A Genetic Algorithm for Linear Programming With Fuzzy Constraints

摘要: We examine a linear programming problem formulation in which the constraint coefficients are not precisely given work. investigate possibility of applying GAs to solve this kind fuzzy without defining membership functions for numbers, using extension principle, interval arithmetic, and α-cut operations computations, penalty method violations. The proposed approach simulates every number by distributing it into certain partition points. then used evolve values each point. As result, final represent grade that number. After calculating estimated uncertain coefficient, we obtain defuzzified problem. crisp can be solved following GA stage. empirical results show very good solutions within bounds thereby accomplishing flexible programming.