A novel version of the TODIM method based on the exponential model of prospect theory: The ExpTODIM method

Autor(es)

Leoneti, Alexandre Bevilacqua ^[1] ; Autran Monteiro Gomes, Luiz Flavio ^[2]

Afiliação do(s) autor(es)

^[1] Univ Sao Paulo, Sch Econ, Business Adm & Accounting, Ribeirao Preto - Brazil

^[2] Ibmec Univ Ctr, Sch Econ & Business, Rio De Janeiro - Brazil

• Tipo de documento: Artigo Científico
• Fonte: European Journal of Operational Research; v. 295, n. 3, p. 1042-1055, DEC 16 2021.
• Citações Web of Science: 1

Resumo

The adherence of the TODIM method and its variations, including two new versions of TODIM with the use of the exponential and logarithmic functions, to prospect theory was compared based on performance indicators with a very frequently used MCDM method, namely TOPSIS and its variation, Behavioral TOPSIS. It was hypothesized that the use of methods with mathematical models more adherent to prospect theory would provide more accurate predictions of individual decision making. A hundred students from the University of Sao Paulo in Ribeirao Preto were invited to participate in a field study where three different cases should be solved without the support of any method. Then, performance indicators were used to evaluate the prediction capacity of the methods with the ones provided by the volunteers, having the TODIM method with the new mathematical function presented in this paper, hereafter named Exponential TODIM (ExpTODIM), reached the best scores in all three performance indicators. The main contribution of the present paper is the proposal of a novel version of TODIM method that fits among the low implementation complexity multicriteria methods with high predictive power, since it is based on a value function that is more adherent to prospect theory. (c) 2021 Elsevier B.V. All rights reserved. (AU)

• Processo FAPESP: 16/03722-5 - Desenvolvimento estrutural para a utilização da Teoria dos Jogos na solução de problemas multicritério de tomada de decisão em grupo nas organizações
• Beneficiário: Alexandre Bevilacqua Leoneti
• Modalidade de apoio: Auxílio à Pesquisa - Regular