Functional-based Approaches for Supporting Product Development

Autor: Montecchi, Tiziano
Rok vydání: 2013
Předmět:
Popis: The process of product development consists of a series of stages to transform an idea into a new product ready for the market. Starting from the earliest stages of identification of market opportunities, till the final stages of product manufacturing, information plays a crucial role. Every stage of product development is supported by many dedicated methods and tools. On the contrary, knowledge management strategies for supporting product development have not reached the same level of development. Often users rely on the text mining tools without a proper strategy, using indeed their own intuition and knowledge of field experts. The objective of this thesis is to propose a systematic searching approach based on the functional decomposition of the product. In the first place, the development of this method consisted of the study and selection of the main methodologies of design, problem solving and design rationale (such as FBS, TRIZ, EMS model, and others), and later a reworking of them has been done in order to transform them into research targets. Each stage of product development process requires strategies, therefore different research targets. At present, strategies to support problem reformulation, problem solving, construction of the state of the art and technological transfer have been tested. Since the effectiveness of a documentary search depends on the capability of text mining tools to obtain high recall, after testing several tools on the market, it was decided to develop and patent a special search engine that works with concepts exploiting libraries based design methods as knowledge bases. The thesis is organized as follows: In Chapter 2, we deal with defining the information to be searched. This information is described in the form of functions and they are identified through research targets, i.e. those key concepts that have to be found in documents with appropriate research strategies. These research targets are created using the main design methodologies properly selected. Once these targets are known, different research strategies have been defined according to the goal we want to achieve (decision making and problem solving). Chapter 3. With the aim of supporting the activities of decision making, it has been developed a strategy to provide the proper knowledge to the designer in order to understand whether and how a specific product (or technology) can be innovated. For this purpose, the design methodologies have been used for inventing/creating the research targets to find all the possible alternative products. Subsequently, a patent search uses these targets to understand which of these systems are already present at the state of the art, because they are already patented in our field and which systems are not yet patented, representing an opportunity for development (white space opportunities). In addition, using a further functional search other technological areas are investigated to understand if those systems, not yet present in our field, have already been developed in other technological fields to reach our same goal. This information, properly organized, can be used as support for technological transfer. Chapter 4. As a support to the problem-solving activities, we have been developed a strategy for problem formulation according to the model of TRIZ contradiction, in other terms as a conflict between two functional requirements. In particular, we propose a formulation of the problem based on a process of retrieving and structuring the proper knowledge in order to facilitate the resolution of the problem itself. Chapter 5. In order to test the effectiveness of these research strategies it was necessary to develop a new search engine that searches for concepts, rather than by keywords. In fact, the concept-based search can highly increase recall of the information retrieval process if compared with the more traditional research based on a combination of keywords selected by experts. It has been developed and patented a semantic search engine, called KOM, which integrates the basic text mining techniques such as tagging and parsing analysis. In particular, the search engine uses the patent literature as a source of information from which we can gain the knowledge. The proposed research strategies have been applied to several industrial cases in the field of mechanical engineering, household appliances, biomedical and pharmaceutical industries. In the work of this thesis two explicative examples are presented about: decision making strategies applied to the technology of sterilization of contact lenses, and problem solving strategy applied to the case of the nutcracker. The application using the industrial case studies have allowed to verify the effectiveness of the conceived strategies and tools, showing significant results, especially related to the improvement of recall of information retrieval. It is under definition an exhaustive evaluation (of these strategies) that involves the technical staff of the companies that have provided the industrial case studies. For what concerns the future developments, due to the development of KOM search engine in the last year of my PhD, it will be possible to conduct a validation of the proposed strategies with a more representative sample. In addition, KOM will reduce the time to develop and test new research strategies for supporting other activities of product development.
Databáze: OpenAIRE
Externí odkaz: