Robust inventory ship routing by column generation

Abstract

We consider a real integrated ship scheduling and inventory management problem. A fleet of ships transports a single product between production and consumption plants. The transporter has the responsibility for keeping the inventory level within its limits at all actual plants, and there should be no need to stop the production at any plants caused by missing transportation possibilities. Due to uncertainties in sailing time, we introduce soft inventory constraints and artificial penalty costs to the underlying model. The model is solved by a column generation approach. By introducing some model adjustments, the problem decomposes into a routing and scheduling subproblem for each ship and an inventory management subproblem for each port. The columns in the master problem represent ship schedules and port call sequences. © 2005 Springer Science+Business Media, Inc.