Dry bulk cargo river-sea combined transportation seagoing ship and river ship direct barge scheduling optimization method

Abstract

The invention relates to a dry bulk river-sea combined transportation seagoing ship and river ship direct barge scheduling optimization method, which comprises the following steps: S1, constructing a river-sea combined transportation direct barge scheduling model, and defining and setting related parameters and variables; s2, constructing a target function by taking the total time of all ships at the port and the total operation cost of carriers as targets, and constructing a berthing position constraint, an equipment operation constraint, a river-sea combined transportation connection constraint and a time connection constraint as constraint conditions; and S3, solving the river-sea combined transportation direct barge scheduling model by adopting a heuristic genetic algorithm to obtain a scheduling scheme, and performing scheduling based on the scheduling scheme. Compared with the prior art, an optimization scheme is provided for a dry bulk cargo river-sea combined transport ship-port object integrated dispatching mode, and the logistics transportation cost is reduced to the maximum extent.

Description

Technical Field

[0001] This invention relates to the field of transportation, and in particular to an optimized scheduling method for direct transshipment between ocean-going vessels and river vessels in dry bulk cargo river-sea intermodal transport. Background Technology

[0002] Dry bulk shipping is a crucial component of international maritime trade, accounting for more than one-third of global seaborne freight volume. Characterized by its large volume and long distances, it is highly dependent on waterway transport. Currently, common multimodal transport methods for dry bulk cargo include road-water intermodal transport, rail-water intermodal transport, and river-sea intermodal transport. Among these, river-sea intermodal transport, by fully utilizing the advantages of inland waterway transport and achieving seamless integration in loading, unloading, and transshipment, significantly reduces transshipment costs and transport time, thus becoming a more efficient mode of transport.

[0003] River-sea intermodal transport refers to a transport organization method that, based on the characteristics of cargo transport and combining the conditions of both inland waterways and oceans, adopts a highly efficient transshipment method between river vessels and sea vessels at transit hub ports, thereby achieving a highly efficient river-sea cargo transport system.

[0004] Current dry bulk cargo transshipment operations via river-sea intermodal transport face multiple practical challenges. The business processes involve collaboration among multiple stakeholders—ships, ports, and cargo—leading to delays in operational planning and independent operation of data systems across different transport segments, resulting in inefficient cross-system data exchange. During terminal loading and unloading operations, differences in tonnage and cargo hold structure between river vessels and seagoing vessels easily lead to issues such as insufficient equipment compatibility and inefficient workflow coordination in dry bulk cargo transshipment. Furthermore, traditional operational models rely on manual scheduling with low levels of automation, resulting in prolonged cargo dwell time at port and hindering the full utilization of mechanized operations. These problems manifest as insufficient integrated terminal management capabilities, a lack of coordination mechanisms among multiple transport modes, and limited adoption of intelligent operation scenarios. These issues directly impact the seamless integration and overall efficiency improvement of river-sea intermodal transport, becoming key factors restricting the improvement of port capacity and the optimization of logistics costs.

[0005] While research on integrated river-sea intermodal transport scheduling for dry bulk cargo has yielded some results both domestically and internationally, the following shortcomings still exist:

[0006] (1) Existing research on river-sea intermodal transport mainly focuses on port system construction, competitive advantage enhancement, resource coordination and transportation scheme optimization, but lacks research on the optimization of dry bulk cargo river-sea intermodal transport scheduling mode.

[0007] (2) Existing research on berth scheduling optimization generally adopts a single-objective model, while research on multi-objective ship scheduling mainly involves container ports. For bulk cargo ports with uncertain ship arrival times and complex loading and unloading processes, how to establish a multi-objective equilibrium optimization scheduling model under comprehensive multi-factor conditions still needs in-depth exploration.

[0008] (3) Current research on collaborative scheduling of loading and unloading equipment at dry bulk terminals mainly focuses on equipment operation optimization and system design, using methods such as mixed integer programming and dynamic programming to coordinate the scheduling of ship loaders, stacker-reclaimers and belt conveyors. Existing results are mostly focused on single-item scheduling and lack integrated scheduling strategies with ships, ports and cargo.

[0009] The aforementioned technical problems urgently need to be solved. Summary of the Invention

[0010] The purpose of this invention is to provide an optimized scheduling method for direct transshipment between seagoing vessels and river vessels in dry bulk river-sea intermodal transport, in order to solve the problems existing in the prior art.

[0011] The objective of this invention can be achieved through the following technical solutions:

[0012] An optimization method for direct transshipment scheduling between seagoing vessels and river vessels in dry bulk river-sea intermodal transport includes:

[0013] Step S1: Construct a river-sea intermodal direct transport scheduling model and define and set relevant parameters and variables;

[0014] Step S2: Construct an objective function with the total time all vessels spend in port and the total operating cost of the carrier as the objective, and construct berthing position constraints, equipment operation constraints, river-sea intermodal transport connection constraints, and time connection constraints as constraint conditions;

[0015] Step S3: Use a heuristic genetic algorithm to solve the river-sea intermodal direct transport scheduling model to obtain a scheduling scheme, and then perform scheduling based on the scheduling scheme.

[0016] Step S1 includes:

[0017] Step S1-1: Assume the river-sea intermodal direct transport scheduling model contains n evaluation indicators and a total of m operational modes to be evaluated. Establish the original evaluation matrix as R = [r ij ] m×n , where r ij Let be the evaluation value of the i-th evaluation indicator for the i-th operation mode;

[0018] Step S1-2: Divide the evaluation indicators into positive and negative indicators. The standardization process for positive indicators is as follows:

[0019]

[0020] The standardization process for negative indicators is as follows:

[0021]

[0022] Where: s ij The standardized evaluation index value is min(r).j ) represents the minimum index value for the j-th operation mode, max(r) j ) represents the maximum index value for the j-th operation mode;

[0023] Step S1-3: Determine the objective weight of each indicator based on the amount of information contained in each indicator:

[0024]

[0025] Where: ω j V represents the objective weight of the j-th indicator. j Let σ be the amount of information contained in the j-th indicator. j p is the standard deviation of the j-th indicator. ij To evaluate the correlation coefficient between the i-th and j-th indicators;

[0026] Step S1-4: Construct a weighted standardization matrix, where each element in the weighted standardization matrix is ​​the product of the standardized evaluation index value and the objective weight of the corresponding index.

[0027] Step S1-5: Determine the optimal and worst ideal solutions for the evaluation index as follows:

[0028]

[0029] Among them: A j + For the optimal ideal solution of the evaluation index, A j - The worst-case ideal solution for the evaluation index. For the first index in the optimal ideal solution A j + The maximum or minimum value in, For the second index in the optimal ideal solution A j + The maximum or minimum value in, For the nth index in the optimal ideal solution A j + The maximum or minimum value in, For the first index in the worst ideal solution A j - The maximum or minimum value in, For the second index in the worst ideal solution A j - The maximum or minimum value in, For the nth index in the worst ideal solution A j - The maximum or minimum value in;

[0030] When the evaluation index is a positive index, the optimal and worst ideal solutions for the evaluation index are:

[0031]

[0032] When the evaluation index is negative, the optimal and worst ideal solutions for the evaluation index are:

[0033]

[0034] Where: e ij These are the elements in the weighted standardized matrix;

[0035] Step S1-6: Calculate the Euclidean distance from each job mode to be evaluated to the optimal and worst ideal solutions:

[0036]

[0037] in: To evaluate the Euclidean distance from job mode i to the optimal ideal solution. To evaluate the Euclidean distance from operation mode i to the worst ideal solution;

[0038] Step S1-7: Calculate the relative proximity:

[0039]

[0040] Where: Z i To evaluate the relative closeness of each evaluation indicator under operation mode i.

[0041] The total time all vessels spent in port is:

[0042]

[0043] Where: f1 is the total time all vessels spend in port, qe i Let q be the time when ship i begins to depart from its berth. i For the arrival time of seagoing vessel i, ke j k is the starting departure time of riverboat j. j Let I be the arrival time of riverboat j, I be the assembly time of ocean-going vessels, and J be the assembly time of riverboats.

[0044] The carrier's total operating costs are:

[0045] f2=C1+C2

[0046] Where: f2 is the carrier's total operating cost, C1 is the sum of the ship's port detention cost and the cost of using the loading and unloading machinery, and C2 is the conveyor belt transportation cost.

[0047] Step S3 specifically includes:

[0048] Step S3-1: Serialize the vessels according to their arrival time sequence in the river-sea intermodal transport;

[0049] Step S3-2: First decode the ship's berthing position and the ship's loading and unloading machine assignment, and then decode the ship's berthing time.

[0050] Step S3-3: Based on the estimated arrival time of the river-sea intermodal vessels, construct an initial solution sequence with practical scheduling significance;

[0051] Step S3-4: Select the fitness function F c :

[0052]

[0053] Where: f c η represents the objective function value for an individual, and η is a parameter reflecting the problem size.

[0054] Step S3-5: Perform genetic operator operations to obtain a scheduling scheme, and perform scheduling based on the scheduling scheme. The genetic operators specifically include selection operators, crossover operators, and mutation operators.

[0055] The process of decoding the ship's berthing position and the ship's loading / unloading machine assignment in step S3-2 includes:

[0056] Step S3-2-1-1: Following the principle of minimizing relative berthing distance, after determining the berth for seagoing vessel i, iterate through the candidate berths for riverboat j and select the one that satisfies the minimum requirement. ij The berth combination, in which, mind ij This refers to the minimum relative distance between seagoing vessel i and riverboat j when they are moored.

[0057] Step S3-2-1-2: Following the principle of allocating as many as possible, when the berth loading and unloading machines are idle, allocate the maximum number of idle loading and unloading machines to each ship without exceeding the upper limit of equipment per ship, thereby maximizing equipment utilization.

[0058] The decoding of the ship's berthing time in step S3-2 includes:

[0059] Step S3-2-2-1: When the berth is available and the seagoing vessel is ready, the berthing time is determined by the arrival time of the seagoing vessel and the river vessel.

[0060] Step S3-2-2-2: When the berth is unavailable, i.e., the previous river vessel is being served, the berthing time is determined by the departure time of the previous river vessel and the arrival time of this river vessel.

[0061] The initial solution generation process in step S3-3 includes:

[0062] Step S3-3-1: Read the estimated arrival times of all vessels;

[0063] Step S3-3-2: Sort the vessel group in ascending order of arrival time;

[0064] Step S3-3-3: Based on the sorting results, record the ship number and generate the initial individual;

[0065] Step S3-3-4: Use multiple sequences formed based on sorting perturbation as the initial population to enhance the diversity of the algorithm.

[0066] The selection operator uses the tournament selection method, which randomly samples s individuals from the population and selects the best individual to enter the next generation.

[0067] The execution process of the crossover operator includes:

[0068] Step S3-5-1-1: Select two individuals with different chromosomes from the population selected through the tournament, denoted as parent A and parent B;

[0069] Step S3-5-1-2: Randomly select two intersection points pos1 and pos2 in parent generation A, and extract the gene segment in parent generation A located in the interval [pos1,pos2].

[0070] Step S3-5-1-3: Copy the gene segment of parent A directly to the same position in the offspring, preserving its local temporal characteristics; fill in the remaining genes according to the original order in parent B, skipping the ship numbers already present in the offspring, to ensure that the chromosomes are legally arranged;

[0071] Step S3-5-1-4: Verify whether the offspring meets the ship uniqueness constraint. If a conflict arises due to crossover, initiate the repair mechanism.

[0072] The execution process of the mutation operator includes:

[0073] Step S3-5-2-1: Randomly select an individual with a specific chromosome from the population;

[0074] Step S3-5-2-2: Randomly select two different positions in the chromosome sequence;

[0075] Step S3-5-2-3: Completely reverse the gene sequence between the two positions to generate a new individual.

[0076] Compared with existing technologies, this invention has the following advantages: This application constructs a multi-objective river-sea intermodal direct transport scheduling optimization model with the objectives of minimizing the total port time of all vessels and minimizing the total operating cost of carriers, and designs a heuristic rule-based genetic algorithm to solve the model. By applying this algorithm to calculate actual case data of a port, a complete river-sea intermodal transport scheduling scheme and corresponding operating cost data within the scheduling cycle are obtained; at the same time, for the scheduling optimization model, the model solution results under different objective function weights are analyzed and compared, verifying the effectiveness of the model and providing technical support for river-sea intermodal transport scheduling decisions at port dry bulk cargo terminals. This invention can realize integrated scheduling of ships, ports, and cargo, providing technical support for the integrated management of "sea-ships-river-ships" at port terminals, optimizing the scheduling of ship berths and loading and unloading equipment at dry bulk cargo terminals, overcoming the technical bottleneck of insufficient connection between various operational links such as river-sea intermodal berth scheduling optimization and loading and unloading equipment allocation, realizing the organic synergy of "water-to-water" operations, and minimizing logistics and transportation costs. Attached Figure Description

[0077] Figure 1 This is a scene diagram of river-sea intermodal transport operations at a breakwater terminal for dry bulk cargo, as described in this application.

[0078] Figure 2 This is the timeline diagram for the direct transport scheduling of "seagoing vessels to river vessels" in this application;

[0079] Figure 3 Assign a two-dimensional spatiotemporal diagram to the berths in this application;

[0080] Figure 4 This is a schematic diagram of the improved sequential crossover operator in this application;

[0081] Figure 5 This is a schematic diagram of the inversion variation of the variation operator in this application;

[0082] Figure 6 This is a flowchart of the heuristic genetic algorithm solution described in step S3 of this application;

[0083] Figure 7 This is an iterative convergence graph of the algorithm in the embodiments of this application;

[0084] Figure 8 This is a two-dimensional coordinate diagram of the location-time of the berth allocation for seagoing vessels in the embodiments of this application;

[0085] Figure 9 This is a two-dimensional coordinate diagram of the location-time of the riverboat berth allocation in the embodiments of this application;

[0086] Figure 10 This is a flowchart illustrating the main steps of the method described in this application. Detailed Implementation

[0087] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. These embodiments are based on the technical solution of the present invention and provide detailed implementation methods and specific operating procedures. However, the scope of protection of the present invention is not limited to the following embodiments.

[0088] like Figure 1 The diagram illustrates a scenario for sea-river intermodal transport at a dry bulk cargo breakwater terminal, as described in this application. The port authority schedules terminal resources according to the sea-ship / river-ship matching plan on the dry bulk cargo sea-river intermodal transport information platform. This involves allocating berths and unloading machine resources to each sea-ship and berths and loading machine resources to the river-ships in the matching plan. When a sea-ship berths at the terminal, its matched river-ship berths. At this point, the unloading machine unloads the dry bulk cargo onto a conveyor belt and directly transfers it to the corresponding river-ship berth for loading, achieving direct transshipment between sea-ship and river-ship.

[0089] In bulk cargo ports, seagoing vessels are generally larger, ranging from 100,000 to 200,000 tons. River vessels, on the other hand, are typically 3,000, 5,000, or 10,000 tons. Therefore, in dry bulk river-sea intermodal transport, one seagoing vessel often needs to be matched with multiple river vessels. Bulk cargo ports have limited berth resources, and due to limitations in bulk cargo loading and unloading efficiency and vessel tonnage, in many ports that primarily serve bulk cargo vessels, seagoing vessels do not enter or leave the port frequently and generally arrive on time within the planned period. Unlike the more organized schedules of container river vessels, dry bulk river vessels are often numerous and have unpredictable arrival times.

[0090] like Figure 2 As shown, this represents the ideal state of seamless connection between seagoing vessels and riverboats, enabling the "direct loading and unloading" effect of "water crossing water".

[0091] like Figure 3 As shown, the berth model can be represented using a two-dimensional coordinate graph. In this graph, the X-axis represents the length of the quayfront, and the Y-axis represents the time a vessel spends in port. This coordinate graph is then used to construct the berth model. Figure 3 In the model, there are three ships: Ship 1, Ship 2, and Ship 3. Each ship is represented by a rectangle. On the coordinate axis, t1 represents the berthing time of Ship 1, t2 represents the departure time of Ship 1, and (p1, p2) is the berthing position range of Ship 1 at the quayhead. In the model, the midpoint of the horizontal distance between the ships is used to define the berthing position of the ships at the quayhead; that is, the berthing position coordinates of Ship 1 are (p1+p2) / 2. Similarly, the berthing position coordinates of Ship 2 and Ship 3 can be obtained.

[0092] Based on this, this application provides an optimization method for direct transshipment scheduling between seagoing vessels and river vessels in dry bulk river-sea intermodal transport, such as... Figure 10 As shown, it includes:

[0093] Step S1: Construct a river-sea intermodal direct transport scheduling model and define and set relevant parameters and variables;

[0094] Specifically, the CRITIC-TOPSIS method is used, which is a method that combines the CRITIC method and TOPSIS, including:

[0095] Step S1-1: Assume the river-sea intermodal direct transport scheduling model contains n evaluation indicators and a total of m operational modes to be evaluated. Establish the original evaluation matrix as R = [r ij ] m×n , where r ij Let be the evaluation value of the i-th evaluation indicator for the i-th operation mode;

[0096] Step S1-2: Divide the evaluation indicators into positive and negative indicators. The standardization process for positive indicators is as follows:

[0097]

[0098] The standardization process for negative indicators is as follows:

[0099]

[0100] Where: s ij The standardized evaluation index value is min(r). j ) represents the minimum index value for the j-th operation mode, max(r) j ) represents the maximum index value for the j-th operation mode;

[0101] Step S1-3: Determine the objective weight of each indicator based on the amount of information contained in each indicator:

[0102]

[0103] Where: ω j V represents the objective weight of the j-th indicator. j Let σ be the amount of information contained in the j-th indicator. j p is the standard deviation of the j-th indicator. ij To evaluate the correlation coefficient between the i-th and j-th indicators;

[0104] Step S1-4: Construct a weighted standardization matrix, where each element in the weighted standardization matrix is ​​the product of the standardized evaluation index value and the objective weight of the corresponding index.

[0105] Step S1-5: Determine the optimal and worst ideal solutions for the evaluation index as follows:

[0106]

[0107] Among them: A j + For the optimal ideal solution of the evaluation index, A j - The worst-case ideal solution for the evaluation index. For the first index in the optimal ideal solution A j + The maximum or minimum value in, For the second index in the optimal ideal solution A j + The maximum or minimum value in, For the nth index in the optimal ideal solution A j + The maximum or minimum value in, For the first index in the worst ideal solution A j - The maximum or minimum value in, For the second index in the worst ideal solution A j - The maximum or minimum value in, For the nth index in the worst ideal solution A j - The maximum or minimum value in;

[0108] When the evaluation index is a positive index, the optimal and worst ideal solutions for the evaluation index are:

[0109]

[0110] When the evaluation index is negative, the optimal and worst ideal solutions for the evaluation index are:

[0111]

[0112] Where: e ij These are the elements in the weighted standardized matrix;

[0113] Step S1-6: Calculate the Euclidean distance from each job mode to be evaluated to the optimal and worst ideal solutions:

[0114]

[0115]

[0116] in: To evaluate the Euclidean distance from job mode i to the optimal ideal solution. To evaluate the Euclidean distance from operation mode i to the worst ideal solution;

[0117] Step S1-7: Calculate the relative proximity:

[0118]

[0119] Where: Z i To evaluate the relative closeness of each evaluation indicator under operation mode i.

[0120] The definition and setting of relevant parameters and variables include:

[0121] (1) Set

[0122] I: represents the set of ships, i∈I, in It is the total number of all seagoing vessels arriving at the port;

[0123] J: represents the set of riverboats, j∈J, in This represents the total number of all matching riverboats;

[0124] W: represents the set of sea-side unloading machines, w∈W in This represents the total number of ship unloaders on the sea side;

[0125] Z: Represents the set of ship loaders on the river side, z∈Z in This represents the total number of ship loaders on the river side;

[0126] T: represents the set of discretized time periods, t∈T, t=(1,2,…,,H).

[0127] (2) Relevant parameters:

[0128] q i : Represents the arrival time of ship i, i∈I;

[0129] k j : indicates the arrival time of riverboat j, j∈J;

[0130] ql i : Represents the length of ship i, including the safe distance between ships, i∈I;

[0131] kl j : Represents the length of the riverboat j, including the safe distance between boats, j∈J;

[0132] (3) Auxiliary variables

[0133] qs i : Represents the starting berthing time of each ship i, i∈I;

[0134] ks j : Indicates the starting berthing time of each riverboat j, j∈J;

[0135] qei : Represents the starting departure time of each ship i, i∈I;

[0136] ke j : Indicates the starting departure time of each riverboat j, j∈J;

[0137] qb i : Indicates the berthing position of each ship i, i∈I;

[0138] kb j : indicates the berthing position of each riverboat j, j∈J.

[0139] (4) Decision variables

[0140] qx ii’ Let qx be a 0-1 variable. In a berth-time two-dimensional coordinate graph, if ship i is completely to the left of ship i', then from the perspective of the dock, ship i is completely to the left of ship i'. ii’ =1, otherwise the value is 0, i∈I, i'∈I;

[0141] kx jj’ Let kx be a 0-1 variable. In a berth-time two-dimensional coordinate graph, if riverboat j is completely to the left of riverboat j', then from the perspective of the dock, riverboat j is completely to the left of riverboat j'. jj’ =1, otherwise the value is 0, j∈J, j'∈J.

[0142] Step S2: Construct an objective function with the total time all vessels spend in port and the total operating cost of the carrier as the objective, and construct berthing position constraints, equipment operation constraints, river-sea intermodal transport connection constraints, and time connection constraints as constraint conditions;

[0143] The total time all vessels spent in port was:

[0144]

[0145] Where: f1 is the total time all vessels spend in port, qe i Let q be the time when ship i begins to depart from its berth. i For the arrival time of seagoing vessel i, ke j k is the starting departure time of riverboat j. j Let I be the arrival time of riverboat j, I be the assembly time of ocean-going vessels, and J be the assembly time of riverboats.

[0146] The carrier's total operating costs are:

[0147] f2=C1+C2

[0148] Where: f2 is the carrier's total operating cost, C1 is the sum of the ship's port detention cost and the cost of using the loading and unloading machinery, and C2 is the conveyor belt transportation cost.

[0149] The objectives f1 and f2 of the dry bulk cargo "sea-to-river vessel" direct transport scheduling optimization model have different dimensions. Directly weighting them would lead to the optimization result being biased towards the objective with the larger dimension. To eliminate the influence of dimensions, this application uses normalization before weighting. Since this application is a minimization problem of the total port time of all vessels and the total operating cost of the carrier, the objective function f of the optimization model is expressed as follows:

[0150] minf = λf1' + (1-λ)f2'

[0151] Where: λ is the weighting coefficient, f1' is the normalized standard value of the total time all ships spend in port, and f2' is the normalized standard value of the carrier's total operating cost.

[0152] Step S3: Use a heuristic genetic algorithm to solve the river-sea intermodal direct transport scheduling model to obtain a scheduling scheme, and perform scheduling based on the scheduling scheme. Specifically, this includes:

[0153] Step S3-1: Unify the coding according to the arrival time sequence of the river-sea intermodal vessels. Let the total number of river-sea intermodal vessels be... River-sea intermodal transport vessels are coded according to their arrival time. As shown in Table 1:

[0154] Table 1

[0155]

[0156] Step S3-2: First decode the ship's berthing position and the ship's loading and unloading machine assignment, and then decode the ship's berthing time.

[0157] Specifically, the process of decoding the ship's berthing position and the assignment of ship loading and unloading machinery includes:

[0158] Step S3-2-1-1: Following the principle of minimizing relative berthing distance, after determining the berth for seagoing vessel i, iterate through the candidate berths for riverboat j and select the one that satisfies the minimum requirement. ij The berth combination, in which, mind ij This refers to the minimum relative distance between seagoing vessel i and riverboat j when they are moored.

[0159] Step S3-2-1-2: Following the principle of allocating as many as possible, when the berth loading and unloading machines are idle, allocate the maximum number of idle loading and unloading machines to each ship without exceeding the upper limit of equipment per ship, thereby maximizing equipment utilization.

[0160] Decoding ship berthing time includes:

[0161] Step S3-2-2-1: When the berth is available and the seagoing vessel is ready, the berthing time is determined by the arrival time of the seagoing vessel and the river vessel.

[0162] Step S3-2-2-2: When the berth is unavailable, i.e., the previous river vessel is being served, the berthing time is determined by the departure time of the previous river vessel and the arrival time of this river vessel.

[0163] Step S3-3: Based on the estimated arrival times of river-sea intermodal vessels, construct an initial solution sequence with practical scheduling significance. The initial solution generation process includes:

[0164] Step S3-3-1: Read the estimated arrival times of all vessels;

[0165] Step S3-3-2: Sort the vessel group in ascending order of arrival time;

[0166] Step S3-3-3: Based on the sorting results, record the ship number and generate the initial individual;

[0167] Step S3-3-4: Use multiple sequences formed based on sorting perturbation as the initial population to enhance the diversity of the algorithm.

[0168] Step S3-4: Select the fitness function F c To reduce the number of local optima and increase the probability of global convergence, this application adopts an improved fitness function based on the model's objective function. This function not only ensures that the fitness value is non-negative but also effectively enhances the algorithm's global search capability. Its expression is as follows:

[0169]

[0170] Where: f c η represents the objective function value for an individual, and η is a parameter reflecting the problem size.

[0171] Step S3-5: Perform genetic operator operations to obtain a scheduling scheme, and perform scheduling based on the scheduling scheme. The genetic operators specifically include selection operators, crossover operators, and mutation operators.

[0172] (1) Selection Operator

[0173] The tournament selection method is used, which involves randomly sampling *s* individuals from the population and selecting the best individual to proceed to the next generation. Specifically, the operation of selecting the winning individual and eliminating the inferior ones is called selection. The purpose of selection is to directly pass on the optimized individual (or solution) to the next generation or to generate new individuals through crossover and then pass them on to the next generation. The selection operation is based on the fitness evaluation of individuals in the population. This algorithm uses tournament selection, randomly sampling *s* individuals from the population and then selecting the best one to proceed to the next generation. Only an individual with a fitness value superior to all other competitors can win the tournament.

[0174] (2) Crossover operator

[0175] like Figure 4 As shown, the execution process includes:

[0176] Step S3-5-1-1: Select two individuals with different chromosomes from the population selected through the tournament, denoted as parent A and parent B;

[0177] Step S3-5-1-2: Randomly select two intersection points pos1 and pos2 in parent generation A, and extract the gene segment in parent generation A located in the interval [pos1,pos2].

[0178] Step S3-5-1-3: Copy the gene segment of parent A directly to the same position in the offspring, preserving its local temporal characteristics; fill in the remaining genes according to the original order in parent B, skipping the ship numbers already present in the offspring, to ensure that the chromosomes are legally arranged;

[0179] Step S3-5-1-4: Verify whether the offspring meets the ship uniqueness constraint. If a conflict arises due to crossover, initiate the repair mechanism.

[0180] (5) Mutation operator

[0181] like Figure 5 As shown, its execution process includes:

[0182] Step S3-5-2-1: Randomly select an individual with a specific chromosome from the population;

[0183] Step S3-5-2-2: Randomly select two different positions in the chromosome sequence;

[0184] Step S3-5-2-3: Completely reverse the gene sequence between the two positions to generate a new individual.

[0185] like Figure 6 The diagram shown is a flowchart of the heuristic genetic algorithm solution described in step S5 above.

[0186] In this application, a direct transshipment scheduling model for seagoing and river vessels is constructed based on the production scheduling data of a dry bulk cargo breakwater terminal at a seaport. The port has continuous berths with an average water depth of over 10 meters, capable of simultaneously accommodating multiple bulk carriers. The seagoing vessel berthing area is located at a quay-type terminal with a shoreline length of 500 meters, capable of accommodating two 100,000-ton seagoing vessels simultaneously. The river vessels used for transshipment are berthed at the breakwater terminal, which has one breakwater, each 250 meters long, capable of accommodating two 5,000-ton river vessels simultaneously on one side. The two sides of the breakwater are uniformly numbered in a linear merging manner (0-500 meters), with 0-250 meters designated for deep-sea vessels and 251-500 meters for near-sea vessels. Figure 1 As shown.

[0187] The matching plan and parameters for river-sea intermodal vessels at a certain seaport's dry bulk cargo breakwater terminal are shown in Table 2.

[0188] Table 2

[0189]

[0190]

[0191] The specifications of the loading and unloading equipment in the river-sea intermodal transport area of ​​a dry bulk cargo breakwater terminal in a certain seaport are shown in Table 3:

[0192] Table 3

[0193] Equipment Name efficiency quantity cost Ship unloader 625t / h 8 units 500 yuan / unit·h Ship loader 700t / h 8 units 300 yuan / unit·h belt conveyor 1250t / h 1200mm bandwidth 0.49 yuan / ton·km

[0194] Other costs of river-sea intermodal transport for a dry bulk cargo breakwater terminal at a certain seaport are shown in Table 4:

[0195] Table 4

[0196] Ship type Parking fee (RMB / hour) Environmental cost (RMB / h) 30,000-ton seagoing vessel 1100 800 50,000-ton seagoing vessel 1500 1200 2000-ton riverboat 320 150 3000-ton riverboat 450 300 5000-ton riverboat 600 480

[0197] In this embodiment, the main parameters of the heuristic genetic algorithm are set as follows: the maximum number of iterations is set to 1000, the population size is 200, the crossover probability is 0.8, and the mutation probability is 0.2.

[0198] like Figure 7 As shown, combining the above parameters and data, the solution is obtained by running the river-sea intermodal direct transport scheduling model. The fitness function value change curve during the algorithm's 1000 iterations is shown in the figure. Figure 7 As shown in Table 5, the combined river-sea transport scheduling plan for dry bulk cargo is obtained after simultaneous operation; the combined river-sea transport operation time is shown in Table 6; and the combined river-sea transport operation cost is shown in Table 7.

[0199] Table 5

[0200]

[0201]

[0202]

[0203] Table 7

[0204]

[0205]

[0206] As shown in Table 5, the joint scheduling plan for dry bulk river-sea intermodal transport demonstrates that the model achieves efficient docking and optimal resource allocation between seagoing and river vessels while meeting the requirement of matching vessel arrival times. Regarding berth allocation, river vessels are concentrated in the 340-430 meter range on the breakwater, indicating that the scheduling model prioritizes utilizing the near-shore berths on the breakwater for ore transshipment (RB1), while coal transshipment (RB2) is mainly distributed in the 50-190 meter range on the far-shore side of the breakwater. This north-south zoning strategy effectively avoids cross-contamination between cargo types and shortens the conveyor belt distance. The distribution of unloading and loading machines is also relatively balanced, ensuring simultaneous loading and unloading operations.

[0207] like Figure 8 and Figure 9 As shown, in order to more intuitively represent the berthing time and position of arriving vessels, a two-dimensional coordinate diagram of the berth allocation plan for seagoing vessels and river vessels was drawn based on Table 5.

[0208] Table 6 shows the close coordination between sea-river intermodal transport operations and the berthing and loading / unloading of seagoing vessels and river vessels. Taking SV1-ore as an example, its loading / unloading operation lasts 15.40 hours, and the river vessels RB1-1 and RB1-2, which are matched with it, immediately begin loading / unloading operations after berthing. However, the waiting time for the two matched river vessels to berth is 0.08 hours, indicating room for optimization. In the future, a more closely coordinated arrival time schedule can be developed to reduce the waiting time for river vessels to berth. During the SV1-ore operation cycle (01:15-16:25), an average of one river vessel is completed for direct transshipment every 1.71 hours, and during the SV2-coal operation cycle (02:10-23:30), an average of one vessel is processed every 1.52 hours, indicating that the system maintains a stable throughput even with the parallel scheduling of two seagoing vessels.

[0209] Table 7 shows the operating costs of the river-sea intermodal transport system, indicating that the relative berthing positions of vessels play a role in equipment costs. Looking at the river vessel data, the RB1 series vessels berth near the sea (370-440 meters) in the breakwater area. This berthing arrangement shortens the horizontal transport distance of the conveyor belts between the seagoing and river vessels, reducing corresponding equipment costs. Conversely, the RB2 series river vessels berth near the sea (50-190 meters), resulting in a longer horizontal transport distance and relatively higher conveyor belt operating costs, thus increasing equipment costs. Furthermore, there is a positive correlation between vessel port costs and equipment costs, indicating that vessel waiting time, operational procedures, and equipment utilization directly affect overall costs.

[0210] To further verify the effectiveness of the model, the solution results under varying objective function weights will be analyzed. In solving the model, this application considers the importance of all vessels' total port time and the carrier's total operating cost in a balanced manner, using a value of 0.5 for the example analysis. To compare the impact of the objective function weights on the results, λ was set to 0.9, 0.7, 0.3, and 0.1 respectively. A larger value indicates that the optimization objective focuses more on reducing the total port time of all vessels; conversely, a smaller value indicates a tendency to reduce the carrier's total operating cost. To verify the solution results for the five values, corresponding numerical experiments were conducted using data from the same example, yielding the solution results for the "sea-to-river" direct transshipment scheduling operation plan for dry bulk river-sea intermodal transport under different values, as shown in Table 8.

[0211] Table 8

[0212] λ value Total time spent in port by all vessels (h) Total operating costs for the carrier (RMB) 0.9 107.78 284265.5 0.7 108.25 280871.3 0.5 108.53 278165.2 0.3 109.16 276168.1 0.1 109.31 275638.3

[0213] As shown in Table 8, the parameter λ affects both the total time a vessel spends in port and the carrier's total operating cost. As the λ value decreases, the carrier's total operating cost tends to decrease, while the total time all vessels spend in port increases. For example, when the λ value decreases from 0.9 to 0.7, the total time spent in port increases slightly from 107.78 hours to 108.25 hours, while the total operating cost decreases from RMB 284,265.5 to RMB 280,871.3. Further decreasing to 0.5, the change in time spent in port is not significant, but the cost continues to decrease to RMB 278,165.2. When the λ value continues to decrease to 0.3 and 0.1, the vessel's time spent in port increases to 109.16 hours and 109.31 hours respectively, while the total operating cost decreases significantly to RMB 276,168.1 and RMB 275,638.3. This indicates that when the optimization objective focuses more on reducing operating costs, the total time vessels spend in port will increase, potentially leading to longer waiting times for some vessels and impacting overall operational efficiency. Conversely, when the λ value is large, the optimization objective tends to reduce vessel time in port to improve operational efficiency, but this may increase additional scheduling costs, such as berth adjustments and equipment allocation, resulting in higher operating costs.

[0214] Overall, different values ​​of λ can strike a balance between operating costs and operational efficiency. The result when λ = 0.5 is relatively balanced. While maintaining low costs, the total time of ships in port is controlled within a reasonable range, providing a reasonable scheduling reference for optimizing the scheduling of river-sea intermodal direct transport.

[0215] In this embodiment, a multi-objective river-sea intermodal direct transport scheduling optimization model was constructed to minimize the total port time of all vessels and the total operating cost of the carrier. A heuristic genetic algorithm was designed to solve the model. Then, the algorithm was used to solve the model on actual port data to obtain the river-sea intermodal transport scheduling scheme and operating cost within the scheduling cycle. At the same time, the model solution results under different objective function weights were analyzed and compared to verify the effectiveness of the model, providing theoretical support for river-sea intermodal transport scheduling decisions at port dry bulk cargo terminals.

[0216] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A method for optimizing the scheduling of direct transshipment between seagoing vessels and river vessels in dry bulk river-sea intermodal transport, characterized in that, include: Step S1: Construct a river-sea intermodal direct transport scheduling model and define and set relevant parameters and variables; Step S2: Construct an objective function with the total time all vessels spend in port and the total operating cost of the carrier as the objective, and construct berthing position constraints, equipment operation constraints, river-sea intermodal transport connection constraints, and time connection constraints as constraint conditions; Step S3: Use a heuristic genetic algorithm to solve the river-sea intermodal direct transport scheduling model to obtain a scheduling scheme, and then perform scheduling based on the scheduling scheme.

2. The method for optimizing the scheduling of direct transshipment between seagoing vessels and river vessels in dry bulk river-sea intermodal transport according to claim 1, characterized in that, Step S1 includes: Step S1-1: Assume the river-sea intermodal direct transport scheduling model contains n evaluation indicators and a total of m operational modes to be evaluated. Establish the original evaluation matrix as R = [r ij ] m×n , where r ij Let be the evaluation value of the i-th evaluation indicator for the j-th work mode; Step S1-2: Divide the evaluation indicators into positive and negative indicators. The standardization process for positive indicators is as follows: The standardization process for negative indicators is as follows: Where: s ij The standardized evaluation index value is min(r). j ) represents the minimum index value for the j-th operation mode, max(r) j ) represents the maximum index value for the j-th operation mode; Step S1-3: Determine the objective weight of each indicator based on the amount of information contained in each indicator: Where: ω j V represents the objective weight of the j-th indicator. j Let σ be the amount of information contained in the j-th indicator. j p is the standard deviation of the j-th indicator. ij To evaluate the correlation coefficient between the i-th and j-th indicators; Step S1-4: Construct a weighted standardization matrix, where each element in the weighted standardization matrix is ​​the product of the standardized evaluation index value and the objective weight of the corresponding index. Step S1-5: Determine the optimal and worst ideal solutions for the evaluation index as follows: Among them: A j + For the optimal ideal solution of the evaluation index, A j - The worst-case ideal solution for the evaluation index. For the first index in the optimal ideal solution A j + The maximum or minimum value in, For the second index in the optimal ideal solution A j + The maximum or minimum value in, For the nth index in the optimal ideal solution A j + The maximum or minimum value in, For the first index in the worst ideal solution A j - The maximum or minimum value in, For the second index in the worst ideal solution A j - The maximum or minimum value in, For the nth index in the worst ideal solution A j - The maximum or minimum value in; When the evaluation index is a positive index, the optimal and worst ideal solutions for the evaluation index are: When the evaluation index is negative, the optimal and worst ideal solutions for the evaluation index are: Where: e ij These are the elements in the weighted standardized matrix; Step S1-6: Calculate the Euclidean distance from each job mode to be evaluated to the optimal and worst ideal solutions: in: To evaluate the Euclidean distance from job mode i to the optimal ideal solution. To evaluate the Euclidean distance from operation mode i to the worst ideal solution; Step S1-7: Calculate the relative proximity: Where: Z i To evaluate the relative closeness of each evaluation indicator under operation mode i.

3. The method for optimizing the scheduling of direct transshipment between seagoing vessels and river vessels in dry bulk river-sea intermodal transport according to claim 2, characterized in that, The total time all vessels spent in port is: Where: f1 is the total time all vessels spend in port, qe i Let q be the time when ship i begins to depart from its berth. i For the arrival time of seagoing vessel i, ke j k is the starting departure time of riverboat j. j Let I be the arrival time of riverboat j, I be the assembly time of ocean-going vessels, and J be the assembly time of riverboats. The carrier's total operating costs are: f2=C1+C2 Where: f2 is the carrier's total operating cost, C1 is the sum of the ship's port detention cost and the cost of using the loading and unloading machinery, and C2 is the conveyor belt transportation cost.

4. The method for optimizing the scheduling of direct transshipment between seagoing vessels and river vessels in dry bulk cargo river-sea intermodal transport according to claim 2, characterized in that, Step S3 specifically includes: Step S3-1: Serialize the vessels according to their arrival time sequence in the river-sea intermodal transport; Step S3-2: First decode the ship's berthing position and the ship's loading and unloading machine assignment, and then decode the ship's berthing time. Step S3-3: Based on the estimated arrival time of the river-sea intermodal vessels, construct an initial solution sequence with practical scheduling significance; Step S3-4: Select the fitness function F c : Where: f c η represents the objective function value for an individual, and η is a parameter reflecting the problem size. Step S3-5: Perform genetic operator operations to obtain a scheduling scheme, and perform scheduling based on the scheduling scheme. The genetic operators specifically include selection operators, crossover operators, and mutation operators.

5. The method for optimizing the scheduling of direct transshipment between seagoing vessels and river vessels in dry bulk river-sea intermodal transport according to claim 4, characterized in that, The process of decoding the ship's berthing position and the ship's loading / unloading machine assignment in step S3-2 includes: Step S3-2-1-1: Following the principle of minimizing relative berthing distance, after determining the berth for seagoing vessel i, iterate through the candidate berths for riverboat j and select the one that satisfies the minimum requirement. ij The berth combination, in which, mind ij This refers to the minimum relative distance between seagoing vessel i and riverboat j when they are moored. Step S3-2-1-2: Following the principle of allocating as many as possible, when the berth loading and unloading machines are idle, allocate the maximum number of idle loading and unloading machines to each ship without exceeding the upper limit of equipment per ship, thereby maximizing equipment utilization.

6. The method for optimizing the scheduling of direct transshipment between seagoing vessels and river vessels in dry bulk river-sea intermodal transport according to claim 4, characterized in that, The decoding of the ship's berthing time in step S3-2 includes: Step S3-2-2-1: When the berth is available and the seagoing vessel is ready, the berthing time is determined by the arrival time of the seagoing vessel and the river vessel. Step S3-2-2-2: When the berth is unavailable, i.e., the previous river vessel is being served, the berthing time is determined by the departure time of the previous river vessel and the arrival time of this river vessel.

7. The method for optimizing the scheduling of direct transshipment between seagoing vessels and river vessels in dry bulk river-sea intermodal transport according to claim 4, characterized in that, The initial solution generation process in step S3-3 includes: Step S3-3-1: Read the estimated arrival times of all vessels; Step S3-3-2: Sort the vessel group in ascending order of arrival time; Step S3-3-3: Based on the sorting results, record the ship number and generate the initial individual; Step S3-3-4: Use multiple sequences formed based on sorting perturbation as the initial population to enhance the diversity of the algorithm.

8. The method for optimizing the scheduling of direct transshipment between seagoing vessels and river vessels in dry bulk river-sea intermodal transport according to claim 4, characterized in that, The selection operator uses the tournament selection method, which randomly samples s individuals from the population and selects the best individual to enter the next generation.

9. The method for optimizing the scheduling of direct transshipment between seagoing vessels and river vessels in dry bulk river-sea intermodal transport according to claim 8, characterized in that, The execution process of the crossover operator includes: Step S3-5-1-1: Select two individuals with different chromosomes from the population selected through the tournament, denoted as parent A and parent B; Step S3-5-1-2: Randomly select two intersection points pos1 and pos2 in parent generation A, and extract the gene segment in parent generation A located in the interval [pos1,pos2]. Step S3-5-1-3: Copy the gene segment of parent A directly to the same position in the offspring, preserving its local temporal characteristics; fill in the remaining genes according to the original order in parent B, skipping the ship numbers already present in the offspring, to ensure that the chromosomes are legally arranged; Step S3-5-1-4: Verify whether the offspring meets the ship uniqueness constraint. If a conflict arises due to crossover, initiate the repair mechanism.

10. The method for optimizing the scheduling of direct transshipment between seagoing vessels and river vessels in dry bulk river-sea intermodal transport according to claim 8, characterized in that, The execution process of the mutation operator includes: Step S3-5-2-1: Randomly select an individual with a specific chromosome from the population; Step S3-5-2-2: Randomly select two different positions in the chromosome sequence; Step S3-5-2-3: Completely reverse the gene sequence between the two positions to generate a new individual.

Images