2SM: método de triagem multicritério linguístico para aplicação no processo de seleção de materiais

Abstract

Due to the existence of many materials, combined with complex relationships between the parameters involved, material selection has been developed by different methods and approaches. Despite this diversity, the screening and sorting steps are common to virtually all of these methods and approaches. In the screening stage, criteria are defined for filtering and eliminating unsuitable materials for the application. Only those with high potential for use are enabled, reducing the pool of candidate materials for the classification stage. A screening method must be able to evaluate a large number of materials quickly and accurately. This work aims to develop a linguistic multicriteria method (2SM) for the screening step of the material selection process, to make the material selection process more agile and accurate. The methodology used in the development of the dissertation consisted of the following steps: a systematic literature review on screening methods in the process of material selection; elaboration of a database with alternative materials and their respective properties; creation of a procedure to group similar materials into classes for each property; a method was established to associate the classes with linguistic 2-tuple variables and to generate a linguistic scale for each property of the materials; the linguistic multicriteria screening method was applied to solve material selection problems in the literature. The results obtained by applying the 2SM method in three case studies resulted in a total of 82 candidate materials; of these, 19 correspond to the materials screened by the selected works for a tool holder, a bicycle frame, and a flywheel, of which, three were classified among the first five positions when they went through the classification step together with the candidate materials screened by the 2SM method, thus corresponding to 12% of representativeness for the first five positions. In contrast, the 2SM method resulted in 22 materials occupying the first five positions, corresponding to 88% representativeness—this way, evidencing its effectiveness. In addition, it was possible to verify that the proposed method presented speed in processing and greater accuracy for the screening of candidate materials compared to the screening methods used in the case studies analyzed.