Research On The Space Allocation Of Export Containers In Stacking Yards Based On Improved Firefly Algorithm

With the rapid development of global economy,international and interregional freight trade have become more frequent and the throughput of containers rapidly increased,which cause the contradiction between the space resources and container throughput of the terminal yard becoming more severe.Therefore it is an economical and effective way to improve the utilization ratio of existing space resources in the storage yard.Distributions of storage for outlet containers directly affect ship berthing time,loading and unloading efficiency,even production efficiency of the whole port.Under the background mentioned above,research on distributions of storage for outlet containers becomes one of the hot issues on marine transport,automated container terminals and other related areas of research.In this paper,the objective to make full use of the yard space resources was achieved by optimizing the stacking location in yard for outlet containers.According to the basic principle of the container allocation in the container yard and considering turning-box volume,equilibrium of container distribution,transport distance and some relevant constraints in a mixed storage model of different destination ports and weight classes,a mathematical model which can optimize the allocation of outlet containers in yard was proposed.It will be difficult to solve this problem by conventional methods because it has NP-hardness.A novel firefly algorithm was applied to solve the problem and a parallel adaptive firefly algorithm(PAFA)is proposed in order to avoid premature falling into local optimum and to increase the diversity of population.Main idea of the algorithm is introducing crossover,mutation and adaptive strategies of genetic algorithm based on the principle of the essence of discrete issues and the basic principle of firefly algorithm.The detailed procedures are as below,firstly discrete standard firefly algorithm and redefine distances and positions between individuals to make it suitable for solving discrete problems.Then according to the idea of self-adaption,the position update process was transformed into self-adaptive position update mode.In the meanwhile parallel strategy was adopted,in which the population is divided into two sub groups,one is focusing on optimizing individual learning to speed up convergence,the other is responsible for both search and development at different stages of the algorithm.To improve the overall performance of the algorithm,the two sub groups conduct information exchange regularly.The algorithm proposed in this paper was verified by two examples of TSP,whose feasibility and effectiveness was showed in the results.Further,taking outlet container allocation problem of relevant routes in container port area of DAYAO Bay port as engineering background,algorithm proposed in this paper was applied,and two kinds of working conditions was tested and simulated.It is showed in the result that the algorithm proposed in this paper has obvious advantages in the allocation of containers,a better container allocation scheme was given which can effectively improve the space utilization and reduce the transportation and operation cost.The research work in this paper has inspiration and referential significance for the further research on the problem of the allocation of the container and the practical engineering of the algorithm,which have certain theoretical significance and applicable value.