Proposta de um modelo hesitant fuzzy linguistic TOPSIS com distribuições de possibilidades para avaliação de riscos de fornecedores

Abstract

Supply risk management has gained ground in the business environment, as it helps companies survive in a scenario of globalization and outsourcing. Factors such as delivery problems, quality, economic scenario, geopolitical issues and adverse events generate uncertainties that can affect the supply of supply chains (Supply Chains – Scs). In this context, supply risk assessment began to encompass more and more attributes and is addressed in the literature as support for decision-making support techniques. The method called Hesitant Fuzzy Linguistic TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) with distributions of possibilities (PDHFLTSTOPSIS) has potential to be explored to support risk analysis, as it provides group decision support in situations of uncertainty and hesitation, based on the judgment of experts. Given this, this study proposes a decision-making model for assessing supply risks, based on PDHFLTS-TOPSIS. To this end, bibliographical research was carried out in order to verify the state of the art in relation to decision-making models subsequent to risk assessment. The procedures of the proposed decision-making model and the results of applying this model in a public organization that has a humanitarian aid chain are presented. During the pilot application of the model, the criteria of import instability and delivery time (risk-intensifying factors) and contract stability and logistics availability (risk-reducing factors) respectively presented the highest weights. In the classification stage, a threshold value of 0.50 was initially adopted, but by adopting a value of 0.60 it was possible to better differentiate suppliers and highlight potential for improvement in supplier management by the contracting company. At the end of the study, sensitivity analysis tests were carried out to verify the effect of varying the weights of the criteria on the results of the risk analysis. This study has the possibility of contributing to the literature on supply risk analysis by using a new approach for defining criteria weights and for evaluating and classifying suppliers. The proposed classification matrix is flexible in allowing the adoption of the segmentation threshold value according to the company’s requirements. The results provided by the model can help public and private companies, without restrictions on the size of the organization, to segment their suppliers based on supply risk factors, providing support for the implementation of risk mitigation strategies.