Enhancing Decision Support Systems development with the EUL Framework applied in the Agricultural Sustainability Assessment domain

• Ano de defesa: 2023
• Autor(a) principal: Menezes, Wanner Martins de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Biblioteca Digitais de Teses e Dissertações da USP
• Palavras-chave em Português: Automação; Automation; Avaliação de interface do usuário; Decision support system; Domain specific language; Framework; Geração de interface do usuário; Linguagem específica de domínio; Ontologia; Ontology; Sistema de apoio à decisão; User interface Evaluation; User interface generation
• Link de acesso: https://www.teses.usp.br/teses/disponiveis/55/55134/tde-04012024-165946/

Decision Support Systems (DSSs) play a vital role in organizing and processing data and information to facilitate decision-making in specific domains within organizations. The development of DSSs traditionally requires close collaboration between software developers and domain experts. However, maintaining a dedicated development team can be financially burdensome for many projects, resulting in slow and costly development and maintenance processes that hinder the agile development of DSS. To address this challenge, our group created the EUL framework, which proposes the utilization of ontologies and scripts based on Domain-Specific Languages (DSL) to automatically generate web-based DSS. The ontologies serve to describe the knowledge domain, while the scripts define the operations to be executed on user-supplied domain data. Developers and experts are only jointly concerned with these 2 artifacts. From them, the EUL framework automates the generation of the DSS, including its User Interface (UI). This research analyses the quality of DSS generated using the EUL framework through its application in the agricultural domain, the SustenAgro DSS. SustenAgro creates sustainability assessments of sugarcane agricultural production systems located in south-central Brazil. In this work, the SustenAgro DSS was evaluated to validate its usability. This evaluation was conducted with different techniques. First, the Think Aloud method was used with four end users who were tasked with performing a set of predefined tasks. Additionally, these users provided feedback on system usability through the System Usability Scale (SUS). The results of the usability tests indicate that the generated UI for the DSS is well-structured, self-explanatory, and easy to use, demonstrating satisfactory functionality. Second, a qualitative evaluation was performed with a sustainability domain expert. He was able to modify the parameters of the DSS and generate new versions. The findings highlight the successful application of the framework in the agricultural domain and validate its usability through user and expert evaluations. The automatic generation of DSS, using ontologies and DSL scripts, offers the potential for efficient and cost-effective development, enabling greater agility in the deployment and maintenance of DSS across various domains.