Parallel Lagrangean approximation procedure | Journal of the Operational Research Society | DeepDyve