Social annotation as a tool for dynamic capture process of cad design intent in the context of knowledge based engineering

  1. Alducin Quintero Mármol, Gerardo
Dirigida por:
  1. Manuel Roberto Contero González Director/a
  2. David Guerra Codirector/a

Universidad de defensa: Universitat Politècnica de València

Fecha de defensa: 09 de diciembre de 2011

Tribunal:
  1. José María Gomis Martí Presidente/a
  2. José Luis Saorín Pérez Secretario
  3. Pablo Bermell García Vocal
  4. Carlos Vila Pastor Vocal
  5. Jorge Martín Gutiérrez Vocal

Tipo: Tesis

Teseo: 318082 DIALNET lock_openTESEO editor

Resumen

This thesis focuses on the New Digital Product Development (NDPD) process area to contribute to increasing the productivity of CAD users by means of an improved design intent communication using a socialannotation technique. In this work this design intent communication is also considered as ¿tacit knowledge transference¿. Part of this type of knowledge is within CAD models that are not just only a 3D geometry representation. They also reflect the result of a specific modelling strategy, which usually constrains the future capability for modification and reuse of the existing 3D model. The literature review led us to investigate on the Web 2.0 technologies based on the principle of participation and collaboration of the users on the Internet. Previous research works show that the design teams interact like social networking on the Web. These technologies are transforming the product development process in latest years, contributing new methodologies specifically oriented to support decision-making processes during product design and shortening the development cycle. Product CAD modelling process must be agile and efficient to provide enough flexibility to adapt to a changing market. In general CAD tools let product designers experiment with many alternative solutions, providing better quality products in less time and at a lower cost. In this context, best practices of digital product modelling can be used to document those CAD modelling strategies that provide the best performance with respect to a specific objective. This objective usually is related to downstream applications, such as CAM and CAE systems or the ¿engineering change process¿, that exploit the 3D CAD models created previously. These CAD models can be characterized with a specific "data quality" depending of the next downstream user (e.g. molding, manufacturing, analysis, and so on). The quality of the virtual model is based on the CAD Modelling Strategy¿ used to build it. Traditional knowledge management systems, such as Knowledge Based System (KBS) / Knowledge Based Engineering (KBE) are not very well suited to capture the `tacit knowledge¿ that embeds the virtual models created with the CAD tools because the most of them have a very rigid structure, and are better suited to manage contexts where the knowledge is static. But organizational knowledge is generated in the daily work in dynamic learning processes. Then the following research questions were stated: Can we improve the traditional processes of knowledge capture/transfer? Can we define a dynamic process of knowledge capture/transfer in a complex process as Digital Product Development process? One exploratory study and three main experimental stages were conducted in order to assess the impact and utility of social annotations in the ¿engineering change process¿ in order to determine its influence in the user performance during this process. The experimental results obtained from experiences with Spanish and Mexican engineering students indicate that it is possible to reduce the time needed to perform ¿engineering changes¿ in existing models into a 15-25% range. This reduction is feasible if those CAD models provide annotations explaining the original design intent for who is modifying the model. The results show that this tagging process behaves in a similar way to tagging processes performed by Internet users (highly dynamic, very easy to use and learn). This means that social annotation/social tagging represents a feasible approach to support dynamic knowledge collecting/transferring from best CAD modelling practices. Thus, the extrapolation of these results means that social technology can be used as best practices collector in Digital New Product Development (DNPD) process in other phases of product life-cycle. Finally it is possible affirm that it is feasible to collect and to transfer engineering knowledge dynamically, and be considered as a complementary process to populate KBS / KBE systems through traditional method.