Loading...
Publication
Integrated Approach to Data in Aquaponics Systems
| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 20.12 MB | Adobe PDF |
Authors
Abstract(s)
Este projeto consiste sobre o desenvolvimento de uma solução de Business Intelligence
aplicada ao contexto de um sistema de aquaponia numa instituição de ensino superior,
com o objetivo de facilitar a recolha, integração, análise e visualização de dados
operacionais e ambientais. A fragmentação e a falta de padronização dos dados, quer
dos ficheiros, convenções de nomenclatura ou estruturas das tabelas, dificultavam a
consolidação eficiente da informação e a tomada de decisão. O objetivo principal foi
desenvolver uma arquitetura baseada num Data Lakehouse para automatizar a recolha,
integração, análise e visualização desses dados, favorecendo a monitorização contínua
dos parâmetros e a partilha de resultados com a comunidade científica.
Para isso, concebeu-se uma arquitetura de dados baseada em paradigmas de Data
Lakehouse, que integra Apache Spark para processamento distribuído e Power BI para
criação de um modelo semântico e dashboards interativos. O trabalho inclui a conceção
de um modelo dimensional, a implementação de um pipeline de ETL (Extração, Transformação
e Carregamento) para limpeza e unificação de arquivos heterogéneos (com
formatos, convenções de nomenclatura e estruturas de tabelas inconsistentes) e o desenvolvimento
de relatórios visuais orientados ao desempenho do sistema aquapónico.
A principal contribuição consiste na demonstração do potencial das tecnologias
analíticas e de visualização de dados na gestão sustentável de sistemas de aquaponia,
mostrando como práticas de engenharia de dados aliadas a ferramentas de BI permitem
superar desafios de qualidade, volumetria e escalabilidade dos dados.
Por fim, a usabilidade da solução Power BI foi avaliada, recorrendo à aplicação de
um questionário, através da Escala de Usabilidade do Sistema, obtendo-se a classificação
de Aceitável. Destacaram-se como pontos fortes a clareza do conteúdo e a facilidade
de navegação, enquanto a estética do design e o desempenho foram apontados como
oportunidades de aperfeiçoamento em desenvolvimentos futuros.
This project focuses on the development of a Business Intelligence solution applied to the context of an aquaponics system within a higher education institution. Its objective is to facilitate the collection, integration, analysis, and visualization of operational and environmental data. The fragmentation and lack of standardization of data, including files, naming conventions, and table structures, hindered the efficient consolidation of information and the decision-making process. The main goal was to develop a Data Lakehouse, based architecture to automate the collection, integration, analysis, and visualization of this data, enabling continuous monitoring of key parameters and the sharing of results with the scientific community. To achieve this, a data architecture based on Data Lakehouse paradigms was designed, integrating Apache Spark for distributed processing and Power BI for the creation of a semantic model and interactive dashboards. The work includes the design of a dimensional model, the implementation of an ETL (Extract, Transform, Load) pipeline for cleaning and unifying heterogeneous files (with inconsistent formats, naming conventions, and table structures), and the development of visual reports focused on the performance of the aquaponic system. The main contribution lies in demonstrating the potential of analytical and data visualization technologies in the sustainable management of aquaponics systems, showing how data engineering practices combined with BI tools can overcome challenges related to data quality, volume, and scalability. Finally, the usability of the Power BI solution was evaluated using a questionnaire based on the System Usability Scale, resulting in a rating of Acceptable. Strengths included content clarity and ease of navigation, while areas for improvement identified in future developments were the visual aesthetics and performance.
This project focuses on the development of a Business Intelligence solution applied to the context of an aquaponics system within a higher education institution. Its objective is to facilitate the collection, integration, analysis, and visualization of operational and environmental data. The fragmentation and lack of standardization of data, including files, naming conventions, and table structures, hindered the efficient consolidation of information and the decision-making process. The main goal was to develop a Data Lakehouse, based architecture to automate the collection, integration, analysis, and visualization of this data, enabling continuous monitoring of key parameters and the sharing of results with the scientific community. To achieve this, a data architecture based on Data Lakehouse paradigms was designed, integrating Apache Spark for distributed processing and Power BI for the creation of a semantic model and interactive dashboards. The work includes the design of a dimensional model, the implementation of an ETL (Extract, Transform, Load) pipeline for cleaning and unifying heterogeneous files (with inconsistent formats, naming conventions, and table structures), and the development of visual reports focused on the performance of the aquaponic system. The main contribution lies in demonstrating the potential of analytical and data visualization technologies in the sustainable management of aquaponics systems, showing how data engineering practices combined with BI tools can overcome challenges related to data quality, volume, and scalability. Finally, the usability of the Power BI solution was evaluated using a questionnaire based on the System Usability Scale, resulting in a rating of Acceptable. Strengths included content clarity and ease of navigation, while areas for improvement identified in future developments were the visual aesthetics and performance.
Description
Keywords
Aquaponia Business intelligence Data lakehouse