Uma Simulação multiagente para alocação de tarefas em projetos de software baseada no truck factor

• Ano de defesa: 2022
• Autor(a) principal: Vieira Neto Segundo, Caetano
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Palavras-chave em Português: Sistemas multiagentes; Algoritmos genéticos; Software - Desenvolvimento
• Link de acesso: http://www.repositorio.ufc.br/handle/riufc/70817

In the development of software, a project several steps must be carried out for it to be completed. Task allocation is one such step. Over the years, techniques have been researched to optimize this process to minimize cost and development time. However, it is believed that the Truck Factor (TF) is a metric that can determine the risk to a project and should also be taken into account when distributing tasks among team members. The TF concerns the distribution of knowledge about the project among the members of the development team, thus ensuring that there is no concentration of knowledge by only one part of the team, this theme is extremely relevant since, with globalization and demand from the IT market, the rotation of its professionals has become common. It is known that resource allocation in software teams does not have only an exact solution, since it is an Np-hard problem and, from that, over the years Search-Based Software Engineering (SBSE) techniques, which have a diversity optimization algorithms, such as genetic algorithms, have been applied in several types of research to solve this problem. In multi-agent environments, it is understood that there is a certain amount of agents that have perception and communicate to achieve their goals. Research in the area of multi-agents has used simulated environments to validate their research since the modeling and simulation must consider the main variables of a real environment. Therefore, this work proposes to build a multi-agent simulation, which uses SBSE techniques, to minimize the impacts caused by Truck Factor in a software development team. In the simulations, we model different configurations for the software development teams. Our results show that the proposed approach minimizes the impacts caused by task allocation in software development teams when we consider the TF metric during task assignment.