The aim of this work is to apply the algebra of fuzzy numbers through the \(\alpha \) levels of fuzzy numbers in Linear Programming Problems (LPP) with parameters and variables being real numbers, thereby expanding the possibilities in the search for the optimal solution. The uncertainty is introduced in each of the coordinates of the initial interior point through a symmetric fuzzy number. In particular, triangular fuzzy numbers are used due to the simplicity of algebraic operations. The modified algorithm is based on the Primal Affine Scaling Algorithm which describes the operations for the execution of a method related to interior points. In the Primal Affine Scaling Algorithm (PASA), only one point is obtained for each iteration in the direction of the optimal solution. In contrast, the modified algorithm, Fuzzy Primal Affine Scaling Algorithm (FPASA), considered ten points. These points are determined by the defuzzification of the \(\alpha \) -level of triangular fuzzy numbers. Numerical simulations were performed for three LPP. The results obtained by FPASA are better than those obtained by PASA in each iteration and require fewer iterations to reach the optimal value compared to PASA, considering the absolute error.
