Diagnóstico de Rentabilidade Empresarial: uma análise fuzzy com dados dos indicadores da margem operacional da DRE

Abstract

This dissertation develops and validates a Mamdani fuzzy inference model to diagnose business profitability based on the operating margin from the 2024 Income Statement of TecnoAmazon Ltda. Four input variables (General and Administrative Expenses, Tax Expenses, Financial Expenses, and Financial Revenue) were fuzzified into Low, Medium, and High using trapezoidal membership functions calibrated on historical quartiles. An 81-rule IF–THEN base, built using the Wang-Mendel method and validated by experts, fed the Mamdani core. Centroid defuzzification generated crisp scores in the 0–30 range. The model achieved a mean absolute error below 8% compared to accounting ROE and a Kendall’s W of 0.85 with analysts’ internal rankings, confirming that fuzzy logic captures performance nuances overlooked by crisp metrics. Limitations include single-year scope, subjective parameter tuning, and lack of macroeconomic variables. Future work should expand historical data, test type-2 fuzzy sets, and integrate ESG indicators