Heuristicas relax-and-fix e fix-and-optimize aplicadas ao problema de dimensionamento de lotes de múltiplos itens com restrições de capacidade e preservação de capacidade e preservação de preparação

Abstract

This work deals with applying relax-and-fix and fix-and-optimize heuristics to solve the Capacitated Lot-Sizing Problem with Setu Carryover. Such problem consists in determining the quantities to be produced of a set of items to satisfy deterministic dynamic demands in each period of a planning horizon without backlogs, in order to minimize total costs. Computational experiments consist of: (i) obtain solutions by the means of using five mathematical formulations from the literature with a Mixed Integer Programming solver; (ii) obtain solutions by relax-andfix followed by fix-and-optimize (RFFO) with all formulations; (iii) obtain solutions by a new approach that combines relax-and-fix with fix-and-optimize (RFO) also using all formulations. RFFO and RFO approaches are compared using a subset of the instances in which none of the five formulations were able to find an optimal solution whithin a given time limit of 600 seconds. Results show that the RFO approach were able to find better solutions than RFO in more than half of the tested instances.