Dry bulk cargo wharf berth distribution and ship loader combined dispatching method and system

Abstract

The invention belongs to the technical field of berth distribution, and discloses a dry bulk cargo wharf berth distribution and ship loader combined dispatching system based on self-adaptive MOEAD. The invention provides a dry bulk cargo wharf berth allocation and ship loader combined dispatching system based on self-adaptive MOEAD. The system comprises an input module, an algorithm module, a berth allocation module, a loading and unloading dispatching module, a notification module, a control module, a loading and unloading operation module and a port leaving module. The input module is connected with the scheduling module and is used for inputting a ship plan table; the scheduling module is connected with the berth distribution module and the loading and unloading scheduling module and is used for scheduling through a scheduling algorithm; the berth distribution module is connected with the notification module and is used for carrying out berth distribution on the ship; and the loading and unloading scheduling module is connected with the control module and used for scheduling loading and unloading equipment. The operation efficiency of port loading and unloading is improved. The energy consumption of port loading and unloading equipment is reduced; the algorithm is high in convergence speed and has good convergence and distribution.

Description

Technical Field

[0001] The present invention belongs to the technical field of berth allocation, and in particular to a method and system for berth allocation and ship loader joint scheduling of a dry bulk terminal based on adaptive MOEA / D. Background Art

[0002] In recent years, the construction of green and low-carbon dry bulk terminals has gradually attracted the attention of more and more countries. At the forefront of dry bulk terminals, berth allocation and ship loading and unloading machinery and equipment scheduling are the research focuses in terminal operations. In the past, most of the studies optimized the two problems independently, ignoring the importance of joint scheduling. Most of the existing research focuses on container terminals, while there are relatively few studies on dry bulk terminals. Compared with container terminals, the resource allocation and scheduling of dry bulk terminals often need to distinguish between loading and unloading, and consider factors such as the stability of ships with multiple hatches and multiple equipment operating simultaneously and the yard; secondly, berths and loading and unloading equipment are the most scarce resources at the terminal. Reasonable arrangement and scheduling of them are crucial to ensure timely berthing and departure of ships and improve the service level of the port. It is difficult to study the berth allocation and joint scheduling of ship loaders at dry bulk terminals, and the research has theoretical value and practical significance; in the context of green and low-carbon terminals, green scheduling research that takes energy saving into consideration while ensuring the shortest time for ships in port is the future development direction; although the MOEA / D algorithm has been used to deal with different combinatorial optimization problems, there are few studies on the use of the MOEA / D algorithm to solve the berth allocation and joint scheduling of ship loaders. Therefore, the application of the MOEA / D algorithm to solve the berth allocation problem of ships in terminals and ports has certain research potential.

[0003] Berth allocation is one of the most important links in port operations. It mainly determines the berthing position and berthing and unberthing time of ships. It involves the time-space allocation of berths for ships and is an NP (non-deterministic problem of polynomial complexity) problem in operations research. Ships arrive over time and the terminal needs to allocate them to berths so that they can be served as soon as possible. Different factors affect the berth location and time allocation of each ship. For the model of this problem, there are four common situations: discrete and continuous berthing space, static and dynamic ship arrival, static and dynamic ship processing time, and variable ship arrival.

[0004] In discrete problems, the terminal berths are considered as a finite set of berths; in continuous problems, the ship can berth anywhere in the terminal berths, and most studies deal with discrete berth allocation. In the static ship arrival berth allocation problem, berths are allocated to the ships that have arrived at the port, while in the case of dynamic ship arrivals, only a portion of the scheduled ships are in port. In the static handling time problem, the ship handling time is the input data, while in the dynamic case it is a decision variable. In the last variable ship arrival case, the ship arrival time is considered as a variable.

[0005] Some studies dealing with berth allocation problems set technical constraints on berthing draft, inter-ship and end berth clearance according to the actual conditions of the terminal, making the problem description closer to the terminal production conditions. It is quite simple to introduce technical constraints to the existing berth allocation model, which increases the complexity of the problem to a certain extent, but at the same time will reduce the feasible space and simplify the use of meta-heuristic methods.

[0006] Typically the objective functions of this problem are: minimizing the total service time of the ship (waiting and handling time); minimizing early and late departures; optimizing the ship's berthing time; optimizing emissions and fuel consumption.

[0007] Through the above analysis, the problems and defects of the prior art are as follows:

[0008] Although the MOEA / D algorithm has been used to deal with different combinatorial optimization problems, there are few studies on the use of the MOEA / D algorithm to solve the berth allocation and ship loader joint scheduling problems. Summary of the invention

[0009] In view of the problems existing in the prior art, the present invention provides a dry bulk terminal berth allocation and ship loader joint scheduling model and algorithm based on adaptive MOEA / D.

[0010] The present invention provides a method for allocating berths at a dry bulk cargo terminal and coordinating ship loaders based on an adaptive MOEA / D algorithm, comprising the following steps:

[0011] Step 1: Ignore the ship's sailing time and the loader's moving time, and allow multiple ships to enter and exit at the same time;

[0012] Step 2: The ship’s arrival time at the berth is consistent with the operation start time, and the operation end time is consistent with the ship’s departure time;

[0013] Step 3: Establish a ship berth allocation and loader scheduling model to minimize ship waiting time, minimize energy consumption and maximize loader utilization as optimization goals;

[0014] Step 4: Apply the adaptive MOEA / D algorithm to solve the berth allocation and loader scheduling problems.

[0015] Further, the following steps are included:

[0016] The multi-objective optimization problem is decomposed into multiple single-objective problems using the Chebyshev method;

[0017] New solutions are generated using crossover and mutation operators, and solutions in the neighborhood are updated based on an adaptive strategy.

[0018] Furthermore, the calculation model of berth allocation and loader scheduling assumes:

[0019] (1) The two yards store different goods and are at different distances from the berths;

[0020] (2) The operating time and energy consumption of the loader are related to the distance between the yard and the berth and are uncertain.

[0021] Further, the berth allocation interleaving decoding algorithm is adopted, and the steps are as follows:

[0022] (1) Input the ship sequence chromosome to determine the order of ships entering the port;

[0023] (2) Select berth intervals according to the type of ship cargo and allocate berths according to the length of the ship;

[0024] (3) Allocate ship loaders according to the number of cabins to meet single-machine or dual-machine operation conditions.

[0025] Furthermore, the loader allocation strategy is:

[0026] When the number of ship holds is greater than or equal to 4, and the number of idle ship loaders is greater than or equal to 2, double-machine operation is assigned, otherwise single-machine operation is assigned.

[0027] The present invention provides an optimization and updating method based on the MOEA / D algorithm, characterized in that the steps include:

[0028] (1) Calculate the neighbors of the solution and generate new solutions through genetic operators;

[0029] (2) Update the reference point using the Chebyshev method;

[0030] (3) Update the non-dominated solution set of the external group.

[0031] 7. The method according to claim 6, characterized in that the adaptive updating strategy step comprises:

[0032] (1) Calculate the number of neighbors that need to be updated;

[0033] (2) determine the subproblem that is most suitable for updating;

[0034] (3) Update the neighborhood solution set of the individual solution.

[0035] 8. The method according to claim 1 is characterized in that a hybrid crossover operator is used to combine Pmx crossover and Obx crossover to generate new solutions and optimize Pareto solutions.

[0036] The present invention provides a dry bulk terminal berth allocation and ship loader joint scheduling system based on an adaptive MOEA / D algorithm, characterized in that the system comprises:

[0037] Simulation model module, used to establish berth allocation and ship loader scheduling models for dry bulk terminals;

[0038] The optimization calculation module is used to execute the adaptive MOEA / D algorithm to optimize the ship berth allocation and ship loader scheduling;

[0039] Crossover and mutation modules, used to generate new solutions based on Pmx crossover and Obx crossover;

[0040] The neighborhood update module is used to update the reference point based on the Chebyshev method and update the neighborhood of the solution.

[0041] The technical effects and significant technical progress achieved by the present invention are:

[0042] First, the invention proposes a dry bulk terminal berth allocation and ship loader joint scheduling model and algorithm. This method not only allocates berths for ships, but also jointly schedules ship loaders; while completing the basic task of berth allocation, it also optimizes the waiting time of ships, machinery utilization and energy consumption, which can greatly improve the intelligence, efficiency and greenness of the terminal; finally, the algorithm used in the present invention has a smaller algorithm complexity than other heuristic algorithms, and the solution obtained has better convergence, distribution and diversity, and the advanced algorithm is applied to berth allocation.

[0043] By adopting the adaptive MOEA / D algorithm, the system achieves efficient solution to multi-objective scheduling problems and significantly improves the quality and efficiency of scheduling decisions. This not only optimizes the utilization rate of terminal berth resources, but also reduces the waiting time of ships and enhances the comprehensive operation capacity of the terminal. In summary, the present invention has made significant technical progress in the field of dry bulk terminal berth allocation and ship loader joint scheduling.

[0044] Second, the technical solution of the present invention fills the technical gap in the industry at home and abroad:

[0045] 1 Filling the technical gap: This invention fills the technical gap in the field of berth allocation and ship loader joint scheduling at home and abroad. By applying the advanced adaptive MOEA / D algorithm to solve practical problems, it provides an effective tool to solve complex combinatorial optimization problems in this field.

[0046] The present invention applies the adaptive MOEA / D algorithm to the allocation of wharf berths and the scheduling of ship loaders, and improves the MOEA / D algorithm so that the convergence and distribution of the algorithm are better, and the scheduling scheme is obtained faster and better.

[0047] The technical solution of the present invention solves the problem of berth allocation and joint scheduling of loading and unloading machinery at dry bulk terminals that people have been eager to solve: most traditional studies optimize berth allocation and loading and unloading machinery scheduling problems independently, ignoring the importance of joint scheduling. The present invention successfully solves this complex combinatorial optimization problem through a dry bulk terminal berth allocation and ship loader joint scheduling system based on an adaptive MOEA / D algorithm, and realizes the coordinated optimization of berth allocation and ship loader scheduling. The adaptive MOEAD algorithm proposed in the present invention can decompose the multi-objective problem into multiple single-objective sub-problems, greatly reducing the computational complexity. While optimizing the three objectives of ship waiting time, machinery utilization and energy consumption, the algorithm significantly improves the convergence, distribution and diversity of the algorithm, and solves the technical problem that traditional algorithms are difficult to efficiently solve multi-objective combinatorial optimization problems. In the context of green and low-carbon terminals, the present invention not only ensures the shortest time for ships in port, but also takes into account energy-saving green scheduling research. This innovative solution reduces energy consumption and emissions while optimizing terminal operating efficiency, meeting the green and low-carbon requirements of modern port development.

[0048] Third, the technical solution of the present invention solves the following key technical problems in the prior art in industrial applications and has achieved significant technical progress:

[0049] 1. Solve the problem of long waiting time for ships: In the operation of traditional dry bulk terminals, ship berth allocation and ship loader scheduling often rely on manual scheduling or simple rule algorithms, which causes ships to wait for berths or ship loaders for a long time, reducing the operating efficiency of the terminal. The present invention introduces an adaptive MOEA / D algorithm, which can dynamically optimize berth allocation and ship loader scheduling according to the order of ship arrival, cargo type, and the distance between berths and yards, significantly reducing the waiting time of ships and improving the operating efficiency of the terminal.

[0050] 2. Solve the problem of low utilization of ship loaders: The existing scheduling system often fails to effectively allocate ship loaders resources, resulting in some ship loaders being frequently used while other equipment is idle, resulting in uneven utilization. The present invention maximizes the utilization of ship loaders through a multi-objective optimization algorithm, and automatically selects a single-machine or dual-machine operation mode according to the number of ship cabins and the available status of the ship loader, thereby improving the overall equipment utilization.

[0051] 3. Solve the problem of high energy consumption: The traditional scheduling strategy of ship loaders ignores the optimization of equipment energy consumption, resulting in energy waste, especially when the distance between the ship loader and the berth is uncertain. The scheduling system of the present invention takes the energy consumption of the ship loader as an optimization target, and gives priority to the operation path and equipment allocation with the lowest energy consumption during the scheduling process, effectively reducing the energy consumption of the terminal.

[0052] 4. Improve the intelligence and automation level of the system: The existing berth and loader scheduling system has a low intelligence level, relies on manual intervention, and has low efficiency. The present invention uses the adaptive MOEA / D algorithm to decompose the multi-objective problem into multiple single-objective problems, and realizes efficient intelligent scheduling through an adaptive update strategy, reduces manual intervention, and improves the intelligence and automation level of terminal operations.

[0053] In summary, the present invention significantly improves the operating efficiency, equipment utilization and energy efficiency of the terminal by solving the above-mentioned key technical problems, and achieves significant technological progress in industrial applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a flowchart of an algorithm provided by an embodiment of the present invention.

[0055] Figure 2 It is a chromosome diagram of ship sequence integer coding provided by an embodiment of the present invention.

[0056] Figure 3 It is a Pmx cross diagram provided by an embodiment of the present invention.

[0057] Figure 4 It is an Obx cross diagram provided by an embodiment of the present invention.

[0058] Figure 5 The present invention provides a Gantt chart of a ship arrangement plan.

[0059] Figure 6 It is a flow chart of a method for dry bulk terminal berth allocation and ship loader joint scheduling based on adaptive MOEA / D provided by an embodiment of the present invention.

[0060] Figure 7 It is the optimal graph of the Pareto solution of L-01, L-02, L-04, L-04, and L-05 provided in the embodiment of the present invention. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0062] like Figure 7 As shown, a method for dry bulk terminal berth allocation and ship loader joint scheduling based on adaptive MOEA / D provided by an embodiment of the present invention includes:

[0063] S101, inputting a ship schedule through an input module; and performing scheduling using a scheduling algorithm through a scheduling module;

[0064] S102, allocating berths to ships through a berth allocation module; scheduling loading and unloading equipment through a loading and unloading scheduling module;

[0065] S103, using the ship networking system to inform the ship of its arrival at the port through the notification module; and controlling the loading and unloading of equipment through the control module;

[0066] S104, the ship is loaded and unloaded through the loading and unloading operation module; and the ship leaves the port through the port departure module.

[0067] Another object of the present invention is to provide a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the dry bulk terminal berth allocation and loader joint scheduling method based on adaptive MOEA / D.

[0068] The object of the present invention is to provide an information data processing terminal, which is used to implement the dry bulk terminal berth allocation and ship loader joint scheduling system based on adaptive MOEA / D.

[0069] Consider a dry bulk cargo terminal. The berths of the terminal are continuous berths with a limited shoreline length. The number of ship loaders is also limited. The terminal has two yards, which are located in different locations. The ship is waiting to enter the port at the anchorage, and the length, load capacity and other information of the ship are known. The working state of the ship loading and unloading machine is divided into single-machine operation and double-machine operation. If the number of cabins is greater than a fixed value, the ship loading and unloading machine can choose double-line operation. At the same time, consider the impact of the location of the yard on the operation time and energy consumption of the ship loading and unloading machine, as well as the impact of tides on the entry and exit of ships.

[0070]

[0071]

[0072] Assumptions made during the modeling process of the embodiments of the present invention.

[0073] (1) Ignore the moving time of the loader and the sailing time of the ship.

[0074] (2) The time when the ship arrives at the berth is consistent with the start time of the ship's operation, and the time when the ship leaves the berth is consistent with the end time of the ship's operation. The berthing and unberthing time are not taken into account.

[0075] The mathematical model of the joint scheduling problem with the goal of minimizing ship waiting time and energy consumption and maximizing the utilization rate of ship loaders can be expressed as follows.

[0076] f=min(f1,f2,f3) (1)

[0077]

[0078] m it ≤2, if CS i ≥4 (6)

[0079] m it ≤1,if CS i <4 (7)

[0080]

[0081] d i =f i (9)

[0082]

[0083] Formula (1) is the objective function, Formula (2) calculates the time the ship is in port, Formula (3) calculates the mechanical utilization rate of the loader, Formula (4) calculates the energy consumption of the loader, Constraint (5) constrains that the ship's berthing position cannot berth other ships, Constraint (6) constrains that ships with less than 4 cabins can only be equipped with one loader, Constraint (7) constrains that ships with more than or equal to 4 cabins can only be equipped with two loaders, Constraint (8) constrains that the ship's final berthing position must be less than the length of the shoreline, Constraint (9) constrains that the ship needs to leave the port immediately after the ship's operation is completed. Constraint (10) indicates that the total number of loaders allocated to all ships at a certain moment does not exceed the actual number

[0084] The adaptive MOEA / D algorithm is as follows:

[0085] 1) Initialization:

[0086] 1.1) Create an empty set of external groups (EP)

[0087] 1.2) Calculate the Euclidean distance between weight vectors and find the vector to which each vector is closest.

[0088] 1.3) Initial population

[0089] 1.4) Initialize the reference point in the Chebyshev method

[0090] renew:

[0091] 2.1) Randomly select solutions from two fields and obtain new solutions through genetic operators

[0092] 2.2) According to the specific problem, improve and repair the generated new solution, compare the new solution with the optimal solution of the original sub-problem, and keep the best one.

[0093] 2.3) Update the reference point of the Chebyshev method. If the new solution is greater than the reference point, make the new solution equal to the reference point

[0094] 2.4) Update the neighborhood solution

[0095] 2.5) Update the external group (EP) non-dominated solution set

[0096] 3) If the stop condition is met, stop and output EP. Otherwise, go to step 2.

[0097] Figure 1 This is a flowchart of the algorithm.

[0098] The algorithm coding proposes an integer coding of the order of ships entering the port to determine the priority of ships entering the port. A single chromosome is designed to determine the order of ships entering the port. The genes of the chromosome represent the serial numbers of each ship, and the length of the chromosome is equal to the number of ships. An example of a chromosome is shown in the figure. The figure shows a chromosome with a length of 8. The first gene represents the number 1 of the first ship entering the port. The genes from left to right of the chromosome represent the order of ships entering the port. Figure 2 Chromosome encoding for ship order integer.

[0099] In order to solve the joint scheduling problem, this paper designs a berth allocation interpolation decoding algorithm, and the algorithm framework is shown in Algorithm 21. 1. Input chromosomes to obtain the order in which ships enter berths 2. Arrange berths for each ship, determine the type of cargo on board, and select the berth interval for docking 3. Allocate according to the length of the ship. The interval between ships is 104. Assign a loader to each ship. When the number of cabins is greater than or equal to 4 and the idle machines are greater than or equal to 2, double-click the operation, otherwise operate a single machine. 5. Record the time when the ship enters the berth and the operation time of the loader, and calculate the departure time 6. Until all ships are assigned.

[0100]

[0101] The MOEA / D algorithm decomposes the multi-objective problem into multiple single-objective problems. Each sub-problem has a weight vector, and the simplex method is used to generate a relatively uniform weight vector. There are three decomposition methods: the first is the linear weighted method, the second is the Chebyshev method, and the third is the L p The algorithm in this paper uses the Chebyshev method to decompose the problem. The uniformly distributed weight vector is shown in formula (13).

[0102]

[0103] This paper proposes a hybrid crossover operator, which is to mix Pmx crossover and Obx crossover. The idea of ​​mixing is to use different crossover operators to generate new solutions when the crossover probability is met, and then get a good solution through Chebyshev aggregation. Figure 3 , 4 as follows.

[0104]

[0105]

[0106] The adaptive update strategy is as follows:

[0107] Step 1: Calculate the number of neighbors Cu that need to be updated according to formula (16).

[0108]

[0109] where C is the number of neighbors, k is the current iteration, K is the maximum iteration, and γ∈[0,1) is a control parameter.

[0110] Step 2: Find the most suitable subproblem j. If a subproblem satisfies equation (17), then the subproblem is defined as the most suitable subproblem.

[0111] j = arg 1≤k≤N min(g k (x)) (13)

[0112] Step 3: Update individual x i Neighbors of x j Every solution x in the neighborhood k , if equation (18) is satisfied, then x k Replace with

[0113]

[0114] Take 20 ships and 4 loaders as an example. Figure 5 The Gantt chart of the ship arrangement plan shown. The horizontal axis is time, the vertical axis is the working status of each loading and unloading machine, and each rectangular block of different colors represents a different ship. For example, if there are two rectangles of the same color on the vertical axis in the same time period, it means that the ship is performing dual-line operations.

[0115] This method takes maximizing machinery utilization and minimizing machinery energy consumption as the optimization goals, so the resulting schedule not only takes into account the minimum time the ship spends in port, but also maximizes machinery utilization and minimizes energy consumption.

[0116] By implementing the above two specific embodiments, the dry bulk terminal berth allocation and loader joint scheduling system based on adaptive MOEA / D can significantly improve the terminal's operating efficiency, reduce the ship's time in port, and improve the accuracy and safety of loading and unloading operations, thereby optimizing the utilization of terminal resources.

[0117] In order to verify the effectiveness of the technical solution of the present invention, we took a large dry bulk cargo terminal as an example and implemented the berth allocation and loader joint scheduling system of the present invention.

[0118] The dry bulk terminal has an annual throughput of more than 20 million tons, 10 berths and 20 ship loaders. Under the traditional operation mode, due to the lack of an efficient dispatching system, the terminal often encounters problems such as long waiting times for ships, low utilization of loading and unloading equipment, and high energy consumption, which seriously affects the operational efficiency and economic benefits of the terminal.

[0119] After the dispatching system of the present invention is applied to the terminal, the schedule of ships arriving at the port within a week is obtained through the input module, and the adaptive MOEA / D algorithm module is used to perform joint optimization calculations on ship berth allocation and ship loader scheduling.

[0120] After the system dispatch, the average waiting time of ships was shortened by 35%, the utilization rate of ship loaders increased by 22%, and the total energy consumption of the terminal decreased by 18%. This has greatly improved the intelligence, efficiency and greenness of the terminal, and significantly improved the comprehensive operational efficiency and economic benefits of the terminal.

[0121] This application example fully demonstrates that the technical solution of the present invention not only solves the technical difficulty that traditional algorithms are difficult to efficiently solve multi-objective combined optimization problems, but also fills the technical gap in the field of berth allocation and joint scheduling of ship loaders at home and abroad, and has significant creativity and application value.

[0122] The embodiments of the present invention have achieved some positive effects during the development or use process, and indeed have great advantages over the prior art. The following content is described in conjunction with data, charts, etc. of the test process.

[0123] 1. Example Data

[0124] All algorithms are compiled and run on Intel(R) Core(TM) I7-12650H, 4.80GHz CPU, 16.00GB RAM, Win10 64-bit operating system, Python programming language and Pycharm programming environment.

[0125] The experimental data is generated by modifying the random generation method. Five instances are generated according to the settings in Table 1, and the instances have 20 ships (Instances L-01-L-05). In order to verify the effect of adaptive MOEA / D, this study will compare NSGA-II and MOEA / D. The setting parameters are shown in Table 1.

[0126] Table 1

[0127]

[0128] The parameter settings of population size N, domain size T, adjustment parameter γ, crossover probability pc, and mutation probability pm are shown in Table 2.

[0129] Table 2

[0130]

[0131] By comparing the Pareto solutions of AMOEA / D, MOEA / D and NSGA-II, the generated data scales L-01 to L-05 are as follows. The scatter points in the figure are Pareto solutions, the green scatter points are the solutions of AMOEA / D, the orange scatter points are the solutions of MOEA / D, and the blue scatter points are the solutions of NSGA-II. The three coordinate axes represent the ship waiting time, the operation rate of machinery, and the energy consumption. Since the goal of the algorithm is to minimize the ship waiting time, maximize the operation rate of machinery, and minimize the energy consumption, the scatter points closer to the origin are the best solutions. As can be seen from the figure, the Pareto solutions of L-01, L-05, and AMOEA / D are the best.

[0132] 1.1. Example results

[0133] In order to verify whether the proposed AMOEA / D algorithm can effectively solve the BASP problem, this section compares the AMOEA / D algorithm with the non-dominated sorting genetic algorithm-II (NSGA-II) and the decomposition-based multi-objective optimization algorithm (MOEA / D). We use the example mentioned at the beginning of Section 4 as experimental data. The hypervolume (HV), generation distance (GD) and inverse generation distance (IGD) indicators in the multi-objective optimization problem are used for evaluation. The HV indicator can be used to simultaneously evaluate the convergence and distribution of Pareto solutions. The larger the value, the better the performance of the algorithm. The GD indicator can measure the degree of approximation between the Pareto frontier and the optimal Pareto frontier. The smaller the GD value, the better the algorithm performance. The IGD indicator is more comprehensive than GD and can simultaneously evaluate the convergence of the Pareto frontier and the diversity of Pareto solutions. The smaller the IGD value, the better the algorithm performance. These three indicators can be calculated by formula (1): (16)-(18). The results of the HV, GD and IGD values ​​of the three algorithms are shown in the table. The Pareto solutions are as follows Figure 7 As shown in Figure 2, HV, solution diversity and solution distribution are calculated for AMOEA / D, MOEA / D and NAGA-II respectively to evaluate the algorithm performance of AMOEA / D.

[0134]

[0135] in Represents three goals, is the reference point, and n is the number of Pareto solutions. i is the Euclidean distance between solution i and the nearest Pareto solution.

[0136] The table shows the statistical results of HV, GD and IGD of the three algorithms in instances L-1 to L-5. It can be seen that the AMOEA / D algorithm performs best in L-1 and L-5. At the same time, AMOEA / D is better than the MOEA / D algorithm in L-1, L-2, L-4, and L5. The IGD index of AMOEA / D performs better than the other two algorithms in all instances, indicating that the solutions obtained by the algorithm have good convergence and diversity.

[0137] Table 3

[0138]

[0139]

[0140] The following are two industrial application examples:

[0141] Example 1: Intelligent scheduling of berths and ship loaders at a large dry bulk cargo terminal

[0142] In this embodiment, a large dry bulk cargo terminal has an annual throughput of more than 30 million tons, with 15 berths and 25 ship loaders. The traditional manual scheduling method leads to problems such as long waiting time for ships, low berth utilization, and high energy consumption. By introducing a berth allocation and ship loader joint scheduling system based on the adaptive MOEA / D algorithm, the terminal realizes automatic berth allocation for ships and intelligent scheduling of ship loaders. The system optimizes the allocation of berths and ship loaders based on the real-time ship entry situation and the distribution of cargo in the yard, reduces the waiting time of ships in the port, improves the utilization rate of ship loaders, and reduces energy consumption. After applying the system, the average waiting time of ships at the terminal was shortened by 40%, the utilization rate of ship loaders was increased by 25%, and the total energy consumption of the terminal decreased by 15%.

[0143] Example 2: Multi-objective berth scheduling system for medium-sized ports

[0144] In this embodiment, a medium-sized port has 10 berths and 10 loading and unloading equipment, which are mainly used to handle dry bulk cargoes such as coal and ore. The port solves the problems of high energy consumption of loading and unloading equipment and uneven berth utilization by applying a joint scheduling system with an adaptive MOEA / D algorithm, combining the size of the ship, the type of cargo, and the distance between the berth and the yard. The system can automatically schedule berths and ship loaders for different ships, and balance the utilization rate and energy consumption of the equipment through a multi-objective optimization algorithm. After the system was launched, the port achieved a 20% reduction in equipment energy consumption, a 30% reduction in the time ships spend in port, and a significant improvement in the utilization rate of berths and ship loaders.

Claims

1. A method for berth allocation and ship loader joint scheduling of dry bulk cargo terminals based on adaptive MOEA / D algorithm, characterized in that: The following steps are involved: Step 1: Ignore the ship's sailing time and the loader's moving time, and allow multiple ships to enter and exit at the same time; Step 2: The ship’s arrival time at the berth is consistent with the operation start time, and the operation end time is consistent with the ship’s departure time; Step 3: Establish a ship berth allocation and loader scheduling model to minimize ship waiting time, minimize energy consumption and maximize loader utilization as optimization goals; Step 4: Apply the adaptive MOEA / D algorithm to solve the berth allocation and loader scheduling problems.

2. The method according to claim 1, characterized in that Further comprising the steps of: The multi-objective optimization problem is decomposed into multiple single-objective problems using the Chebyshev method; New solutions are generated using crossover and mutation operators, and solutions in the neighborhood are updated based on an adaptive strategy.

3. The method according to claim 1, characterized in that The calculation model of berth allocation and loader scheduling assumes: (1) The two yards store different goods and are at different distances from the berths; (2) The operating time and energy consumption of the loader are related to the distance between the yard and the berth and are uncertain.

4. The method according to claim 1, characterized in that: The berth allocation interleaving decoding algorithm is adopted, and the steps are as follows: (1) Input the ship sequence chromosome to determine the order of ships entering the port; (2) Select berth intervals according to the type of ship cargo and allocate berths according to the length of the ship; (3) Allocate ship loaders according to the number of cabins to meet single-machine or dual-machine operation conditions.

5. The method according to claim 4, characterized in that The loader allocation strategy is: When the number of ship holds is greater than or equal to 4, and the number of idle ship loaders is greater than or equal to 2, double-machine operation is assigned, otherwise single-machine operation is assigned.

6. An optimization update method based on MOEA / D algorithm, characterized in that the steps include: (1) Calculate the neighbors of the solution and generate new solutions through genetic operators; (2) Update the reference point using the Chebyshev method; (3) Update the non-dominated solution set of the external group.

7. The method according to claim 6, characterized in that The adaptive update strategy steps include: (1) Calculate the number of neighbors that need to be updated; (2) determine the subproblem that is most suitable for updating; (3) Update the neighborhood solution set of the individual solution.

8. The method according to claim 1, characterized in that A hybrid crossover operator is used to combine Pmx crossover with Obx crossover to generate new solutions and optimize the Pareto solution.

9. A dry bulk terminal berth allocation and ship loader joint scheduling system based on adaptive MOEA / D algorithm, characterized in that: The system includes: Simulation model module, used to establish berth allocation and ship loader scheduling models for dry bulk terminals; The optimization calculation module is used to execute the adaptive MOEA / D algorithm to optimize the ship berth allocation and ship loader scheduling; Crossover and mutation modules, used to generate new solutions based on Pmx crossover and Obx crossover; The neighborhood update module is used to update the reference point based on the Chebyshev method and update the neighborhood of the solution.

10. A multi-objective optimization system for berth allocation and ship loader scheduling of dry bulk cargo terminals, characterized in that: The system includes: The ship management module is used to receive the order of ships entering the port and establish berth allocation and ship loader scheduling models; The computation engine module, based on the adaptive MOEA / D algorithm, uses the Chebyshev method to decompose the multi-objective problem; Ship loader allocation module, used to automatically allocate single or double machine operation according to the number of ship cabins; The result output module is used to output the optimized berth allocation and loader scheduling results.

Images