Banca de DEFESA: MOACIR FERNANDO MORAIS GALDINO DE LIMA

Uma banca de DEFESA de MESTRADO foi cadastrada pelo programa.
STUDENT : MOACIR FERNANDO MORAIS GALDINO DE LIMA
DATE: 23/02/2022
TIME: 14:00
LOCAL: Online
TITLE:
KEY WORDS:

Multicriteria Decision Making. FITradeoff Method. Supplier Selection. Sorting Problematic. Elicitation of profiles. Sensitivity Analysis.


PAGES: 64
BIG AREA: Engenharias
AREA: Engenharia de Produção
SUMMARY:

In this work, the decision model for sorting problems that implements the FITradeoff method is improved with the introduction of two new steps: the elicitation of the classification problem profiles, and the sensitivity analysis of the final result of the model. The profile elicitation step fills a gap in the model, which requires the direct and unreferenced supply of values that define the classes into which the alternatives will be grouped. This elicitation consists of providing values based on two hypothetical alternatives, the ideal solution and NADIR, which constitute the best and worst possible solution for the decision problem. In addition, the elicitation can be done according to a ratio scale or an interval scale, with the decision maker being able to choose which scale to use. The second proposed improvement consists of evaluating the robustness of the obtained sorting, analyzing how sensitive the result is in relation to variations in the input data through Monte Carlo simulations. The model can be analyzed for robustness in relation to the alternatives’ performances or to the profile values obtained through the profile elicitation step. The result of this step is the robustness index associated with the sorting of each alternative, defined as the proportion of scenarios in which the sorting would not change. The model resulting from these improvements is applied to a supplier sorting problem in order to demonstrate the applicability of the model.


BANKING MEMBERS:
Externo à Instituição - EDUARDO DE FREITAS ROCHA LOURES
Presidente - 3091524 - EDUARDA ASFORA FREJ
Interna - 3184359 - LUCIA REIS PEIXOTO ROSELLI
Notícia cadastrada em: 21/02/2022 15:13
SIGAA | Superintendência de Tecnologia da Informação (STI-UFPE) - (81) 2126-7777 | Copyright © 2006-2024 - UFRN - sigaa03.ufpe.br.sigaa03