Product Knowledge Management: Role of the Synthesis of TRIZ and Ontology in R&D Process

Transcription

1 Product Knowledge Management: Role of the Synthesis of TRIZ and Ontology in R&D Process Hyman Duan, Quentin Xie, Yunmei Hong, Leonid Batchilo, Alp Lin IWINT, Inc. Abstract With the acceptance of Knowledge Management by industries and the mass application of PDM/PLM systems in enterprises in recent years, the issue of Product Knowledge Management (PKM) has drawn remarkable attention. However, current PDM/PLM systems are not able to support knowledge management activities in product lifecycle; and the existing knowledge management theory, such as SECI (Socialization-Externalization-Combination-Internalization) model, also has limitations when applied to R&D process. In this paper, we analyze the difficulties and drawbacks of applying SECI model in product lifecycle. We propose using both TRIZ and ontology to overcome the difficulties. We present a systematic inventive problem solving workflow using TRIZ and ontology-based search scheme, analyze how they interact with product knowledge creation, representation, organization and reuse. Finally we draw the conclusion that SECI model can be enhanced with synthesis of TRIZ and ontology in R&D process. Keywords: Product Knowledge Management, TRIZ, Ontology 1 Introduction Ikujiro Nonaka and Hirotaka Takeuchi [1] propose a SECI model of the knowledge creating process to understand the dynamic nature of knowledge creation, and to manage such a process effectively. They distinguished between two types of knowledge explicit and tacit. As shown in Figure 1, Nonaka and Takeuchi claim that the four conversion processes involving these two types of knowledge constitute the essence of knowledge creation: From tacit to tacit (i.e., socialization), From tacit to explicit (i.e., externalization), From explicit to tacit (i.e., internalization), and From explicit to explicit (i.e., combination). The creation of knowledge is a continuous process of dynamic interactions between tacit and explicit knowledge. Only by tapping into tacit knowledge can new and improved explicit knowledge be created. In turn, better explicit knowledge is essential for stimulating the development of new, higher level, tacit knowledge. The four modes of knowledge conversion interact in the spiral of knowledge creation. The spiral becomes larger in scale as it moves up through organizational levels, and can trigger new spirals of knowledge creation.

2 Figure 1. The SECI cycle of knowledge creation [1] SECI model recognizes the dynamic nature of knowledge and knowledge creation and provides a framework for management of the relevant processes. But it has some disadvantages. It is based on a study of Japanese organizations, which heavily rely on tacit knowledge. So SECI model do not pay more attention to the sharing of explicit knowledge. We think that SECI model is more about an idea of knowledge management than about an operable tool, especially for product knowledge. The knowledge of an organization comprises three types of knowledge, including customer knowledge, product knowledge and management knowledge. They decide, support and implement the enterprise strategy respectively. The main focus of customer and management knowledge is to locate knowledge from both within and beyond an organization, leverage, store and distribute that knowledge in an efficient way. All of these are major components of traditional knowledge management. Product knowledge is quite unique from its abstract nature and high intellect involved, which makes it hard to represent. How to create and utilize new knowledge based on existing ones becomes the main focus, which remains a dark area in traditional knowledge management system. With the acceptance of Knowledge Management by industries and the mass application of PDM/PLM systems in enterprises in recent years, the issue of Product Knowledge Management (PKM) has drawn remarkable attention. PKM emphasize the conscious analysis, optimization and organization of production information in PDM system in order to reuse them in new product development and then transfer them into product knowledge and enterprise intellectual assets. The core idea of PKM is to generate knowledge from information and to guide innovation with the knowledge. PKM is high level application of PDM and PLM. However, current PDM/PLM systems are not able to provide fully support to product knowledge management activities in product lifecycle.

3 Research [2] indicates that, in a typical organization, only 4% of organizational knowledge is available in a structured and reusable format and the rest is either unstructured or resides in peoples minds. The structured knowledge, although small in volume, has high value for companies because it can be accessed easily, mined and used for decision making. Generating structured knowledge, through transformation from tacit form into explicit form, is one of the critical steps of product knowledge management. The remainder of this paper is structured as follows: Section 2 reviews the knowledge creation role of TRIZ in R&D process; Section 3 discusses the product knowledge organization powered by ontology and presents the benefits of using TRIZ and ontology in problem solving process. In section 4, we propose using both TRIZ and ontology to overcome the above difficulties of product knowledge management. We analyze how they interact with product knowledge creation, representation, organization and reuse. Finally in section 5, we draw the conclusion that SECI model can be enhanced with synthesis of TRIZ and ontology in R&D process. 2. Knowledge creation role of TRIZ in R&D process Knowledge is the essence of innovation and competitive advantage in R&D process, and thus knowledge creates the platforms on which a company can expand into new product markets, enabling it to maintain or enhance its competitive advantages [2]. Therefore Effective tools and processes for knowledge creation and management are becoming a critical issue. TRIZ is recognized as the most effective methods and tools for engineering problem solving and innovative concept generation, and also is being aware of its value on knowledge creation. Figure 2. TRIZ concepts and tools

4 TRIZ (Figure 2) was designed for problem solving and provides a logical way of thinking for development of the technical systems through overcoming contradiction and towards the increase of ideality with utilizing resources. Figure 3. TRIZ problem solving process The TRIZ standard problem solving process is illustrated in Figure 3. From Specific Problem to Specific Solution, there are three main steps and six small steps: Abstraction (X1), Reformulation (X2), Transformation (Y1), Selection (Y2), Interpretation (Z1), and Consolidation (Z2). The output and intermediate results of these steps include function model, cause-effect problem reformulation model, Su-F problem model, contradiction definition, standard solutions, inventive principles, integrated concepts, etc. According to TRIZ training and application experience worldwide, the steps Z and X, especially the step Z, need more involvement and efforts of the problem solver and thus are very difficult to TRIZ beginners and normal engineers. We think that the problem solver full of TRIZ spirit is the master of all the three steps X, Y, and Z, thus the systematic problem solving process aided by TRIZ methods and tools is actually a knowledge creation process. All the inputs, outputs and intermediate results of above steps are high level (product and system level) and high value knowledge especially for conceptual design in R&D process. Managing and recording the thinking process and results of problem solvers and R&D engineers having TRIZ spirit and philosophy in their mind is becoming the key issue for product knowledge management in R&D team of a company. Guillermo etc. [3, 4] proposed to integrate TRIZ and Case-Based Reasoning and produce a system and database to apply SECI knowledge conversion cycle. But the limitation of TRIZ-CBR approach is its efficacy as also mentioned the authors. Furthermore, the authors did not disclose how to build the CBR database and how to apply TRIZ-CBR approach in R&D process. We propose to use ontology as the tool for organizing and retrieving the external and internal knowledge of an enterprise and ontology-powered patent-based technical solution

5 knowledgebase as a complementary tool aiding engineers thinking in step X and Z. 3. Knowledge organization role of ontology in R&D process Looking through the TRIZ problem solving process from personal thinking point of view, we find that it is a dialectically unity between divergent thinking and convergent thinking (Figure 4). That means, the divergent thinking process from root problem or TRIZ generalized problems to possible ideas or TRIZ generalized solutions should begin with the convergent point, the right problem; and the diverging degree (i.e. accessibility and understanding of the state of art and related domains/industries knowledge from functional point of view) in above process determines the innovation level of final solution. Figure 4. Divergent thinking and convergent thinking Therefore, we need to simulate the two key elements of divergent thinking and convergent thinking, association and analogy, in computer software, in order to manage and record the thinking process of engineers and supply the gap of accessing and organizing knowledge of TRIZ. Association and analogy of human thinking root in the relations among things in human brain. The method and tool describing universal relations among things in computer is the ontology. An ontology provides a system of fundamental concepts and relations among them in term of which we can describe all pieces of knowledge we are interested in. The system can help engineers build the relations among things in the problem space and the solution space of a specific problem. In general, TRIZ guides the engineers in macro-level along a convergent innovation direction comparing with traditional problem solving methods, such as trial and error, brainstorming. Meanwhile, ontology supports the engineers in micro-level with divergent heuristics in problem reformulation and concept generation. The combination of TRIZ and ontology and their interaction in problem solving process perfectly demonstrate the law of eternal development and universal correlation of the natural world, the basic principle of materialistic dialectics, and make the abstract philosophy principle understandable and operable in real life. Recently, applying ontology or ontology engineering to product knowledge management is drawing more attentions in PDM and knowledge engineering areas. Edmond Chan etc. [5] proposed a framework to promote the reusability of product data and information generated from the past product development. An alternative evaluation engine is integrated into a

6 PDM system to evaluate the usability of existing parts and products for the development of a new product. An ontology editor is utilized to enable the integration of the PDM system and the alternative engine. It is an effort to evaluate and reuse product data and information in detail design stage. Yoshinobu Kitamura etc. [6] proposed to apply ontological engineering to systematization of functional knowledge to avoid ad hoc modeling and obtain consistent functional models. We define a default SVPO format (Table 1) in the knowledgebase of our R&D innovation platform, Pro/Innovator [7], for external (patent) knowledge extraction and acquisition and enterprise internal knowledge organization. End users also can customize the SVPO format to satisfy their own knowledge content. knowledge fields problem situation solution description Table 1. SVPO format of technical solutions in Pro/Innovator SVPO format Attribute Attribute Solution Verb excessive Parameter absence of insufficient (S) (V) unstable (P) presence of undesirable uncontrollable solution title prior art problem analysis problem statement system composition working principles additional info related problem Object (O) Figure 5. A patent-based technical solution example (to purify water) in Pro/Innovator All the instances of default format (Figure 5) and customized formats are stored in one

7 knowledgebase powered by Pro/Innovator ontology. Pro/Innovator ontology is a hierarchically-arranged database of linguistic units which are called synsets. A synset is a set of synonymous terms, representing a certain concepts. Ontology defines a number of relations between the synsets. These relations are referred to as semantic relations of the following types: synonymy, hypernymy, hyponymy and association (Figure 6). Figure 6. Part of water ontology in Pro/Innovator With ontology-powered knowledge organization and retrieval, the TRIZ problem solving process (Figure 3) can be enhanced as Figure 7. Figure 7. TRIZ problem solving process enhanced with ontology-powered search

8 4. Interaction between TRIZ and ontology in R&D process In section 3, we describe the benefits of integrating TRIZ and ontology-powered knowledge organization and retrieval in innovative problem solving process. Here we analyze how they interact with product knowledge creation, representation, organization and reuse in R&D process. Figure 8. Product knowledge management with CAI platform in PLM environment The transformation and exchange of product knowledge with the help of CAI platforms (Pro/Innovator and CBT/NOVA [8]) in PLM environment is shown in Figure 8. The product knowledge transformation can be described according to SECI model: Socialization: The innovation training platform, CBT/NOVA, and R&D innovation platform, Pro/Innovator, provide major means of sharing product tacit knowledge between individuals and governing socialization in different processes during the product life cycle. Externalization: Knowledge engineering implementation team of an enterprise transform the product tacit knowledge embodied in R&D individual experts into explicit with SVPO format knowledge acquisition. Internalization: Knowledge engineering implementation team of an enterprise transform the explicit knowledge stored in internal PDM database and external patent database into tacit knowledge embedded in R&D team and embodied in R&D individual through ontology-powered knowledge organization. Combination: The achievement of R&D process and innovative problem solving process integration TRIZ and ontology can combine and utilize pieces of explicit knowledge in Pro/Innovator knowledgebase to generate new explicit knowledge and thus deliver it to the final product, physical asset of the enterprise.

9 5 Conclusions SECI model of knowledge management can be enhanced with synthesis of TRIZ and ontology in R&D process. Based on TRIZ and ontology, the R&D innovation platform, Pro/Innovator, achieves close integration between R&D process and product knowledge management process. References [1] Nonaka Ikujiro, Hirotaka Takeuchi. The Knowledge-Creating Company: How Japanese Companies Create the Dynamics of Innovation. Oxford: Oxford University Press [2] Philipp Ackermann, Dominik Eichelberg. Product Knowledge Management. Proceedings of International Conference on Economic, Technical and Organisational Aspects on Product Configuration Systems, Technical University of Denmark, Copenhagen, 2004 [3] Guillermo Cortes Robles, Stephane Negny, Jean Marc Le Lann. Knowledge Management and TRIZ: A Model for Knowledge Capitalization and Innovation. Proceedings of the 4th ETRIA TRIZ Future Conference 2004, Firenze University Press (Italy), [4] Guillermo Cortes Robles, Stephane Negny, Jean Marc Le Lann. Innovation and Knowledge Management: TRIZ at the Core of Knowledge Creation Process. Proceedings of the 5th ETRIA TRIZ Future Conference 2005, Press of University of Leoben (Austria), [5] Edmond Chan, Kai Ming Yu. A framework of ontology-enabled product knowledge management. International Journal of Product Development, 2007, 4(3/4): [6] Yoshinobu Kitamura, Riichiro Mizoguchi. Ontology-based systematization of functional knowledge. Journal of Engineering Design, 2004,15(4): [7] Hyman Duan, Alp Lin, Yanling Li, Leonid Batchilo, Huiyu Wang. CAI Tool Case Study of Reducing Vibration of Diesel Engine Lubricating Oil System. Proceedings of TRIZCON 2006, Altshuller Institute for TRIZ Studies [8] Duan Haibo, Lin Yue, Pesetsky Serge. Innovation Education Principles with Computer Based Training System. Proceedings of TRIZCON 2005, Altshuller Institute for TRIZ Studies