Improved Wolf Pack Algorithm And Its Application To Yard Crane Scheduling Optimization At Container Terminals

With the increasing scale of international import and export trade,the container transport industry has developed rapidly.Accompanied by the development is fierce competition among container terminals.In order to improve the competitiveness and operational efficiency of the port itself on the basis of not increasing the cost input.It is necessary to increase the utilization rate of related port resources and equipment.The yard crane is the main loading and unloading equipment in container yard.It is meaningful and valuable to study and discuss the yard cranes scheduling problem.It concerns the operation efficiency and cost of the whole container terminals.In this paper,the yard cranes scheduling problem in container yard while import and export containers are stored in same block is the engineering background.Different waiting rates,charging time and longest waiting time are set for internal and external trucks,as well as the priority of being served.At the same time,practical constraints such as the work quantity balance and interference or spanning of yard cranes are taken into account when multiple yard cranes are operating at the same time.Mathematical optimization models of single and multiple yard cranes scheduling problems in mixed container area are established,whose aim is the sum of waiting cost of trucks and moving cost of yard cranes.In order to solve the above scheduling problem model,a novel wolf pack algorithm(WPA)is selected for further study.Then,an improved wolf pack algorithm(IWPA)is proposed.The improved algorithm introduces adaptive parameter adjustment strategy,random initialization strategy of stagnation state and wolf pack update strategy based on hunger value.The adjustment of adaptive parameters can make the step size of the three intelligent behaviors in the algorithm change adaptively with the iteration.In the early stage,larger step size is conducive to speeding up global search.In the later stage,the step size is small,which can strengthen the local search and make it easier to find the optimal solution.Random initialization of stagnation state can prevent algorithm prematurity and keep population diversity.The wolf pack update strategy based on hunger value is helpful to retain the better solution and speed up the algorithm to find the optimal solution on the basis of maintaining the population diversity.In this paper,I further discretize the above strategies and present a discrete version of the improved wolf pack algorithm to solve the above discrete yard cranes scheduling problem.In order to verify the performance of the proposed algorithm,it is used to solve function optimization problems and traveling salesman problems.Then we compare the results of the proposed algorithm,the basic wolf pack algorithm and other algorithms in the literature.The feasibility and effectiveness of IWPA algorithm be verified in solving continuous and discrete combinatorial optimization problems.Then,the proposed algorithm is applied to the above yard cranes scheduling model,and the examples of single and multiple yard cranes scheduling problems are solved respectively.By comparison with the results of basic wolf pack algorithm and other algorithms given in literature,the proposed IWPA algorithm has lower cost,faster convergence speed,better accuracy and precision.It shows that the proposed algorithm has better performance in solving the scheduling problem and can obtain a satisfactory scheduling scheme.In this paper,relevant explorations are made on the solution of yard cranes scheduling problems at container terminals,which has inspiration and reference significance for similar scheduling problems such as port quay cranes and urban bus.