单向航道下连续泊位分配与船舶调度集成优化

为改善单向航道连续泊位港口的运营效率,研究泊位分配与船舶进出港调度集成优化.考虑潮汐、进出港时段交替与偏好泊位的影响,建立0-1整数线性规划模型,以船舶偏离偏好泊位成本和滞期成本为优化目标,确定各艘船舶的靠泊位置与进出港时刻.针对问题情境和其特有的约束条件,将原数学模型通过Dantzig-Wolfe分解方法分成主问题模...     

低碳经济下的港口泊位分配模型及其算法实现

航运公司正在进行前所未有的努力以减少船舶的燃油消耗量及碳排放量,而港口所制定的泊位分配计划对于船舶的油耗量和碳排放量有着直接的影响。由于船舶的到港时间是港方制定泊位分配计划的关键参数,因此将船舶到港时间作为决策变量引入传统的泊位分配（BAP）模型中,设计了港口与船方协调调度的新的泊位分配策略--VAT（Variable Arrival Time）策略,同时将船舶油耗和碳排放量融入BAP 模型的目标函数中,建立了船舶油耗量最小和船舶离港延迟时间最短的双目标优化模型。采用多目标遗传算法对该模型进行求解,并用仿真算例验证了该策略的有效性。计算结果表明,VAT策略可以大大削减航运公司的燃油消耗和船舶的碳排放,同时可以提高港口的服务水平,缩短船舶在港等待时间。     

多目标约束处理方法求解泊位岸桥联合分配问题

为了制定合理高效的泊位岸桥联合分配方案,加快船舶周转,本文针对船舶动态到港的连续泊位建立了以船舶总在港时间最短为目标的泊位岸桥联合分配混合整数非线性模型.通过多目标约束处理策略将复杂约束的违反程度转化为另一个目标,从而将原单目标优化模型转化为双目标优化模型,并用基于快速非支配排序的多目标遗传算法(NSGA-II)对其进行求解.同时,针对问题特点,分别设计了基于调整、惩罚函数、可行解优先和综合约束处理策略的单目标遗传算法对原模型进行求解.通过多组不同规模的标准算例对本文的方法进行测试,验证了基于多目标约束处理策略的方法求解效果相较于单目标约束处理策略的方法更加高效和稳定.     

Collaborative mechanisms for berth allocation

This paper proposes two collaborative mechanisms between container shipping lines and port operators to facilitate port operators to make proper berth allocation decisions. In the first mechanism, assuming no transshipment, a shipping line needs to provide the port operator with the utilities associated with the start operation days of each liner route. The total utilities for all start operation days must be 0. A higher bunker and inventory cost for the shipping line means a lower utility. The port operator compensates the shipping line if its ship is scheduled on a day with negative utility and charges additional fees if the ship is scheduled on a day with positive utility. The second mechanism accounts for the utilities related to the inventory cost of transshipment containers. These two mechanisms ensure that shipping lines have no incentive to overstate or undervalue the utilities. The utilities estimated by shipping lines are much more accurate than those estimated by port operators. Hence, models for the tactical berth allocation problem incorporating the utilities provided by shipping lines lead to more efficient and equitable berth allocation plans. The utilities provided by shipping lines can also guide the decisions on operational berth allocation.     

A distributed agent system for port planning and scheduling

This paper describes a distributed agent system for dynamic port planning and scheduling. The proposed system comprises four agents, namely a port planning manager (PPM), a berth control agent (BCA), a shuttle allocation agent (SAA) and a yard storage agent (YSA). These agents communicate and cooperate with one another to work out the schedule for berth allocation and requirements for shuttles. Basically, the PPM maintains all the necessary information and provides a negotiation and communication locale for the BCA, SAA and YSA, which represent the berths, the shuttles and the container yard, respectively, to resolve their conflicts under its supervision. With the help of a GA-enhanced dynamic scheduler, the BCA attempts to optimise the berth allocation schedule using the data such as ship arrival information retrieved from the PPM. As soon as the SAA receives the schedule from the BCA, it proceeds to check for the availability of shuttles and work out the requirement for shuttles. On the other hand, the YSA generates the yard storage allocation schedule and the truck/train schedule for the transportation of containers.A case study is conducted to illustrate the capability of the distributed agent system. It has been shown that the proposed system is able to successfully generate the schedules for both the BCA and the SAA.     

桥吊可动态分配的连续泊位分配问题算法

研究在允许桥吊动态分配的情况下集装箱码头的连续泊位动态分配问题,并建立以船舶在港时间最小为目标的动态泊位分配模型;然后基于兄弟-儿子方法对船舶的位置进行调整以规划桥吊使桥吊不交叉。在相同算例下,比现有方法得到的船舶在港时间更少,从而验证了模型及算法的有效性。     

离散泊位布局下的泊位岸桥动态协调调度

为了制定合理的集装箱码头泊位岸桥资源调度计划,提高码头作业效率和客户满意度,基于离散泊位布局,建立了以在港集装箱船总的服务成本最小为优化目标的动态泊位岸桥协调调度模型。设计了遗传算法对模型求解,通过将部分约束条件嵌入算法结构简化了模型求解难度,并对算法迭代过程中的不可行解采用逐时刻基因调整策略进行修复。经过若干算例的数值实验,验证了模型和算法的可行性和有效性。     

Hybrid rolling-horizon optimization for berth allocation and quay crane assignment with unscheduled vessels

Port operations usually suffer from uncertainties, such as vessels’ arrival time and handling time and unscheduled vessels. To address this, this study presents a dynamic berth allocation and crane assignment specific problem (BACASP) when unscheduled vessels arrive at the port, which is branded the berth allocation and quay crane assignment specific problem with unscheduled vessels (UBACASP). A rolling-horizon based method is proposed to decompose the UBACASP into a multi-stage static decision BACASP, wherein a rescheduling margin-based hybrid rolling-horizon optimization method is developed by incorporating the event-driven and periodical rolling-horizon strategies as the urgency of dynamic events is evaluated. In each rolling horizon, a mixed integer linear programming model (MILP) is presented for the BACASP to minimize the total port stay time of vessels and the penalties of delays associated with the spatial and temporal constraints, such as the length of continuous berth, number of quay cranes (QCs) and non-crossing of QCs. A discretization strategy is designed to divide the continuous berth into discrete segments, and convert the BACASP to a discrete combinatorial optimization problem, which is efficiently solved by the proposed adaptive large neighborhood search algorithm (ALNS). Case studies with different problem characteristics are conducted to prove the effectiveness of the solution methods proposed in this study. Moreover, the performances of the ALNS and the existing methods for solving the BACASP are compared, and the advantages and disadvantages of different rolling strategies under different degrees of uncertainties are deeply analyzed.     

Integrated intelligent techniques for remarshaling and berthing in maritime terminals

Maritime container terminals are facilities where cargo containers are transshipped between ships or between ships and land vehicles (trucks or trains). These terminals involve a large number of complex and combinatorial problems. Two important problems are the container stacking problem and the berth allocation problem. Both problems are generally managed and solved independently but there exist a relationship that must be taken into account to optimize the whole process. The terminal operator normally demands all containers bound for an incoming vessel to be ready in the terminal before its arrival. Similarly, customers (i.e., vessel owners) expect prompt berthing of their vessels upon arrival. This is particularly important for vessels from priority customers who may have been guaranteed berth-on-arrival service in their contract with the terminal operator. To this end, both problems must be interrelated.In this paper, a set of artificial intelligence based-techniques for solving both problems is presented. We develop a planning technique for solving the container stacking problem and a set of optimized allocation algorithms for solving the berth allocation problem independently. Finally we have developed an architecture to solve both problems in an integrated way. Thus, an algorithm for solving the berth allocation problem generates an optimized order of vessels to be served meanwhile our container stacking problem heuristics calculate the minimum number of reshuffles needed to allocate the containers in the appropriate place for the obtained ordering of vessels. Thus combined optimal solutions can be calculated and the terminal operator could decide which solution is more appropriate in each case. These techniques will minimize disruptions and facilitate planning in container terminals.     

A network flow model for the dynamic and flexible berth allocation problem

This paper presents a new model for the dynamic berth allocation problem (BAP). The model is developed using a berth-flow network modeling approach and is formulated as an integer multi-commodity network flow problem. In addition, an innovative flexible berth-space utilization scheme, based on blocking plans, is incorporated into the proposed model. This is referred to as the dynamic (vessel arrivals) and flexible (berth space) BAP model (or DFBAP), and is designed to better utilize wharf space in a container port. Computational experiments conducted on an instance generated using actual data show that the DFBAP model is more effective and efficient than the method currently used by port authorities. A set of scenario analyses is also performed to obtain insights into important model parameters.     

基于作业链的泊位与岸桥协同调度研究

针对泊位与岸桥协同调度问题,引入"链式优化"思路,用作业链的方法分析集装箱装卸作业过程,首先将泊位计划作为开始链单元,采用资源节点优化策略进行分析,以最小化船舶在港总成本为目标建立模型;然后将岸桥卸船作业作为结束链单元,采用任务节点优化策略进行分析,以最小化岸桥最大完工时间为目标建立模型.考虑到作业链的整体性能,设计嵌...     

基于改进遗传算法的泊位岸桥协调调度优化

针对集装箱码头资源调度不合理造成资源浪费的问题,在考虑岸桥装卸成本的基础上,以在港集装箱船总的作业成本最小为优化目标,建立了基于非线性混合整数规划的泊位岸桥协调调度优化模型。为使模型更加接近码头操作的实际情况,模型假设船舶装卸时间依赖于为其分配的岸桥数。采用基于可拓关联函数的改进遗传算法对模型进行求解。改进算法强调了不可行解的重要性,用可拓关联度来衡量种群中不可行解的优劣程度,通过在种群迭代中始终保持一定数量的不可行解来维持种群多样性,从而克服传统算法局部搜索能力较差的缺陷。数值实验验证了模型和算法的可行性和有效性,与不考虑岸桥装卸成本的模型相比,能够有效减少港口资源的浪费。     

Quay crane and yard truck dual-cycle scheduling with mixed storage strategy

In order to enhance the efficiency of port operations, the scheduling problem of the quay cranes and yard trucks is crucial. Conventional port operation mode lacks optimization research on efficiency of port handling operation, yard truck scheduling, and container storage location. To make quay crane operations and horizontal transportation more efficient, this study uses a dual-cycle strategy to focus on a quay crane and yard truck scheduling problem in conjunction with a mixed storage strategy. A dispatching plan for yard trucks is considered, as well as the storage location of inbound containers. Based on the above factors, a mixed-integer programming model is formulated to minimize vessels’ berth time for completing all tasks. The proposed model is solved using a particle swarm optimization-based algorithm. Validation of the proposed model and algorithm is conducted through numerical experiments. Additionally, some managerial implications which may be potentially useful for port operators are obtained.     

Model and heuristic for berth allocation in tidal bulk ports with stock level constraints

We consider the problem of allocating berth positions for vessels in tidal bulk port terminals. A berth is defined as a specific location alongside a quay where a ship loader is available for loading or unloading vessels, accommodating only one vessel at a time. In tidal ports, draft conditions depend on high tide conditions, since available depth at low tide is not adequate for the movement of ships. Some port terminals are associated with important transnational enterprises which maintain strong control over the stock level of their goods. Since the stock level sometimes depends on a continuous process of consumption or production of minerals, the decision to load or unload vessels must consider the amount of the bulk cargo stored in the port yards. Therefore, a basic criterion for decision making is to give priority to the vessels related to the most critical mineral stock level. A second basic criterion is to decide what sequence of vessels reduces the overall demurrage within a given planning horizon. This paper presents an integer linear programming model based on the transportation problem to represent the Berth Allocation Problem in Tidal Bulk ports with Stock level conditions (BAPTBS). Problem instances are solved by a commercial solver and by a Simulated Annealing-based algorithm (SA). The SA employs a problem-specific heuristic, becoming a valid alternative for finding out good solutions for difficult instances.     

潮汐影响下泊位和岸桥双目标联合调度仿真

针对集装箱船舶大型化导致的港口航道现有水深无法满足大型船舶安全吃水深度,需要借助潮水上涨进出航道的现状,研究了潮汐影响下连续型泊位和动态岸桥联合调度问题。建立了以最小化船舶周转时间和岸桥在船舶间移动次数的双目标混合整数规划模型。基于问题特点,设计了Epsilon约束精确算法和带精英策略的快速非支配排序遗传算法(NSGA-Ⅱ)分别求解小规模和大规模算例的Pareto最优解集,所得结果验证了模型和算法的正确性与有效性。通过潮汐周期灵敏度分析评估了潮汐周期长度对岸桥工作效率和港口服务质量的影响。仿真结果表明,建立的优化模型能够帮助港口企业有效降低潮汐对生产作业的影响,同时提供一组高效的Pareto最优泊位岸桥调度方案提高工作效率和经济效益。     

A bi-objective model for robust berth allocation scheduling

Berth allocation is an important port operation problem for container terminals. This paper studies how to develop a robust schedule for berth allocation that incorporates a degree of anticipation of uncertainty (e.g., vessels’ arrival time and operation time) during the schedule’s execution. This study proposes a bi-objective optimization model for minimizing cost and maximizing robustness of schedules. A heuristic is also developed for solving the bi-objective model in large-scale problem cases. Numerical experiments are conducted to validate the effectiveness and efficiency of the proposed model and method. Managerial implications are also discussed.     

求解动态停泊计划问题的拉格朗日松弛算法

研究钢铁企业原料码头动态停泊计划问题,其动态特征主要体现在原料船动态到达并有两个或两个以上连续泊位且在停泊计划开始执行时每一泊位上仅有部分泊位长度可利用。针对这个问题,建立了一个数学模型并设计了改进拉格朗日算法在很短的时间内求得了近优解。在改进算法中使用了所提出的四个性质来分别加速求解子问题、乘子更新和获得可行解的过程。通过包含50个实际规模问题的算法性能实验表明改进的拉格朗日松弛算法相比未改进算法减少了80%的运行时间。     

The berth allocation problem with mobile quay walls: problem definition,solution procedures,and extensions

The berth allocation problem (BAP), which defines a processing interval and a berth at the quay wall for each ship to be (un-)loaded, is an essential decision problem for efficiently operating a container port. In this paper, we integrate mobile quay walls into the BAP. Mobile quay walls are huge propelled floating platforms, which encase ships moored at the immobile quay and provide additional quay cranes for accelerating container processing. Furthermore, additional ships can be processed at the seaside of the platform, so that scarce berthing space at a terminal is enlarged. We formalize the BAP with mobile quay walls and provide suitable solution procedures.     

Simulation-based investment planning for Humen Port

This paper addresses an investment planning problem for a container terminal in Humen Port using simulation with Arena software. We propose a simulation model considering various types of container ships and cranes, flexible berth allocation and dynamic crane scheduling. We carry out simulation experiments in order to identify a parameter setting that can minimize the total investment cost while maintaining a required service level. The experiment results show that the proposed economic investment plan can reduce the cost compared with the current investment plan.     

基于蚁群算法的泊位调度优化与仿真

在全球贸易经济聚焦在中国的同时,港口的吞吐能力成为目前港口业的主要矛盾。提高泊位这个环节的运作能力,减少船舶在港时间,增加港口的吞吐能力成为主要研究对象。本文采取仿真模型与优化算法相结合的研究方法,把泊位调度问题转化为旅行商问题,建立了一个泊位岸桥协调调度,通过蚁群算法建立数学模型,使船舶在港时间最短为目标建立函数,求得最佳调度方案。用ProModel建立船舶到港停泊及离港仿真模型。验证泊位调度优化的有效性,以便指导港口实际的泊位调度。