Kajian Metode Penalti pada Optimisasi Linear dengan Kendala Non-Linear

Abstract

This research aims to examine the penalty methods used in non-linear optimization. The main objective is to identify the advantages and disadvantages of each penalty method, as well as determine the specific context in which each method is most effective. This research belongs to the literature research type. The study is conducted on prepared cases to obtain the optimal result of each penalty method. Penalty methods and barrier methods are interrelated in constrained optimization. The penalty method is used in the early stage to quickly generate an initial solution, while the barrier method is applied in the late stage to improve and tighten the solution. The barrier method is better at approaching and meeting the constraint boundary with higher precision, so the initial solution of the penalty method can be optimized and refined by the barrier method to achieve more optimal and precise results.