This paper deals with the capacitated lot-sizing problem with multiple items, setup time and unrelated parallel machines. The aim of the article is to propose an efficient population-based metaheuristic to tackle this problem. In the proposed approach, a dynamic constructive heuristic is proposed for generating a set of initial solutions. A path-relinking strategy between solutions in the current population is integrated in the diversification procedure, allowing us to explore the solution space further. The population is then intensified using a local search procedure which uses an operator based on machine capacity relaxation coupled with a repair procedure. The proposed approach is evaluated from the instances introduced in Toledo, F. M. B., and V. A. Armentano (2006.“A Lagrangian-Based Heuristic for the Capacitated Lot-Sizing Problem in Parallel Machines.” European Journal of Operational Research 175 (2): 1070–1083). The results of the experiment show that the proposed approach yields competitive results in terms of solution quality for most of the instances.