A river-sea intermodal vessel collaborative scheduling method and system based on multi-objective optimization

Abstract

The application discloses a kind of based on multi-objective optimization's river-sea intermodal ship collaborative scheduling method and system, the method includes: the dynamic data of river ship and sea ship, static data, scheduling application data and channel environment data are collected and fused, and fusion multi-source heterogeneous data are obtained;Channel digital model is constructed, and the dynamic traffic capacity of each channel is calculated according to fusion multi-source heterogeneous data;With the minimum of all ship total scheduling time and total waiting time as the goal to build multi-objective optimization model;Multi-objective optimization model is solved based on improved DE-NSGA-II algorithm, and the pareto optimal solution set is obtained;Final scheduling scheme is selected from the pareto optimal solution set and executed.The application significantly reduces ship operating cost, reduces carbon emissions, and helps green port construction.The reduction of ship waiting time and total voyage directly reduces fuel consumption and ship operating cost.

Description

Technical Field

[0001] This invention relates to the field of intelligent port logistics management technology, and in particular to a method and system for collaborative scheduling of river-sea intermodal vessels based on multi-objective optimization. Background Technology

[0002] In the actual operation of river-sea intermodal ports, the formulation of vessel traffic organization plans is often constrained by multiple factors such as the structure of the complex channel, the layout of the port basin, and navigation rules. Currently, most port production departments and dispatch centers still rely mainly on manual experience for vessel traffic dispatch. With the continuous increase in the number of arriving vessels, river-sea intermodal terminals face multiple challenges, including complex vessel types, high traffic density, limited channel depth, and tidal changes. The traditional manual dispatch mode is no longer sufficient to meet the needs of efficient operation. Furthermore, current vessel dispatching technologies have the following pain points:

[0003] Single-objective limitations: Traditional scheduling systems only optimize single objectives such as waiting time or throughput, and cannot balance ship turnaround efficiency and port operation cycle; Insufficient constraint handling: Tidal window constraints are not dynamically associated with ship type (riverboat / seaboat), and channel conflict detection uses a static safety distance model, which does not consider speed differences;

[0004] Real-time limitations: Mathematical programming methods time out when scheduling large-scale ships (unable to solve for more than 300 ships).

[0005] Lack of coordination: Independent decision-making among berth allocation, channel passage, and tidal windows leads to resource conflicts. Summary of the Invention

[0006] To address the aforementioned shortcomings in existing technologies, this invention provides a multi-objective optimization-based method and system for collaborative scheduling of river-sea intermodal vessels. This solution addresses the problems of structural resource mismatch in complex waterways, low scheduling efficiency, difficulty in simultaneously optimizing conflicting total scheduling time and total waiting time objectives, and lack of rapid response capability to dynamic environments (such as tides and vessel delays).

[0007] To achieve the above-mentioned objectives, this invention provides a multi-objective optimization-based method for coordinated scheduling of river-sea intermodal vessels, comprising:

[0008] Dynamic data, static data, scheduling request data, and waterway environment data of river vessels and sea vessels are collected and fused to obtain fused multi-source heterogeneous data;

[0009] Construct a digital model of the waterway and calculate the dynamic traffic capacity of each waterway based on the fusion of multi-source heterogeneous data;

[0010] A multi-objective optimization model is constructed with the goal of minimizing the total scheduling time and total waiting time of all ships.

[0011] The multi-objective optimization model was solved using the improved DE-NSGA-II algorithm, and the Pareto optimal solution set was obtained.

[0012] Based on the dynamic traffic capacity of the waterway, the final scheduling scheme is selected from the Pareto optimal solution set and executed.

[0013] Secondly, this invention also provides a multi-objective optimization-based river-sea intermodal vessel collaborative scheduling system, comprising:

[0014] The multi-source heterogeneous data acquisition and fusion module is used to collect dynamic data, static data, scheduling application data, and waterway environment data from river vessels and sea vessels, and then fuse them to obtain fused multi-source heterogeneous data.

[0015] The dual-channel state perception and modeling module is used to construct a digital model of the channel and calculate the dynamic passage capacity of each channel based on the fusion of multi-source heterogeneous data.

[0016] A multi-objective collaborative optimization scheduling engine is used to construct a multi-objective optimization model with the goal of minimizing the total scheduling time and total waiting time of all ships; and the multi-objective optimization model is solved based on the improved DE-NSGA-II algorithm to obtain the Pareto optimal solution set;

[0017] The scheduling scheme generation and dynamic adjustment module is used to select the final scheduling scheme from the Pareto optimal solution set based on the dynamic traffic capacity of the waterway and execute it, and trigger re-optimization when deviations or environmental changes are detected.

[0018] Visualized decision support module: used to display waterway status, ship dynamics, scheduling plans and key performance indicators, and supports human interaction.

[0019] The beneficial effects of this invention are as follows:

[0020] 1. The dynamic channel allocation model and collaborative optimization mechanism provided by this invention significantly change the imbalance in the utilization rates of primary and secondary channels under the original scheduling model, greatly improving the feasibility and safety of the scheme, and avoiding channel conflicts or uneven resource utilization caused by model simplification. It effectively solves the structural resource mismatch problem of "deep-water channels being idle and shallow-water channels being congested."

[0021] 2. A balance was found between total scheduling time (TST) and total waiting time (TWT) through a multi-objective optimization algorithm, significantly reducing the ineffective waiting time of ships in the connection area and the total time spent in port. This significantly reduces ship operating costs, carbon emissions, and contributes to the construction of green ports. The reduction in ship waiting time and total voyage directly reduces fuel consumption and ship operating costs. At the same time, the reduction in fuel consumption directly leads to a reduction in carbon dioxide (CO2) and other harmful gas emissions.

[0022] 3. Through the forward-looking spatiotemporal conflict prediction and intelligent resolution mechanism of the present invention, spatiotemporal conflicts between ships at key nodes such as waterway intersection areas can be avoided during the scheduling plan generation stage, which greatly improves the level of navigation safety.

[0023] 4. By employing an integrated berth-channel scheduling strategy oriented towards river-sea vessel collaboration, the actual occupancy of berths as a scarce resource is accurately modeled, ensuring the accuracy of subsequent vessel scheduling plans while guaranteeing that the berth occupancy constraints of departing vessels are met. This achieves true collaboration between vessels, port, and channel, significantly reducing vessel waiting time at anchor (TWT index) and improving the overall port turnaround efficiency (TST index). Attached Figure Description

[0024] Figure 1 The flowchart illustrates a multi-objective optimization-based collaborative scheduling method for river-sea intermodal transport vessels, provided as an example. Detailed Implementation

[0025] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0026] This embodiment uses a complex channel hub area in Beibu Gulf Port as an application scenario to describe in detail the specific implementation process of the present invention, such as... Figure 1 As shown, it includes the following steps:

[0027] S1. The dynamic data, static data, scheduling application data and waterway environment data of river vessels and sea vessels are collected and fused through the multi-source heterogeneous data acquisition and fusion module to obtain fused multi-source heterogeneous data.

[0028] The system periodically collects dynamic data (GPS / BeiDou position, speed, heading), static data (draft, length, beam, load capacity), scheduling request data (request time, destination berth), and waterway environmental data (real-time water depth, current speed, wind speed) from intelligent terminal devices deployed on ships, waterways, and ports. Subsequently, the data fusion center performs time synchronization, coordinate unification (e.g., mapping to WGS-84 electronic charts), and ship data association based on MMSI codes on the aforementioned multi-source heterogeneous data, ultimately forming a global information view with a consistent spatiotemporal reference serving the optimization engine. Its function is to provide accurate, consistent, and comprehensive input for subsequent high-precision modeling and optimization decisions, overcoming the problem of inaccurate decisions caused by data silos.

[0029] S2. Construct a digital model of the waterway through the dual-channel state perception and modeling module, and calculate the dynamic passage capacity of each waterway based on the fusion of multi-source heterogeneous data.

[0030] In the server memory, channel zoning modeling is performed based on electronic nautical chart data, clearly defining the main channel, secondary channel, and key functional areas (connecting area, port basin area, and confluence area).

[0031] The formula for calculating the dynamic capacity of each waterway is as follows:

[0032]

[0033] in, For waterway The dynamic traffic capacity index. For waterway Real-time available water depth, For waterway Design water depth, For waterway The number of ships within the zone, For waterway Ship capacity threshold within the zone, For waterway The current flow rate, For flow rate threshold, , , All are weighting coefficients that are dynamically adjusted according to channel priority, and .

[0034] This step clearly defines the different navigation distance constants for the main channel and secondary channels, and designs multiple functional zones such as the connecting area, port basin, and confluence area. Differentiated route segments are constructed for different vessel types (riverboats / seagoing vessels) and different directions (entering / leaving port), with higher safety time limits set for the confluence area segments. The complex, compound waterways of the physical world are transformed into a quantifiable, dynamically evolving digital twin that can be understood by algorithms, providing precise decision-making basis for subsequent intelligent scheduling and solving the problem that general models cannot handle the structural characteristics of compound waterways.

[0035] S3. Construct a multi-objective optimization model with the goal of minimizing the total scheduling time and total waiting time of all ships through a multi-objective collaborative optimization scheduling engine;

[0036] This invention addresses the scenario of combined river-sea transport via a multi-channel system, explicitly establishing two conflicting objective functions that require coordinated optimization. The specific expressions are as follows:

[0037]

[0038]

[0039] in, Let the first objective function be... This represents the maximum scheduling end time among all vessels. The minimum value at the start time of all ship scheduling. The second objective function is... Riverboat The start scheduling time, Riverboat The application scheduling time, For seagoing vessels The start scheduling time, For seagoing vessels The application scheduling time.

[0040] The aforementioned objective function clearly defines two conflicting objectives that need to be optimized collaboratively, laying the foundation for finding the Pareto optimal solution set. The first objective function aims to maximize the navigation efficiency of the hub area and shorten the total time from vessel application to completion of operations. Its core meaning is to minimize the scheduling time for all vessel types (including river vessels and sea vessels), specifically by minimizing the difference between the maximum scheduling end time and the minimum scheduling start time for all vessels. The maximum scheduling end time refers to the largest time point between the scheduling end times of inbound and outbound vessels. For inbound vessels, the scheduling end time is based on their arrival time at the berth, and for outbound vessels, it is based on their arrival time at the channel entrance. The minimum scheduling start time is the smallest time point between the scheduling start times of inbound and outbound vessels. For inbound vessels, the scheduling start time is their departure time from the anchorage, and for outbound vessels, it is their departure time from the berth.

[0041] The second objective function aims to improve the ship experience and reduce operational costs incurred due to scheduling wait times. Its core meaning is to minimize ship waiting time, specifically by minimizing the total time difference between the start time and the time of requesting scheduling for all ships. The "time of requesting scheduling" specifically refers to the exact point in time when a ship waiting to enter port at anchor or waiting to leave port submits its request for port entry or exit scheduling.

[0042] In addition, the present invention refines the scheduling rules by establishing a group of constraints in the modeling process, including berth allocation constraints, tidal window constraints, navigation continuity constraints, and conflict resolution constraints.

[0043] The berth allocation constraints are as follows:

[0044]

[0045]

[0046]

[0047]

[0048]

[0049]

[0050]

[0051]

[0052]

[0053]

[0054] in, , All are riverboat indexes. ; , All are ocean-going vessel indexes. ; Indicates berth; As a decision variable, when the berth The berthing capacity meets the needs of riverboats When there is a need for docking, ,otherwise ; As a decision variable, when the berth The berthing capacity meets the needs of seagoing vessels When there is a need for docking, ,otherwise ; As a decision variable, when the riverboat Assigned to a berth hour, ,otherwise ; As a decision variable, when the seagoing vessel Assigned to a berth hour, ,otherwise ; As a decision variable, when the riverboat River boat docked at the same berth And riverboats On the riverboat When it docked before, ,otherwise ; As a decision variable, when the seagoing vessel and sea ships docked at the same berth And seagoing vessels On a sea ship When it docked before, ,otherwise ; Riverboat Actual departure time Riverboat Actual berthing time; To indicate a seagoing vessel Actual departure time Indicates seagoing vessel Actual berthing time; For the maximum value, among all constraints, the maximum value involved is... M It is a sufficiently large positive constant. M The value of is determined based on the scheduling system unit and scheduling cycle of this invention. For example, if minutes are used as the unit and days as the cycle, then M The value is set to 1440 (minutes). This value should be greater than the maximum possible difference between any time variables in the model within a single scheduling period to ensure that the relevant constraints can automatically relax and fail when the logical conditions are not met.

[0055] The above berth allocation constraints ensure that: the berthing capacity of each river vessel or sea vessel is not less than its tonnage; for each vessel and each berth, allocation can only be made when the berth capacity meets the requirements; each vessel entering the port is allocated a berth; for vessels leaving the port, the berth allocation variable is automatically set to zero, while vessels needing to enter the port require a berth allocation; only one vessel can berth at the same berth at a time, and the berthing times of vessels berthing at the same berth cannot overlap.

[0056] When the natural water depth of a port cannot meet the draft requirements of large vessels, it is necessary to use the tide to facilitate vessel entry and exit from the port. This involves utilizing the tidal window when the water depth meets the vessel's draft requirements. Since vessels need to maintain continuous navigation during entry and exit, the entry and exit procedures must be completed within the same tidal window to ensure the safety and continuity of the entire navigation process.

[0057] The tidal window constraint is as follows:

[0058]

[0059]

[0060]

[0061] in, Indicates the right-hand time of the tide-climbing time window; As a decision variable, when the seagoing vessel When the sailing direction is towards the port, ,otherwise ; Indicates seagoing vessel The moment of arrival at the channel entrance, Indicates the leftmost time of the tide-climbing time window; Indicates seagoing vessel Apply for departure time.

[0062] The aforementioned tidal window constraints ensure that: the time for scheduling vessels entering the port at high tide must be strictly limited to the time between the left and right edges of the tidal window, i.e., later than the left edge and earlier than the right edge; the time for scheduling vessels leaving the port at high tide must be strictly limited to the time between the left and right edges of the tidal window, i.e., later than the left edge and earlier than the right edge; and the time for vessels leaving the port at high tide to depart from their berths must not be earlier than the start time of the tidal window, nor earlier than the departure time declared by the vessel; thus achieving precise control over vessel navigation at high tide.

[0063] The specific constraints on navigation continuity are as follows:

[0064]

[0065]

[0066]

[0067]

[0068]

[0069]

[0070]

[0071]

[0072]

[0073]

[0074]

[0075]

[0076] in, Riverboat The moment of arrival at the channel entrance; Indicates the distance from the anchorage to the waters at the entrance to the channel; Riverboat The speed of navigation; As a decision variable, when the riverboat When the navigation direction is towards the port, ,otherwise ; For seagoing vessels The speed of navigation; Riverboat The moment of arrival at the entrance waters of the harbor basin; The distance from the main channel entrance waters to the harbor basin entrance waters; It indicates the width of the main channel, which is the distance that a riverboat needs to cross to reach the entrance waters of the harbor basin; For seagoing vessels The moment of arrival at the entrance waters of the harbor basin; Riverboat The time when the scheduling is completed; Riverboat The distance from the entrance waters of the harbor basin to the berth; Indicates seagoing vessel The time when the scheduling is completed; Indicates seagoing vessel The distance from the entrance waters of the harbor basin to the berth.

[0077] The aforementioned navigation continuity constraints ensure the following: navigation continuity during the ship's entry into port; navigation continuity during the ship's departure from port; navigation continuity for entering ships, i.e., the continuity of entering ships from their berths to the port basin entrance waters; navigation continuity for departing ships, i.e., the continuity of ships from the port basin entrance to the channel entrance waters. With the help of these constraints, the specific times when ships navigate to the channel entrance waters, cross waters, and the port basin entrance waters can be deduced in sequence.

[0078] The conflict resolution constraints are categorized into overtaking conflict resolution constraints, encounter conflict resolution constraints, and crossover conflict resolution constraints. By introducing safe time distances and 0-1 decision variables, the three typical navigational conflicts that may occur between ships—overtaking, encounters, and crossovers—are avoided in advance at the model level. This not only ensures safety but also greatly reduces the risk of the plan failing due to conflicts during the execution phase.

[0079] The specific constraints for resolving conflicts during overtaking are as follows:

[0080]

[0081]

[0082]

[0083]

[0084] in, Riverboat The moment of arrival at the channel entrance; The minimum safe time interval to avoid overtaking conflicts; As a decision variable, when the riverboat On the riverboat In the past, ,otherwise ; Indicates a seagoing vessel The moment of arrival at the channel entrance; As a decision variable, when the seagoing vessel When the sailing direction is towards the port, ,otherwise ; As a decision variable, when the seagoing vessel On a sea ship In the past, ,otherwise ; Riverboat The moment of arrival at the entrance waters of the harbor basin; For seagoing vessels The moment of arrival at the entrance waters of the harbor basin.

[0085] Overtaking conflict mitigation constraints prevent overtaking incidents between the two types of vessels while navigating the channel. Specifically, vessels entering the port must maintain a minimum safe time interval, and the speed of a following vessel must not exceed that of the vessel in front. Secondly, this constraint controls vessels leaving the port basin, requiring them to maintain a necessary safe time interval and limiting the speed of following vessels to no more than that of the vessel in front, thereby avoiding overtaking conflicts between departing vessels.

[0086] The conflict resolution constraint requires that two vessels traveling in opposite directions on the same segment of the voyage maintain a sufficient safe time interval. Specifically:

[0087]

[0088]

[0089]

[0090]

[0091] in, To avoid the minimum safe time interval in case of conflict; Riverboat The time when the scheduling is completed; As a decision variable, when the riverboat Riverboats entering the port at the same time When departing, ,otherwise ; Indicates seagoing vessel The time when the scheduling is completed; As a decision variable, when the seagoing vessel Entering port and at the same time, seagoing vessels When departing, ,otherwise ; As a decision variable, when the riverboat Departing port and at the same time sea vessel Upon entering the port, ,otherwise .

[0092] The above constraints ensure that a minimum safe time interval must be maintained between vessels entering and leaving the port at the channel entrance; between sea vessels entering and river vessels leaving the port in the port basin entrance waters; between sea vessels entering and river vessels leaving the port in the port basin entrance waters; and between river vessels entering and river vessels leaving the port in the port basin entrance waters.

[0093] Cross-conflict resolution constraints ensure that the minimum safe time interval for cross-conflict is maintained between river vessels departing from the port and sea vessels entering the port in the harbor basin entrance waters. Specifically:

[0094]

[0095] in, The minimum safe time interval to avoid cross-conflicts.

[0096] S4. Solve the multi-objective optimization model based on the improved DE-NSGA-II algorithm to obtain the Pareto optimal solution set.

[0097] The specific method is as follows:

[0098] Each propagation scheduling sequence is encoded as a chromosome, where each chromosome represents a ship sequence, such as [Ship5, Ship1, Ship3, ...], indicating scheduling priority. The population is initialized based on the encoded chromosomes using a combination of heuristic and random generation strategies.

[0099] Fitness Assessment and Repair: The scheduling process for each scheduling sequence (chromosome) is simulated, and the Total Scheduling Time (TST) and Total Waiting Time (TWT) are calculated. During the simulation, for each ship in the sequence, an optimal channel is selected from the available channel set based on its draft and dynamic throughput index. When allocating berths to arriving ships, it is checked whether their expected berthing time falls within the berth's idle time window. Berth occupancy time is calculated as: berthing time + service time. During simulated ship navigation, the time window for occupancy in key areas such as junctions is predicted and compared with the occupancy time windows of already scheduled ships. If a conflict occurs, the solution is not directly abandoned; instead, a time-shifting strategy is adopted: the ship's planned departure time is postponed by one step, and the simulation is repeated until a conflict-free time point is found. This greatly improves the efficiency of finding feasible solutions, solves the problem of preventing spatiotemporal deadlock in complex channel junctions under high-density traffic flow, achieves conflict avoidance, and ensures the safety and reliability of the scheduling scheme. Furthermore, because it is a dynamic adjustment rather than a complete rescheduling, the computational efficiency is higher, meeting the requirements of real-time response. Traditional methods often involve post-event conflict resolution or have extremely high computational complexity. The method of this invention achieves online, real-time, and forward-looking conflict resolution.

[0100] During the simulation, targeted remediation strategies are employed to handle constraint violations. The remediation strategies are as follows:

[0101] (1) Strategies to correct the rationality of berth tonnage allocation: Adjust the berths allocated to ships according to the principle that the ship's deadweight must be less than or equal to the berth's tonnage. For example, if a ship entering the port is allocated a tonnage that is less than its deadweight, the berth needs to be reallocated until the berth's berthing capacity meets the ship's needs.

[0102] (2) Strategy for rationalizing the allocation of the same berth: Adjust the order of ships entering and leaving the port according to the principle of first departure and then berthing at the same berth. There are two situations: ① When two ships entering the port are assigned to the same berth, the second ship can only berth after the first ship has departed. If the allocation time of the two ships overlaps, the berth will be reassigned to the ships according to the principle of first application and first occupancy and berth allocation constraints; ② When a ship applying to enter the port is assigned to the berth of a ship applying to leave the port, and the ship entering the port is dispatched before the ship leaving the port, the order of the two ships entering and leaving the port needs to be reversed, with the ship leaving the port dispatched first and the ship entering the port dispatched later, to ensure that only one ship is allowed to berth at the same berth at the same time.

[0103] (3) The repair strategy is to meet the requirements of the tide time window: the start time and end time of the scheduling of the sea vessel must be within the tide time window. For example, if the start time and end time of the scheduling of the sea vessel are not within the tide time window, the schedule will be postponed for 10 minutes until it is within the tide time window.

[0104] (4) Safety time interval repair strategy: In order to avoid traffic conflicts between ships, the safety time interval should be met. For example, when the time interval between ships scheduled one after the other does not meet the safety time interval, the scheduling time of the ship scheduled later is postponed by 5 minutes until the safety time interval is met.

[0105] Genetic operations: An improved discrete differential evolution strategy is adopted, and parent chromosomes in the population are randomly selected for differential mutation; sequential crossover is performed on the differentially mutated chromosomes; a binary tournament selection of elite solutions based on fast non-dominated sorting and crowding calculation is adopted to ensure the diversity and convergence of the Pareto front.

[0106] Repeat fitness evaluation, repair, and genetic operations until the maximum number of iterations is reached, and output a set of Pareto optimal solutions.

[0107] By combining the global search capability of differential evolution with the multi-objective processing framework of NSGA-II, and supplementing it with a domain knowledge-driven repair strategy, the convergence speed and success rate of the algorithm in solving this ultra-complex, strongly constrained practical engineering problem are significantly improved. It can stably generate a set of high-quality Pareto optimal solutions for the complex problem of complex waterway scheduling, for decision-makers to choose from.

[0108] S5. Based on the dynamic traffic capacity of the waterway, the final scheduling scheme is selected from the Pareto optimal solution set and executed.

[0109] Plan issuance: The dispatcher selects a final plan from the Pareto front, and the server breaks it down into specific instructions (such as channel allocation, time window, recommended speed) and issues them to the corresponding ship terminals.

[0110] Rolling Time-Domain Optimization (RHO): The system starts a new round of optimization at fixed intervals (e.g., 5 minutes), using the latest time as the starting point and regenerating the scheduling scheme with new data, so that the system can adapt to the dynamically changing environment.

[0111] Abnormal rescheduling: When a major deviation is detected (such as ship failure or sudden weather change), a temporary re-optimization is immediately triggered, generating an emergency adjustment command.

[0112] To verify the technical effectiveness of this invention, a comparative simulation experiment was conducted in the Beibu Gulf Port dual-channel simulation environment, comparing this invention with the traditional FSFC scheduling method. The experimental results show that, compared with the traditional FSFC scheduling method, this invention achieves the following quantitative effects: Channel utilization: The main channel utilization rate increased from 75% to 92%, and the secondary channel utilization rate increased from 50% to 85%, a comprehensive improvement of approximately 41%, effectively solving the structural resource mismatch problem of "idle deep-water channels and congested shallow-water channels"; Efficiency improvement: The average total vessel turnaround time (TST) was reduced by 28.2%, and the total waiting time (TWT) was reduced by 38.7%; Safety: The occurrence rate of spatiotemporal conflicts in the channel intersection area was reduced by more than 95%. During the three-month simulation period, a safety record of zero conflicts in key areas was achieved, greatly improving the level of navigation safety. Closed-loop control and continuous optimization of the system were realized, ensuring the real-time performance and robustness of the scheduling scheme.

[0113] In summary, this invention significantly alters the imbalance in the utilization rates of main and secondary channels under the original scheduling model, greatly improving the feasibility and safety of the solution and avoiding channel conflicts or uneven resource utilization caused by model simplification. It significantly reduces ship operating costs, decreases carbon emissions, and contributes to the construction of green ports. The reduction in ship waiting time and total voyage directly lowers fuel consumption and ship operating costs. It achieves true synergy between ships, ports, and channels, drastically reducing ship waiting time at anchor (TWT index) and improving the overall port turnaround efficiency (TST index).

Claims

1. A method for collaborative scheduling of river-sea intermodal vessels based on multi-objective optimization, characterized in that, include: Dynamic data, static data, scheduling request data, and waterway environment data of river vessels and sea vessels are collected and fused to obtain fused multi-source heterogeneous data; Construct a digital model of the waterway and calculate the dynamic traffic capacity of each waterway based on the fusion of multi-source heterogeneous data; A multi-objective optimization model is constructed with the goal of minimizing the total scheduling time and total waiting time of all ships. The multi-objective optimization model was solved using the improved DE-NSGA-II algorithm, and the Pareto optimal solution set was obtained. Based on the dynamic throughput capacity of the waterway, the final scheduling scheme is selected from the Pareto optimal solution set and executed; The formula for calculating the dynamic traffic capacity of each waterway is as follows: in, For waterway The dynamic traffic capacity index. For waterway Real-time available water depth, For waterway Design water depth, For waterway The number of ships within the zone, For waterway Ship capacity threshold within the zone, For waterway The current flow rate, For flow rate threshold, , , All are weighting coefficients that are dynamically adjusted according to channel priority, and .

2. The method according to claim 1, characterized in that, The specific expression for the objective function of the multi-objective optimization model is: in, Let the first objective function be... This represents the maximum scheduling end time among all vessels. The minimum value at the start time of all ship scheduling. The second objective function is... Riverboat The start scheduling time, Riverboat The application scheduling time, For seagoing vessels The start scheduling time, For seagoing vessels The application scheduling time.

3. The method according to claim 2, characterized in that, The multi-objective optimization model includes berth allocation constraints, specifically: in, , All are riverboat indexes. ; , All are ocean-going vessel indexes. ; Indicates berth; As a decision variable, when the berth The berthing capacity meets the needs of riverboats When there is a need for docking, ,otherwise ; As a decision variable, when the berth The berthing capacity meets the needs of seagoing vessels When there is a need for docking, ,otherwise ; As a decision variable, when the riverboat Assigned to a berth hour, ,otherwise ; As a decision variable, when the seagoing vessel Assigned to a berth hour, ,otherwise ; As a decision variable, when the riverboat River boat docked at the same berth And riverboats On the riverboat When it docked before, ,otherwise ; As a decision variable, when the seagoing vessel and sea ships docked at the same berth And seagoing vessels On a sea ship When it docked before, ,otherwise ; Riverboat Actual departure time Riverboat Actual berthing time; To indicate a seagoing vessel Actual departure time Indicates seagoing vessel Actual berthing time; It is the maximum value. It is a positive constant. M The value is determined based on the scheduling system unit and scheduling period. M The value of is greater than the maximum possible difference between any time variables within a single scheduling cycle in the multi-objective optimization model.

4. The method according to claim 3, characterized in that, The multi-objective optimization model also includes tidal window constraints, specifically: in, Indicates the right-hand time of the tide-climbing time window; As a decision variable, when the seagoing vessel When the sailing direction is towards the port, ,otherwise ; Indicates seagoing vessel The moment of arrival at the channel entrance, Indicates the leftmost time of the tide-climbing time window; Indicates seagoing vessel Apply for departure time.

5. The method according to claim 4, characterized in that, The multi-objective optimization model also includes navigation continuity constraints, specifically: in, Riverboat The moment of arrival at the channel entrance; Indicates the distance from the anchorage to the waters at the entrance to the channel; Riverboat The speed of navigation; As a decision variable, when the riverboat When the navigation direction is towards the port, ,otherwise ; For seagoing vessels The speed of navigation; Riverboat The moment of arrival at the entrance waters of the harbor basin; The distance from the main channel entrance waters to the harbor basin entrance waters; The width of the main channel; For seagoing vessels The moment of arrival at the entrance waters of the harbor basin; Riverboat The time when the scheduling is completed; Riverboat The distance from the entrance waters of the harbor basin to the berth; Indicates seagoing vessel The time when the scheduling is completed; Indicates seagoing vessel The distance from the entrance waters of the harbor basin to the berth.

6. The method according to claim 5, characterized in that, The multi-objective optimization model also includes conflict resolution constraints, namely, chasing conflict resolution constraints, encountering conflict resolution constraints, and cross-conflict resolution constraints; The specific constraints for resolving conflicts during overtaking are as follows: in, Riverboat The moment of arrival at the channel entrance; The minimum safe time interval to avoid overtaking conflicts; As a decision variable, when the riverboat On the riverboat In the past, ,otherwise ; Indicates a seagoing vessel The moment of arrival at the channel entrance; As a decision variable, when the seagoing vessel When the sailing direction is towards the port, ,otherwise ; As a decision variable, when the seagoing vessel On a sea ship In the past, ,otherwise ; Riverboat The moment of arrival at the entrance waters of the harbor basin; For seagoing vessels The moment of arrival at the entrance waters of the harbor basin; The specific constraints for conflict resolution are as follows: in, To avoid the minimum safe time interval in case of conflict; Riverboat The time when the scheduling is completed; As a decision variable, when the riverboat Riverboats entering the port at the same time When departing, ,otherwise ; Indicates seagoing vessel The time when the scheduling is completed; As a decision variable, when the seagoing vessel Entering port and at the same time, seagoing vessels When departing, ,otherwise ; As a decision variable, when the riverboat Departing port and at the same time sea vessel Upon entering the port, ,otherwise ; The specific steps for resolving cross-conflict constraints are as follows: in, The minimum safe time interval to avoid cross-conflicts.

7. The method according to claim 6, characterized in that, The specific method for solving the multi-objective optimization model based on the improved DE-NSGA-II algorithm is as follows: Each propagation scheduling sequence is encoded as a chromosome; the population is initialized using a combination of heuristic and random generation strategies. Fitness assessment and repair: Simulate the scheduling process of each scheduling sequence and calculate TST and TWT. During the simulation, adopt targeted repair strategies to handle constraint violations. Genetic operations: An improved discrete differential evolution strategy is adopted to randomly select parent chromosomes in the population for differential mutation; sequential crossover is performed on the differentially mutated chromosomes; and a binary tournament selection method based on fast non-dominated sorting and crowding calculation is used to select elite solutions. Repeat fitness evaluation, repair, and genetic operations until the maximum number of iterations is reached, and output a set of Pareto optimal solutions.

8. A system based on the multi-objective optimization-based river-sea intermodal vessel collaborative scheduling method according to any one of claims 1 to 7, comprising: The multi-source heterogeneous data acquisition and fusion module is used to collect dynamic data, static data, scheduling application data, and waterway environment data from river vessels and sea vessels, and then fuse them to obtain fused multi-source heterogeneous data. The dual-channel state perception and modeling module is used to construct a digital model of the channel and calculate the dynamic passage capacity of each channel based on the fusion of multi-source heterogeneous data. A multi-objective collaborative optimization scheduling engine is used to construct a multi-objective optimization model with the goal of minimizing the total scheduling time and total waiting time of all ships. The improved DE-NSGA-II algorithm was used to solve the multi-objective optimization model, and the Pareto optimal solution set was obtained. The scheduling scheme generation and dynamic adjustment module is used to select the final scheduling scheme from the Pareto optimal solution set based on the dynamic traffic capacity of the waterway and execute it, and trigger re-optimization when deviations or environmental changes are detected. Visualized decision support module: used to display waterway status, ship dynamics, scheduling plans and key performance indicators, and supports human interaction.

9. The system according to claim 8, characterized in that, The system initiates a new round of optimization at fixed intervals, regenerating the scheduling scheme based on the latest data; it also performs re-optimization when the system detects deviations or environmental changes.

Images