Kerem Akartunali, Postdoctoral Research Fellow, The University of Melbourne

Despite significant amounts of research, big bucket lot-sizing problems remain notoriously difficult to solve. We consider a two-period model, which is the simplest single-machine, multi-period, capacitated submodel. We propose a methodology that would approximate the closure of the convex hull of this submodel by generating violated inequalities using a distance function. We discuss the polyhedral characteristics of this submodel, as well as how to apply them to the original problem, and we conclude with detailed computational results.

Joint work with Andrew J. Miller.