requires participants from various departments. Those team members have different task functions: (1) the project manager controls and coordinates the project, (2) the marketing analyzer conducts the business analysis, (3) the designer is responsible for product design, (4) the salesperson focuses on product commercialization, (5) the inspector carries out the quality assurance tasks, and (6) the sourcing planner oversees the management of outsourced parts. Based on different knowledge-needs, KF designers can design virtual KF for individual participants.
The following discussion pertains to the sourcing planners at this company whose task function is parts outsourcing. First, KF designers make the concealing criteria for the sourcing planners, as required by the information security policy of the company and in consideration of the information granularity suggested by domain experts. Hence, the Essential Knowledge Nodes are identified based on the concealing criteria and all knowledge concepts in the Essential Knowledge Nodes should be included in an Essential Concept Set. Then, a virtual knowledge node is obtained by the order-preserving approach to ensure that the ordering in the base knowledge flow is retained. Finally, the KF designers abstract the knowledge concepts in the Essential Concept Set using the domain ontology and the minimum generalization policy.
The knowledge flow in Figure 12 includes nine knowledge nodes, k0 to k8, where each knowledge node contains multiple knowledge concepts. The knowledge concepts of k2 are too specific for the sourcing planners, so the KF designers make a concealing criterion
<sourcing planner, k2, Y> to meet their knowledge-needs. Another concealing criterion
<sourcing planner, k4, Y> is also made because the knowledge concepts of k4 are confidential for the sourcing planners. Following the two concealing criteria, the KF
37
designers select two knowledge nodes, k2 and k4, as the essential knowledge nodes for the sourcing planners.
Figure 12. Base KF for sourcing planners (nodes k2 and k4 are essential KNs).
The KF designers apply the procedure in Figure 9 to obtain a virtual knowledge node.
Initially, the neighboring knowledge node set NNS = {k1, k3, k7} and ENS (essential knowledge node set) = ES (expanding knowledge node set) = {k2, k4}. Knowledge node k3 is added into ES since the ordering of k2 is higher than the ordering of k3 (i.e. k2 > k3), but satisfy the ordering preservation rule. Therefore, the execution stops. The minimum expanding knowledge node set MES includes knowledge nodes {k2, k3, k4}, and a virtual knowledge node vk1 is derived as shown in Figure 13.
38 7. Service level agreement of 3rdparties 8. Design rule of RF and Baseband 1. Portability
Figure 13. A virtual KN, vk1, obtained after applying order-preserving approach.
After discovering the minimum expanding knowledge node set MES, the KF designer uses the procedure in Figure 11 and the ontology in Figure 14 to derive the knowledge concepts of vk1 based on the minimum generalization policy.
Figure 14. A partial domain ontology of mobile phone development.
Initially, the essential concept set ECS equals to {consumer behavior, display options, battery options, card options, design rule of RF and Baseband}. After performing the Generalization function on ECS, the KF designer obtains the abstracted knowledge
39
5. Service level agreement of 3rd parties 6. IC components knowledge concepts hardware design alternatives, price/performance of parts and service level agreement of 3rd parties are added in AKC, since knowledge node k3 is in MES but not in ENS. Hence, AKC = {consumer analysis, mechanical parts, IC components, hardware design alternatives, price/performance of parts, service level agreement of 3rd parties}. The knowledge concepts of vk1 can thus be obtained, as shown in Figure 15.
Figure 15. Knowledge concepts of vk1 after applying the minimum generalization policy.
Figure 15 show that vk1 has a redundant knowledge concept, hardware design alternatives, that can be inferred by mechanical parts and IC components. Figure 16 shows the result after removing the concept hardware design alternatives. The final knowledge concepts of vk1 are: consumer analysis, mechanical parts, IC components, price/performance of parts and service level agreement of 3rd parties.
40
5. Service level agreement of 3rd parties 1. Portability
Figure 16. Knowledge concepts of vk1 after removing implied knowledge concept.
The example given demonstrates the virtual knowledge flow from the sourcing planners’ perspective. If the sourcing planners are not satisfied with the view that has been generated, KF designers can repeat the same steps to abstract a new virtual knowledge flow by identifying other knowledge nodes as essential knowledge nodes, and then generating new MES to form another virtual knowledge node. Similarly, it is possible to create other virtual knowledge flows from other members’ perspectives. Hence, the proposed KFV model can enhance conventional knowledge flow models by supporting different team members with various knowledge-needs. Finally, every team member can obtain a proper virtual knowledge flow to support his/her knowledge-needs in the collaborative knowledge support platform.
To test the practical implications of this study, a preliminary analysis was conducted.
Several professionals were invited to examine the case and related concepts to investigate whether the theoretical model could benefit them. Overall, there was general agreement regarding the feasibility of the KFV model and its practical value. They thought that the KFV model could enhance typical knowledge flows to serve all team participants with their various knowledge-needs. For example, an interviewee mentioned that he would be able to communicate with a hardware designer more efficiently because the ontology of the theoretical model provided a common understanding of the general knowledge of hardware design. By referring to their different knowledge-flow views, both participants would be able to better understand their different knowledge-needs. Such understanding has the
41
potential to improve the quality of communication and increase the efficiency of knowledge sharing.
The results of the preliminary analysis are summarized as follows: (1) the visualized knowledge flows and knowledge-flow views help team members to easily formulate their knowledge-needs and quickly obtain consensus under common domain ontology. Thus, the quality of their communication and decision making is improved; (2) knowledge-flow designers can produce concealing criteria to protect confidential knowledge from unauthorized access and solve the information overload problem by abstracting detailed knowledge; and (3) in organizations, knowledge-flow views extend the efficiency of knowledge flows and improve the effectiveness of knowledge sharing and knowledge support.
4.6 Discussion
This chapter investigates the shortage of knowledge support in collaborative teams.
The workers in a team usually have different knowledge-needs according to their task functions. For example, the mobile phone development process involves six task functions.
The workers in these task functions need to access different knowledge concepts at different conceptual levels to conduct their work and communicate with each other.
However, conventional knowledge flow models do not provide different views of a knowledge flow that are required to address individual needs. The proposed KFV model meets this and related challenges. According to the proposed model, KF designers select some base knowledge nodes from a base knowledge flow to generate virtual knowledge nodes that conceal confidential or detailed information. Through an order-preserving approach and a knowledge concept generalization mechanism, a virtual knowledge flow is generated. The proposed innovative model allows various virtual knowledge flows to be generated that meet the individual knowledge-needs of different workers. These virtual knowledge flows not only comply with organizational information security policy but also reflect the granularity of knowledge-needs. Thus, this study advances the conceptual applicability of knowledge flow research to cooperative knowledge support environments.
42
Practical implications can be derived from the KFV model, including knowledge support facilitation in cooperative teams and team productivity and communication quality improvement. Moreover, the KFV model can be applied to any knowledge-based organization where business processes are conducted by cooperative teams in a dynamic working environment.
43
Chapter 5 Role-based knowledge-flow view model
According to the knowledge-flow view (KFV) model described in Chapter 4, knowledge-flow designers (KF designers) identify essential knowledge nodes based on concealing criteria and abstract base knowledge nodes (base KN) to virtual knowledge node (virtual KN) to build virtual knowledge flows (virtual KF) for the purpose of facilitating cooperative knowledge support. Except considering concealing criteria, tasks are often assigned to dedicated roles based on the characteristics of tasks to ensure quality and security. Hence, workers have different knowledge-needs in terms of roles. This leads to develop a role-based knowledge-flow view (r-KFV) model to illustrate role-based knowledge-needs properly in teamwork environments. This chapter extends the KFV model to the r-KFV model by introducing a role-based framework and role-based knowledge flow abstraction methods.
The r-KFV model analyzes the levels of knowledge required by workers based on their roles, and develops role-based knowledge flow abstraction methods that generate virtual knowledge nodes to provide the appropriate level of knowledge for each role. The purpose of the r-KFV model is to derive role-based virtual knowledge flows (VKFs) from a base knowledge flow (BKF). In the model, virtual knowledge nodes (virtual KNs) are generated from base knowledge nodes (base KNs) based on the relevance degrees between roles and base KNs. In addition, a concept abstraction method is developed to abstract knowledge concepts of base KNs for a virtual KN.
5.1 Concepts of role-based virtual knowledge flows
This section introduces a role-based framework in the r-KFV model and presents the key concepts of role-based virtual knowledge flows.
A role-based virtual knowledge flow comprises a set of virtual KNs that are aggregated from base KNs according to their relevance to a role. Some base KNs may be more relevant to a role than others. In other words, a role may refer to some knowledge nodes frequently but refer to others rarely due to its responsibilities and authorities. The
44
relevance degrees of a role to base knowledge nodes are distinct and indicate how important the base knowledge nodes to the role are. A virtual KN of a role-based virtual KF denotes a meaningful knowledge unit of interest to the role; thus, it should be relevant to the role. The process of identifying a role-based virtual KN involves aggregating base KNs based on their relevance to the role.
Once the relevance of the aggregated knowledge nodes to the role reaches a certain threshold, a virtual knowledge node is identified for the role. The relevance of a knowledge node to a particular role can be specified by KF designers or derived from the relevance degrees between the role and the operations associated with the knowledge nodes. Based on the relevance degrees of all base knowledge nodes, procedures can be clearly defined to generate appropriate role-based virtual KFs for different organizational roles.
After a virtual KN has been identified, KF designers can derive its knowledge concepts. The objective is to obtain abstractions of the knowledge concepts in the virtual KN’s member knowledge nodes that do not conform to the knowledge required by the role.
The role-based knowledge requirements are defined through domain ontology, which is a hierarchy comprised of knowledge concepts. The lower levels of the hierarchy contain specific knowledge, while the upper levels contain knowledge that is more general. The various roles in a team have different knowledge requirements. Participants need specific knowledge about their own roles and tasks, but needs less specific knowledge about other roles’ tasks. For example, Workers with a researching role design and develop products, so they must have specific technical skills and knowledge. Worker with a marketing role, who launch and promote products, may not have specific technical knowledge about the products, but they must have general knowledge about the technical aspects. The granularity of knowledge such as general or specific can be represented by knowledge concept levels in domain ontology. Consequently, building virtual knowledge flows from a roles perspective and considering the granularity of knowledge are necessary to illustrate knowledge-needs properly in teamwork environments.