n this paper a further improvement of the methodology presented by Iannone et al. (2010) is proposed. In particular, the procedure of lot sizes calculation has been made more time efficient by means of the introduction of a specific exit condition able to stop more quickly the algorithm when there is the certainty to having found the optimal solution. In this way one of the main limits evidenced during the past experimentations has been surmounted.
In addition, some useful indications about the regulation parameters of the algorithm are furnished. Although there were not significant differences in the goodness of the proposed solution, which remains generally high, the configuration with logarithmic variation and high resolution presents a remarkable criticality in presence of strict space constraint. This condition, together with a more precise evaluation of the convenience domains for the parameters, will be object of further studies.
In conclusion, this work represents a step forward in the definition and the tuning of a robust methodology able to assure good solutions without excessive complexity of implementation to a classical stock management problem.