Çok aileli dinamik parti büyüklüğü belirleme problemleri ve çözüm önerileri

Abstract

ABSTRACT In this dissertation a Lagrangean relaxation algorithm for Multi-Family Capacitated Dynamic Lot Sizing Problem with Coordinated Replenishments (MFCDLC) was developed. Furthermore, the dissertation includes the surveys about the dynamic lot sizing problems and the solution algorithms developed for them. As MFCDLC has the capacity constraints which have nonlinear structure, there are some difficulties in solving this problem. A Lagrangean problem, therefore, was created by dualizing the capacity constraints which are the side constraints. Using the subgradient optimization the Lagrangean multipliers were updated and the Lagrangean dual was solved. The Lagrangean relaxation algorithm was tested over randomly generated 10 problems. Computational experience shows that the foregoing algorithm is very efficient and provides near-optimal solutions to the problem.