This article considers the lot-sizing problem in multi-stage productio
n settings with capacity-constrained resources. This problem deals wit
h the determination of a production plan for the end item and its comp
onents in order to meet the forecast demand in each period of a planni
ng horizon. The production plan should minimize the sum of production,
setup and inventory costs. A heuristic method build upon a formulatio
n of the problem in terms of echelon stock is developed. Computational
results are reported and the solutions' quality is evaluated through
Lagrangean lower bounds. (C) 1997 Elsevier Science Ltd.