Research On Scheduling Optimization Of Container Stowage And Transportation Process

Containers, as the leading modern logistics carrier, brought a revolution to the world bulk cargo transportation, and its typical operation shape is intermodal in the form of sea, rail, road, inland waterways and other modes of transport organically constitute a continuous, comprehensive and integrated cargo transport networks, to achieve a "door to door" logistics services. Because of the complication of the container transportation, it is necessary to process the related links at all levels and information using advanced and intelligent technology and methods.This paper considered the repositioning of empty containers as the starting point for the optimization. The repositioning of empty containers is a complex problem which has perplexed shipping companies for a long time. This paper analyses container shipping network and takes a digital depiction for the empty container repositioning network in Bohai bay as an instance based on the database of the empty container repositioning in this area, which is constructed by Access. Then, the whole network is visualized by Mapinfo, which lays the groundwork for intellectualizing the decision-making on empty container repositioning in Bohai bay. Then this paper presented an improved empty container allocation process in order to reduce the allocation cost. The empty container problem is formulated as nonlinear integer programming and an immune genetic algorithm (IGA) is designed to optimize empty container scheduling. Finally, an empirical analysis for Qingdao port was conducted to simulate and evaluate the proposed allocation process, which showed the effectiveness of proposed model and algorithm.Then container stowage was discussed. A MIP model, to focus on the containers loading problem under the realistic constraints, is proposed. Based on pre-allocation strategy, a hybrid algorithm, which combines genetic and heuristic algorithms, has been put forward to solve the problem, which not only generates efficient loading patterns but also satisfies some releastic constraints such as cargoes bearing capacity, container loading stability and mixed loading etc. Numerical experiments, from the benchmark problems, showed that the proposed algorithm was superior to those in the literature.Since there is disagreement between stochastic arrival sequence of export containers and their loading sequence, it is inevitable to relocate containers. Besides, inappropriate relocation will lead to another relocation once more, which heavily affects the loading efficiency of the container terminal. Based on the given container yard and containership stowage plan and actual constraints for the customs examination, a mixed integer programming model for containership loading sequence is proposed to aim at the least relocation. Furthermore, a heuristic algorithm is developed to efficiently get the near optimal solutions. The experiment, from a container terminal, showed the effectiveness and practicality of both the model and algorithm.The container multimodal transportation is an advanced organization manner, by which the coordinate organization and management among all kinds of transportation modes and among all participants of the whole network is an objective requirement. The graded structure multimodal transportation planning is analyzed firstly. Then, combining these data analysis and mining above, the conceptual model of graded optimization is put forward. Besides, mathematical models are proposed for the container regular liner shipping and inland transshipment planning subproblems. By means of the analysis on the structural characteristic of the problem solution, two corresponding genetic algorithms are designed for the two subproblems. Simulation tests show their efficiency of the proposed models and algorithms.Finally, based on the analysis of all the process during the multimodal transportation, this paper constructed a multi-dimensional database. The multi-dimensional data analysis and data mining were made on the database. The application show proves the feasibility and effectiveness of the multi-dimensional analysis and data mining in the container multimodal transportation information processing. Decision support can be included from the application.