Due to the fairly recent deregulation of the sea cargo industry, carriers now engage in private contracts with shippers as opposed to using the same contract for all shippers. The length of a contract is typically six months to a year. There is an opportunity, therefore, for carriers to act strategically about several of their activities including pricing and equipment/capacity management.

Our goal is to develop a strategic planning tool that uses as input: the current voyages, current equipment quantities and locations, current capacities, rental prices, and estimated demand, and determines as output: the commodities which they should focus on, the prices that they should offer over the year to shippers, the quantity of containers that should be rented and the amount of capacity that needs to be reserved on other carriers.

This research will also deal with the issue of capacity allocation. Most containerized cargo is transported under contract. Many carriers and shippers enter into these contracts once per year, and the contract typically specifies the price to be paid for moving a container with specified cargo from specified origins to specified destinations. Often, the contract also specifies a minimum quantity commitment, which is the minimum quantity of freight that the shipper will offer to the carrier during the contract period. The contract also specifies a service commitment, stating that the carrier will serve the shipper to the extent of the minimum quantity commitment, and possibly more, "reasonably spread over the duration of the contract", as long as the shipper "gives 14 days booking notice, but not less than 7 days".

In practice, booking requests are often made less than 14 days before the freight becomes available for shipment, in which case the carrier may delay moving the freight if insufficient capacity is available. For example, the shipper may request a booking on a voyage that departs 9 days into the future, but if that voyage is already too fully booked, then the carrier may offer to transport the freight on a voyage that departs a week later, thus 16 days into the future. The shipper may then accept the booking offer, or decline the booking offer and request a booking with another carrier. Also, often shippers cancel previously made bookings. It also happens that freight scheduled to depart on a voyage does not arrive at the port in time for the voyage. Approximately 30% of booked freight ends up not transported as originally booked. The booking control decision maker at the carrier has to decide, whenever a shipper requests a booking, what booking to offer the shipper. For example, the carrier may offer the shipper a booking on the voyage 9 days into the future, or on a voyage scheduled to depart a week later. The decision maker should take into account the price (revenue) specified in the contract with the shipper, the probability of the shipper declining the booking offer, as well as uncertain future booking requests, cancellations and no-shows.

Because of the different economic needs in different regions, most liner trade is imbalanced. Liner operators, therefore, often need to reposition their empty containers or to lease containers from vendors to meet a customer's demand. To do this effectively, the liner operators need to take into consideration the voyage schedule, capacity on each vessels, the leasing cost and inventory level at each port.

Many different models have been proposed in the literature to obtain an optimal container repositioning plan. However, most of these approaches lead to cumbersome stochastic integer programming/network flow models that may not be easily implemented due to the lack of data concerning vessel¡¯s capacity and demand distribution. We plan to build in this project a model that requires minimal input and that will produce plans that remain "robust" under different scenarios. We are currently working with an approach that allows us to replace the complicated recourse function in the stochastic programming model by a linear programming model. Further work needs to be done to evaluate and compare this approach with existing methods.

What is the optimal mix of cargo in a vessel run with revenue optimization as the objective? As part of this project, we plan to study the tradeoff in FCL and LCL cargo mix in a typical run and highlight the various strategic and operational issues. The problem is compounded by the fact that different types of cargoes generate different revenue. A FCL of electronics pays higher than a FCL of gypsum board. Furthermore, if the inbound vessel is full, and if this is not balanced by a full outbound, then there is the need to ship empties necessitated by the need to re-position the empties for use by the destination port. This too has to be taken into account in the revenue optimization formulation. On a micro-level, we want to examine if FCL is preferred to LCL cargo. A container load of LCL cargo may generate higher revenue but one has to taken into account the variable cost. A LCL container load involves many consignors (and as many consignees) and with each consignor, the container carrier incurs a variable cost (e.g., documentation). It is therefore not obvious that FCL necessarily pays higher than LCL container loads. Variable cost for LCL is highly variable and there may be some break-even point depending on the number of consignors, beyond which a FCL is preferred. Of course, the administrative process has improved greatly with the wide application of information technology, especially the Internet, in recent years. The structure of administrative cost for FCL and LCL needs to be re-examined. In summary, the proposed study on the optimal mix cargo and the FCL/LCL issue is promising in obtaining new insights that would have real impact on current practices.