From data to value: a model to manage experimental data collected from durability tests within the product development process
Coupled with the pressure from competition, regulations, and the overall need for improvements, organizations started to recognize their data as a business asset. Although companies have more data than ever at their disposal, actually deriving meaningful insights and value from them is easier said t...
| Autor principal: | Larocca, Arthur Eduardo de Freitas |
|---|---|
| Formato: | Dissertação |
| Idioma: | Inglês |
| Publicado em: |
Universidade Tecnológica Federal do Paraná
2021
|
| Assuntos: | |
| Acesso em linha: |
http://repositorio.utfpr.edu.br/jspui/handle/1/25836 |
| Tags: |
Adicionar Tag
Sem tags, seja o primeiro a adicionar uma tag!
|
| Resumo: |
Coupled with the pressure from competition, regulations, and the overall need for improvements, organizations started to recognize their data as a business asset. Although companies have more data than ever at their disposal, actually deriving meaningful insights and value from them is easier said than done. Within the context of the product development process (PDP) and collaborative engineering, the management of experimental data has been identified as a research gap. While many studies have tackled data and collaboration based on computer-aided technologies (Cax) environments, the management of the measured data has been observed as a scarce topic in the literature. Moreover, among all verifications and validations performed during development, further analyses in a multinational manufacturing company indicated a higher prevalence of durability tests performed in prototypes. In this concern, the main objective of this work is to propose a model to manage measurement files collected in durability tests which would enable and foster their reuse across projects and throughout the product development organization. For this purpose, this qualitative applied research tackles this challenge based on the methodological framework of the Design Science Research Methodology (DSRM). Built upon prior literature and considering the constraints that hinder data reuse, a logical data model was developed and documented as a class diagram in the Unified Modeling Language (UML). This artifact comprises a set of approaches that have been designed to enable the complete lifecycle of the most recurrent data in durability testing: load, strain, acceleration, and displacement. The main strategy of the proposed model is the enhancement of metadata into the measurement files, só important information, that is usually unstructured within test reports, is stamped and structured into the test signals. Several concepts were developed based on this strategy, such as the concept of data versions and the novel approach for the documentation of the instrumentation positioning as cartesian coordinates. A proofof-concept has been demonstrated using real measurement files and other related information to key academic and industrial stakeholders, who also took part in the development and evaluation of the solution. Feedback and satisfaction surveys suggested that the proposed model is complete and faithful to real-world phenomena. They also suggested that the proposed artifact is better than the current model (or way of working) regarding its internal consistency, level of detail, and robustness. The relevance of this work lies in its applied nature and potential contributions towards a more collaborative and efficient product development process regarding the verification and validation of new products. |
|---|