Uma metodologia baseada no modelo SCOR® e em inferência fuzzy para apoiar a avaliação de desempenho de fornecedores

Abstract

Resumo: Na literatura acadêmica, tanto a seleção quanto a avaliação para o desenvolvimento de fornecedores vêm sendo abordadas como um problema de tomada de decisão no qual um conjunto de fornecedores é avaliado com base em múltiplos critérios de desempenho. Embora já existam centenas de metodologias quantitativas voltadasAbstract: In the academic literature, supplier selection and evaluation have been addressed as a decision-making problem in which a set of suppliers is assessed based on multiple criteria. Although there are hundreds of quantitative methodologies to support the supplier selection problem, supplier evaluation aiming at developing suppliers is little explored in the literature. Furthermore, most of the existing approaches have some limitations due to the use of inadequate techniques. Thus this study proposes a new methodology to support the assessment of supplier performance, developed from the combination of fuzzy inference systems with some performance indicators of the SCOR  model (Supply Chain Operations Reference). The proposed approach enables the evaluation of aspects related to the performance of operations and costs. The results of this evaluation are used to categorize the suppliers with similar performance and identify guidelines for the management of each supplier group. To demonstrate the modeling process and use, and also to evaluate the suitability of this proposal, a pilot application involving the evaluation of 10 suppliers of a company in the automotive sector was conducted. Four fuzzy inference systems were implemented using MATLAB  and parameterized according to the judgments of two employees of the company. A sensitivity analysis was performed to verify the consistency of the results yielded by these systems. The results support the suitability of the proposed methodology and the parameterization performed during implementation. When compared with the methodologies for the assessment of suppliers in the literature, this approach presents advantages such as the appointment of guidelines for the management of the supplier base; the possibility of integration with the supply chain performance evaluation; the ability to assess simultaneously a non-limited amount of suppliers, the representation and processing of information in a linguistic format, and the mapping and internal storage of all decision scenarios of the problem.

A Hesitant Fuzzy TOPSIS model to supplier performance evaluation

“…The results of the pilot application are somewhat similar to those found by the studies proposed by Lima Junior and Carpinetti [2], Osiro et al [4], and Lima Junior et al [21]. These studies also applied models based on fuzzy logic to assess suppliers of automotive companies and concluded that the vast majority of suppliers presented satisfactory performance, while a few suppliers required development programs and only one should be replaced.…”

“…When compared to other previous models to supplier performance evaluation, the proposed model has advantages such as: i. Unlike all the previous models found in literature [2][3][4][5]11,12,[17][18][19]21], it allows the decision makers to use linguistic expressions, or more than one linguistic term simultaneously, to evaluate the performance of a supplier in a given criterion; ii. Another advantage is dealt with group decision-making employing aggregation of the subjective judgments of several decision makers.…”

5-Item Likert Scale and Percentage Scale Correspondence with Implications for the Use of Models with (Fuzzy) Linguistic Variables

“…Usually, the models express linguistic terms via triangular (or other) fuzzy numbers, and Likert scales are assumed to be equidistant, see e.g. Carrasco et al (2012), Casola et al (2005), Holeček and Talašová (2010), Lima et al (2016), Lin and Yeh (2012), or Yan et al (2017). However, if individuals do not perceive Likert scales to be equidistant, this input assumption might lead to unwanted distortion of results.…”