Research On Stacks Intelligent Allocating Method Of Bulk Yard In Port

The bulk stack is the core resource of the port and the allocation of stacks is an extremely important part of the Stack-Resource-Scheduling.It will reduce the utilization of the stack resources,increase operation costs and increase the pressure of port throughput when the allocation of stacks is unreasonable.It is a momentous problem in the process of building Smart Port to make the allocation of stacks reasonable,improve the matching degree between cargo and stacks,improve the utilization of bulk stack and reduce operation costs.Therefore,the research on the allocation strategy for bulk stack in bulk port is of great theoretical value and practical significance.According to the mentioned problems,this thesis has carried out related research,and the main content is as follows:(1)A formalized model of bulk stack is built based on grid and the model is used to describe the resources in stack and the stack status.Firstly,a gridding method for bulk stack is applied to build the grid coordinate system for the comprehensive bulk stack which has no definite stacks divided.Then,the directed graph which the node is the cargo stacks and the edge is constructed by analyzed the connectivity between cargo stacks is built.Finally,a calculation method for the stack utilization efficiency is proposed to evaluate the state of the stack at a certain moment,this method is based on the directed graph and the actual operating environment of bulk stack.This method includes the capacity of stack road,the efficiency of flexible stacks,the rationality of adjacent stacks and the coverage of stacker,and quantification method is designed for each part.(2)A bulk stack allocation algorithm based on dynamic game(BSAABG)is designed to resolve the competition problem for stacks caused by distributing stacks for multiple cargos.Firstly,the distribution behavior for multiple cargos is modelled as a dynamic game and the notion of satisfaction equilibrium is applied to analyze the game.The stacks allocation for bulk stack model based on the game theory is established and the set of available stacks is strategy set,and the matching degree between cargo and stacks is the utility of the cargo.Then,the BSAABG is used to solve the model,and the convergence of the algorithm is proved theoretically.The algorithm included initial game and the iterative game.Each cargo chooses the max-utility stack from available stack set by greedy strategy in initial game stage.In the subsequent iterative game stage,the game is repeated by adjusting the game order and the strategy until the allocation is balanced and the stack allocation efficiency is maximized.(3)The model and the algorithm in this thesis are experimentally verified.As for the formalized model of bulk stack,the method to calculate the capacity of stack road and the forecast model of cargo turnover period based on Random Forest algorithm in the efficiency of flexible stacks are mainly verified.Firstly,the method to calculate the capacity of stack road is verified based on the method to analyze road capacity in city and the real operation environment in stack,the results show that the method in this thesis could evaluate the capacity of stack road effectively under heavy traffic conditions.Then the validity of the forecast model of cargo turnover period is verified by collecting actual production data.The prediction accuracy is improved by setting the accuracy threshold to screen decision tree and the importance of each feature in affecting the cargo turnover period are analyzed.In the experiments of the stacks allocating method based on dynamic game.The stack allocation efficiency threshold which determine the game order does not need to change is firstly identified.Then,the rationality of using greedy algorithm to simulate the experience of manual scheduling is verified.Finally,compare the BSAABG with Greedy Algorithm and Stacks Allocation Algorithm Based on Rule,the experimental results show that the BSAABG could effectively improve the average satisfaction of the cargo and the stack allocation efficiency when the number of cargos is large.In summary,the work in this thesis is a preliminary exploration on the intelligent construction of bulk ports.It is of great practical significance and value for improving the utilization of bulk stack and reducing operation costs in bulk ports.