Solving the integrated planning and scheduling problemusing variable neighborhood search based algorithms

Abstract

In this paper, we address the Integrated Planning and Scheduling Problem (IPSP) on parallel and identical machines. Planning and scheduling are essential for the efficient management of supply chains. Although both pursue the same general objective, they are usually performed independently mostly because they relate to different timescales. As a consequence, the generated plans and schedules are typically sub-optimal from a global standpoint.

The approaches followed in this paper explicitly consider the interdependence between the planning and scheduling activities by solving them simultaneously in an integrated way. We explore different heuristics based on variable neighborhood search procedures with new and specifically designed neighborhood structures relying on the properties of the IPSP. The quality of these approaches is evaluated through extensive computational experiments performed on a large set of benchmark instances. The results show that the proposed methods achieve high-quality solutions, with a substantially low computation time, outperforming other state-of-the-art results reported in the literature.