We consider two integer linear programming models for the one-dimensional cutting stock problem that include various difficulties appearing in practical real problems. Our primary goals are the minimization of the trim loss or the minimization of the number of master rolls needed to satisfy the orders. In particular, we study an approach based on the classical column-generation procedure by Gilmore and Gomory for solving the linear programming (LP) relaxations and an extra column-generation procedure before solving a final integer problem. For our computational tests we use some data sets from the paper industry and some others generated randomly.
Proceedings of the "2nd International Conference on Engineering Optimization (EngOpt 2010)" CR-ROM Paper nº 1105.