(434b) Decision Support for the Routing, Scheduling, & Bunkering of Multi-Parcel Chemical Tankers

For many years, the US$2.2 trillion global chemical industry has been a key driver of the global economic growth. The global chemical trade, which hit more than US$1.24 trillion in 2006, has achieved an impressive 14% average annualized growth between 2000 and 2006 (World Trade Organization, 2007). Maritime logistics is the workhorse of the growing global chemical trade. To support this trade, new parcel tankers that routinely ship chemical cargos worldwide have been built in record numbers. A parcel tanker differs from other maritime bulk carriers by the presence of tanks or compartments that can carry multiple liquid cargos simultaneously. Parcel tankers are capital-intensive and their daily operating costs can exceed ten thousands of dollars per ship. In a notoriously cyclical industry, efficient cargo assignment and routing and scheduling of parcel tankers are crucial to the financial success of a tanker company.

Fuel expenses claim up to 90% of the daily operating cost of a multi-parcel tanker, and a tanker's fuel consumption varies with the cube of its speed. This combined with high fuel prices makes it crucial that the tanker owners manage tanker speed and fuel purchases in a prudent manner. The volatility in fuel prices and their significant variations across refueling ports make this a difficult task, and systematic optimization can aid the decision-making process. Tanker owners naturally seek low-cost refueling options to reduce their total operating expenses. They may deviate slightly from their normal voyage plans, incur necessary port dues, or even delay the transit through a canal to refuel at a port with attractively priced fuel. Alternatively, they may resort to lowering voyage speeds, since the fuel consumption rate is proportional to the cube of cruising speed. Surprisingly, no literature model addresses this issue for multi-parcel tankers.

Management of cargos, routes, bunkering, and schedules of a fleet of ships is an inherently complex combinatorial problem. Its complexity is further accentuated in the tanker business primarily due to the need to make decisions a few weeks or months in advance and safety regulations related to cargo stowage. In this work, we address two issues. First, we extend a previous static MILP formulation (Jetlund & Karimi, 2004) for the routing and scheduling of multi-parcel tankers to address the dynamic scenario in which cargos arrive at arbitrary intervals and must be assigned dynamically to ships and routing and scheduling must be revised to accommodate the changing scenrios. Second, we propose a novel mixed-integer linear programming model (Oh & Karimi, 2010) that optimizes the operation of a multi-parcel tanker in the presence of uncertain fuel prices. Given the route and cargo details of a tanker, it determines the optimal voyage speeds and a refueling plan that minimizes the expected total operating cost. We use a case study derived from industrial data to illustrate the application of our new models. We also briefly describe a decision support system that addresses the practical needs of ship owners in managing multi-parcel chemical tankers.

References

Jetlund A, Karimi I A. Improving the logistics of multi-compartment chemical tankers. Comput. Chem. Eng. 2004, 28, 1267.

Oh H C, Karimi, I A, Operation Planning of Multi-Parcel Tankers under Fuel Price Uncertainty, Ind Eng Chem Res. 2010, in press.