Application of fuzzy TOPSIS in the Analyze phase of the DMAIC cycle to aid decision-making

Abstract

This paper reports the use of multicriteria analysis in the Analyze phase of the DMAIC (Define-Measure-Analyze-Improve-Control) cycle for continuous improvement. The research was carried out in a tissue paper factory located in southern Brazil. A sample of 64 parts of 16 different reels of recycled paper was used. A problem regarding paper quality variability was detected, presenting a scrap index ranging between 9% and 23%, compromising machine productivity and product sales. This motivated the implementation of a structured project supported by the application of the DMAIC cycle. The project team (machine operators, maintenance staff, supervisor, and data analyst) defined the evaluation criteria and determined the control intervals and their equivalence with linguistic variables to support the necessary evaluations for the application of fuzzy TOPSIS (Technique for Order of Preference by Similarity to Ideal Solution). The criteria were C1-Longitudinal Strength (Kgf), C2-Longitudinal Elongation (mm), C3-Cross-strength (kgf), C4-Cross-elongation (mm), C5-Weight (g/m²), and C6-Humidity (%). The results showed that samples with the same ranking had the lowest scrap indexes in the subsequent process. Also, the criterion C5 had a more significant impact on the quality of the product than the other criteria, which was determined from the DMAIC sequence. Improvements related to C5 should be prioritized. The fuzzy TOPSIS method presented is a flexible tool, adapting itself to the solution of the problem and contributing to the decision-making process.