Beginning in the latter stages of the twentieth century, the forces of globalization that reshaped industrial supply chains were influenced by two primary factors: the very large differential in labor costs between developed and developing countries and the availability of high volume, inexpensive international freight transportation.

When it comes to moving finished goods and intermediate products between continents to consumers and businesses, the dominant mode of transportation is containerized ocean cargo. In fact, since the early 1990s, the use of container shipments worldwide has grown more than three times the rate of the global GDP. Despite the recent economic downturn that has significantly cooled international trade, it is likely that ocean container cargo will continue to be important for years to come.

Researchers within the Stewart School have been active in ocean cargo research throughout the past decade. In this article, we highlight two recent research efforts supported by both the National Science Foundation and the Singapore Economic Development Board.

Professor Ozlem Ergun and Richa Agarwal (PhD IE 2007) recently completed a set of work culminating in Agarwal's thesis, "Network Design and Alliance Formation for Liner Shipping." Today's large ocean carriers, like Maersk, Hapag Lloyd, and NOL, face challenging problems in service network design, the task of determining which routes or services to operate and at what frequencies and how to assign ships to these services. An important new challenge is that most transoceanic routes today are operated jointly by a number of different carriers working together within an alliance. Thus, the research also focuses on how to jointly plan the operations of multiple collaborative carriers, including how to allocate the capacity of different services among carriers.

To solve these problems, Agarwal and Ergun developed a new integrated optimization model to solve the ship scheduling problem and the cargo routing problem simultaneously -- the first such model to incorporate transshipment ports (where cargo is transferred from one service to another). Since the model is too large to be solved directly with commercial optimization solvers, the team developed various customized heuristic and exact solution approaches and demonstrated their applicability on realistically sized problems.

In the second part of the study, Agarwal and Ergun focused specifically on alliance formation among liner carriers, using techniques from optimization and algorithmic game theory to propose an approach to design a collaborative service network. To manage interactions among participating carriers, their approach determines appropriate prices for capacity exchange between carriers to induce participants to follow an optimal collaborative strategy.

Professor Alan Erera and Aykagan Ak (PhD IE 2008) have recently studied another important aspect of ocean container operations: scheduling problems at seaports. Ak's thesis, "Berth and Quay Crane Scheduling: Problems, Models, and Solution Methods," is an integrated study that investigates how to optimally plan ship loading and unloading operations at container seaports. Managing berthing space and the large dockside quay cranes that are used to service vessels is a critical operational challenge; efficiency gains here allow ports to maximize container throughput while meeting the service requirements of carriers.

Berth allocation is the problem of determining where to moor a sequence of arriving vessels over time, while quay crane allocation and scheduling problems focus on which quay cranes to assign to berthed vessels. At large modern seaports, it is possible to simultaneously berth multiple vessels of varying lengths along long linear berthing areas with a shared set of cranes, which can be moved from one vessel to another (without passing each other) at any time, thus leading to very complex scheduling problems.

Ak's work was among the first to consider joint planning of berth allocation and quay crane scheduling, and he developed very fast solution heuristics based on tabu search to quickly find near-optimal solutions to these problems. His methods are currently being used as part of a large research effort focused on risk mitigation at U.S. seaports to simulate how port operators might re-optimize operations if cranes or berth sections are damaged.

Professors Alan Erera and Ozlem Ergun prepared this article fo the Fall 2009 issue of Industrial and Systems Engineering, the alumni magazine for the Stewart School of ISyE.