PERBANDINGAN METODE SEPARABLE PROGRAMMING DAN QUADRATIC PROGRAMMING DALAM PEMECAHAN MASALAH PEMROGRAMAN NONLINIER

Abstract

The Separable programming method solves nonlinear programming problems by transforming a nonlinear shape that consists of a single variable into a linear function and resolved by the simplex method. Meanwhile, the quadratic programming method accomplishes the two degrees nonlinear model by transforming the nonlinear shape into linear function with the Kuhn Tucker Conditions and resolved by the simplex Wolfe method. Both of these methods are applied to the Markowitz’s portfolio model, which is to find the proportion of stock funds to obtain maximum profits by combination of three shares, such as BMRI, GGRM, and ICBP. The completion using the quadratic programming method is more effective and efficient with the same optimum value.