A METHOD FOR SOLVING THE PROCUREMENT OPTIMIZATION PROBLEM BASED ON INVERSE CALCULATIONS

Abstract

The paper describes a method for solving the procurement optimization problem based on inverse calculations. The method involves solving the unconstrained optimization problem and adjusting the obtained values of arguments subject to the constraint. Compared to conventional nonlinear optimization methods, the proposed method is easier to implement with computer software, since it does not require any determination of additional variables and multiple solutions to an unconstrained optimization problem. A solution to a two-constraint problem using inverse calculations has been considered. In this case, the optimization problem is transformed into a single-constraint problem. Therefore, when constructing a system of equations for determining increments of arguments, the transformed form of the objective function and the constraint are taken into account. This paper discusses a solution to a procurement portfolio development problem for a confectionery company with a limited budget and a target value of contribution margin. The obtained solution was compared with the solution produced by the mathematical software package and the penalty method. The result of this comparison is presented in the paper. The proposed algorithm can be used in systems intended to support the procurement decision making process. The algorithm can also be used to solve quadratic programming problems of the considered form in other fields of research.

Keywords

• procurement optimization
• inverse calculations
• inverse task
• quadratic programming