Study On Scheduling Optimization Of Container Terminal Handling System

In the face of the rapidly changing international port market competition environment,how to improve the handling efficiency of container ports and meet more customer needs is the core goal of port enterprises under the restriction of limited resources.In particular,the demand for containers in the early days of the COVID-19 epidemic has soared,even "one box is hard to find",to the ups and downs of the demand for containers in the normalization of the epidemic.The frequent alternation between the port blockades and the "dock sunning" has brought great fluctuations in the handling capacity of containers at the port,which has brought great difficulties and challenges to the port production scheduling.Nansha Port Container Terminal Phase II Project has excellent economic hinterland conditions,location conditions,complete collection and distribution system and route network.Since it was put into operation,the container throughput has increased day by day,and the phenomena of "port congestion" and "sunshine" have occurred occasionally.Facing the competitive pressure of surrounding ports and the production characteristics of container terminals,how to use the non renewable port shoreline resources on the basis of accurate prediction of container throughput,optimize the berth allocation,improve the allocation level of quayside bridges,container trucks and other terminal production resources,and achieve the accuracy of port scheduling,so as to reduce the total time of ships staying in the port and improve the use of berths,Improving the efficiency of port production and operation has become the core issue of container port scheduling.Taking the production and operation of container terminals as the research object,this paper studies the precise optimal scheduling of ports based on the throughput prediction of container terminals.First of all,empirical mode decomposition+support vector machine prediction model(EMD-SVR),autoregressive comprehensive moving average model decomposition+support vector machine prediction model(ARIMASVR)are successively constructed in the container terminal throughput prediction.Relying on the 108 month monthly container throughput data of Guangzhou Nansha Port Phase II Terminal as the training set,absolute percentage error(APE)The ARIMA-SVR container map throughput prediction model with higher prediction accuracy is obtained through comparative analysis of the average absolute percentage error(MAPE)and other indicators.On the basis of this prediction model,the monthly container forecast volume of Guangzhou Nansha Port Phase II Terminal in the next 24 months is predicted.Combined with the number of existing shoreline berths,quayside bridges,trucks and other equipment and facilities of Guangzhou Nansha Port Phase II Terminal,It is estimated that the port will be blocked many times in the next 24 months.Then,in view of the possible problems of port congestion in the future of Guangzhou Nansha Port Phase II Terminal,a port optimal dispatching model is established in the design of the port optimal dispatching model to maximize the operating income of container terminal operators to meet the requirements of ships staying at the port.The objective function in the model considers the costs of ship berthing income,quayside bridge loading and unloading,truck transportation,and truck waiting during the study period,In addition,the penalty cost of ship delay is added(in traditional literature,most of the terminal scheduling optimization is aimed at minimizing the total dwell time of the ship,considering less terminal operating income).Therefore,this problem not only maximizes the operating income of the container terminal,but also meets the requirements of ship stopping at the port.Compared with the traditional port optimal scheduling,it increases its complexity,and achieves the balance between ship and terminal services and costs.Genetic algorithm is used to solve the model.In the numerical example part,the paper uses the actual case data of a ship berthing for 48 consecutive hours in the monthly throughput forecast of Guangzhou Nansha Port Phase II,which will lead to port congestion.The algorithm in the paper is used to set the maximum iteration number of 1000,population size of 200,cross probability of 0.8,and variation probability of 0.08.During the forecast period,the operating income of the container terminal is8305467.914 yuan,and the income of a single container is 232.3 yuan/TEU,The allocation scheme of berth,quayside bridge and container truck for container ships arriving at the port within 48 consecutive hours is obtained.Compared with the production data of similar ports in 2021,the single box income after optimization has increased by 7.94%.It proves the validity of the port optimal precise scheduling model,and effectively solves the container terminal port optimal scheduling problem.In general,the case calculation results conform to the practical logic.The model and algorithm can meet the requirements of container terminal scheduling optimization,and can also provide reference for other similar container terminal enterprises to design terminal scheduling optimization schemes.