Navegando por Autor "Melo, Luciana Vieira de"

Agora exibindo 1 - 1 de 1
  • Imagem de Miniatura
    Item
    Modelos de programação linear inteira para variantes do problema de programação de projetos com restrição de recursos
    (Universidade Federal de Goiás, 2018-11-23) Melo, Luciana Vieira de; Queiroz, Thiago Alves de; http://lattes.cnpq.br/8041183668335400; Queiroz, Thiago Alves de; Mundim, Leandro Resende; Bachega, Stella Jacyszyn
    The resource-constrained project scheduling problem has its importance both in the theoretical part, as in the field of operational research, and in practice, with project management in corporate environments and other applications. In this context, some integer linear programming models, solved with the help of an optimization library, were studied for the resource-constrained project scheduling problem. This problem aims at minimizing the makespan, namely, the total completion time of the project, given the scheduling of activities. In order to achieve this objective, a quantitative approach is used, and the research is classified as descriptive with regard to its objective, and bibliographical and experimental with regard to the technical procedures used. The first model has two types of decision variables, while in the second model there is only one type of variable. When considering the insertion of real constraints, in particular, the multi-skill, the multi-mode and time lags, the third, fourth and fifth models are obtained, respectively, from the second model with the addition of such constraints. The models are analyzed with regard to the runtime and the amount of instances solved in optimality. The results of the computational experiments indicate that the second model is a bit more competitive in comparison with the first one, since it was able to solve a larger number of instances, present solutions with a smaller gap and require less computational time. Therefore, the other models started from the second with the addition of practical constraints. The results of the computational experiments indicate that the models with practical constraints can have better performance (that is, related to the number of instances solved, gap value and computational time) when smaller instances are considered. Therefore, developing models that are capable of solving medium and large size instances is a challenge, but, it can bring great advantages for the corporate environment, helping managers in making decisions, reducing waste, improving costs and thus bringing personal well-being.