Proposta de agregação robusta de múltiplos métodos com incertezas em problemas de tomada de decisão multicritério

• Ano de defesa: 2018
• Autor(a) principal: Marcos Antonio Alves
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: Análise de cenários; Métodos de tomada de decisão multicritério; Agregação robusta; Incertezas; Lógica Fuzzy; Processo decisório; Engenharia elétrica; Incerteza; Lógica difusa
• Link de acesso: http://hdl.handle.net/1843/BUOS-B33H5E

Makingdecisionsisadiculttaskthatconsistsingatheringtheavailable information about the problem and choosing the best alternative in terms of expected results. Many uncertainties are involved in this process, being either ontological or epistemic. The latter, reducible, is often treated in multicriteria decision-making problems with the creation of scenarios and the use of dierent methods with classical or fuzzy logic. However, dierent ordinations can be provided to the decision maker with dierent rankings for each alternative. This dissertation proposes a robust aggregation model of multiple multicriteria decision making methods considering uncertainties. In order to evaluate the proposal, a practical problem of energy generation planning considering hydrothermal dispatch was used. The solution to this problem involved the combination of specialized genetic algorithm of Chu-Beasley and linear programming. Dierent scenarios were created, from the most pessimistic to the most optimistic, varying the parameters of hydrology and energy demand. Four study cases were performed with dierent scenarios and multiple multicriteria methods with classical and fuzzy logic. The robust solutions found indicated a small increase in costs when compared to those of lower cost in each scenario. However, they minimized the eects of uncertainties in the decision-making process, as vagueness and changes in the future scenario. This proposal may help decision-makers to make complex decisions more assertively and minimizing errors