Containerization technique greatly improves the efficiency of international cargo shipping. Container terminals, as the service gateways of hinterland transport and maritime transport, play a crucial role in the process of cargo transport. In presence of growing demand for containerized marine shipping and increasing sizes of container vessels, a swift service at container terminal is vital. To improve service productivity of a container terminal, one option is to increase the capacity of container terminal or purchase new facility. However, the obvious drawback of this option is the high investment cost involved, which discourages terminal operators from taking this option. In contrast, a more viable option is taking advantage of planning tools such as the planning of quay side operations or land side operations, which will not incur a significant amount of investment. In this thesis, we focus on developing more realistic models and efficient solution algorithms for these types of problems in optimizing container terminal. Integer programming is widely used in many application areas and plays a fundamental role in supporting managerial decisions or modeling hard to solve problem. Most of ocean shipping problems can be modeled as integer programming or mixed integer programming, such as cargo routing, fleet deployment, crane scheduling, stowage planning, etc. Different from linear programs which is solvable in polynomial time, integer programming problems are normally NP-hard. Many solution methods have been proposed by the previous studies, one of the well-known and successfully applied methodology is Benders decomposition. This method was mostly used for LP problems or MIP problems.