Quality data mart: development of dashboards with meaningful data about the quality of products and processes

• Autor(a) principal: Mendes, Ana Rita da Silva
• Data de Publicação: 2020
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10773/29932
• topic: Quality data mart; Dashboard; Data visualization; Process control; DMAIC; CRISP-DM

With the emerge of powerful technologies, it is easier nowadays to store and analyse data. Companies can gain a competitive advantage by using the data available to gain valuable information about their production. The project took place in the quality department of an electronic manufacturing plant. It brings together the areas of monitoring products’ characteristics, process control and data analysis and mining. The challenge encountered was that the needed data is not available for use in a meaningful, easy and fast way to give to the stakeholders the necessary insights about the quality of the production. The goal of the project was to develop two quality data marts and to propose a standard procedure for their creation. Therefore, to develop the quality data mart a methodology based on CRISP-DM was used. Simultaneously, to study the challenge presented, the DMAIC phases were applied. The study started by defining the project scope, followed by a characterization of the initial situation. Then the areas of quality, data and dashboard execution were analysed. For each dimension, solutions were designed and implemented. Lastly, a review of the project was done. As a result, the foundations for the development of quality data marts are set, with a defined process, support materials and a pilot project. The creation of dashboards with near-real-time data enables the stakeholders to use it in their decision-making process. Therefore, the results of this project allow stakeholders to make better and quicker decisions and contribute to the development of a data-driven mindset in the company.