A port rolling collaborative scheduling method considering task uncertainty

Abstract

The application discloses a port rolling collaborative scheduling method considering task uncertainty, comprising the following steps: obtaining port resource information, ship information and a last-period baseline plan, and constructing a target function of port collaborative scheduling optimization and berth constraints and sailing sequence constraints based on the port resource information and the ship information; under the berth constraints and the sailing sequence constraints, performing conflict identification on the last-period baseline plan to obtain interference severity of task uncertainty disturbance; determining an optimization strategy of the last-period baseline plan according to the interference severity, and determining an evaluation value according to the target function under the determined optimization strategy, and obtaining a collaborative scheduling plan based on the evaluation value. The application can realize rolling update of ship in-out port and berth allocation in an environment with uncertain task demand, relieve port congestion and reduce ship port pressure level.

Description

Technical Field

[0001] This invention relates to the field of ship scheduling technology, and specifically to a port rolling collaborative scheduling method that takes into account the uncertainty of tasks. Background Technology

[0002] In actual port operations, external disturbances such as weather and sea conditions, waterway traffic, and equipment status can cause unexpected changes to the set of vessel operations that ports need to handle, severely impacting vessel entry and exit sequences and berth planning. On the one hand, changes in waterway control or demand may lead to alterations or even cancellations of previously submitted vessel operation plans. On the other hand, congestion at nearby ports or unforeseen circumstances may cause vessels to temporarily apply for port relocation. This unpredictability and dynamic restructuring of time windows disrupt established vessel operation plans and berth allocations, leading to conflicts in port resources and rendering the original operation plans ineffective.

[0003] Therefore, in order to ensure the orderly and efficient operation of ports, a rolling collaborative scheduling method for ports is urgently needed in an environment where task requirements are uncertain, so as to realize the rolling update of ship entry and exit and berth allocation, alleviate port congestion and reduce the level of ship congestion. Summary of the Invention

[0004] This invention provides a port rolling collaborative scheduling method that considers task uncertainty, so as to realize the rolling update of ship entry and exit and berth allocation in an environment where task demand is uncertain, thereby alleviating port congestion and reducing the level of ship congestion.

[0005] This invention provides a port rolling collaborative scheduling method considering task uncertainty, the method comprising: Obtain port resource information, vessel information, and the baseline plan of the previous cycle, and construct the objective function, berth constraints, and navigation sequence constraints for port collaborative scheduling optimization based on the port resource information and vessel information; Under berth constraints and navigation sequence constraints, conflict identification is performed on the baseline plan of the previous cycle to obtain the severity of disturbances to mission uncertainty. The optimization strategy for the baseline plan of the previous cycle is determined based on the severity of the interference. Under the determined optimization strategy, the evaluation value is determined according to the objective function, and the collaborative scheduling plan is obtained based on the evaluation value.

[0006] In some embodiments of the present invention, determining the optimization strategy for the baseline plan of the previous cycle based on the severity of interference includes: When the interference severity is less than or equal to the interference threshold, the baseline plan of the previous cycle is optimized by using a local interference strategy, and the search of the population is restricted to be adjusted within a preset range of unfrozen tasks. Unfrozen tasks are ship tasks that are in an unexecuted or interruptible state. If the severity of the interference exceeds the interference threshold, a global reordering strategy is used to optimize the baseline plan of the previous cycle, and the order, berth, and resource variables of unfrozen tasks are released for reordering.

[0007] In some embodiments of the present invention, under berth constraints and navigation sequence constraints, conflict identification is performed on the baseline plan of the previous cycle to obtain the severity of disturbances caused by mission uncertainty, including: Under berth constraints and navigation sequence constraints, calculate the channel conflict index, berth conflict index, resource overrun index, and plan stability loss index respectively. The severity of the interference is obtained by weighting and summing the indicators of channel conflict, berth conflict, resource overrun, and plan stability loss.

[0008] In some embodiments of the present invention, an objective function for port collaborative scheduling optimization, along with berth constraints and navigation sequence constraints, are constructed based on port resource information and vessel information, including: The ship waiting time is defined as the difference between the ship's arrival start time and its estimated arrival time. An objective function is constructed to minimize this waiting time, and is expressed as follows: ; In the formula, This is the time when ship i begins to enter the port. This is the estimated arrival time of vessel i. It is the effective set of ships at the k-th decision time; Based on the compatibility between berth length and ship length, the order of operations of different ships on the same berth, and the non-overlapping berth occupancy time, berth constraints are constructed. Based on the temporal relationship between the safe interval between ships entering and leaving the port, tidal windows, estimated arrival time, port entry sailing time, berthing operation time and departure time, a navigation sequence constraint is constructed.

[0009] In some embodiments of the present invention, under a defined optimization strategy, determining the evaluation value based on the objective function includes: Under a given optimization strategy, the objective function value of each candidate scheduling scheme is calculated using the objective function. Perform feasibility checks on berth constraints and navigation sequence constraints, and determine penalty values ​​based on the check results; The evaluation value is determined based on the objective function value and the penalty value.

[0010] In some embodiments of the present invention, obtaining a cooperative scheduling plan based on an evaluation value includes: The population individuals are ranked based on their evaluation values, and elite retention is performed. Based on fitness values, a selection operation is performed on the retained pairs to form a parent set; Based on the parent set, perform crossover and mutation operations on the unfrozen tasks to obtain the child set; The fitness value of the child set is calculated and the constraints are checked. Based on the fitness value and the constraint check results, it is determined whether the current child set meets the preset conditions. If the preset conditions are met, the collaborative scheduling plan is determined from the final child set.

[0011] In some embodiments of the present invention, performing crossover and mutation operations on unfrozen tasks includes: For time-series groups, conflict-guided sequential crossovers are implemented to adjust the order of inbound and outbound operations; For resource groups, perform block crossover or redistribution mutations to adjust berth assignments; If a conflict occurs in the channel safety interval, berth occupancy, or resource overrun, the unfrozen task that caused the conflict is located and removed to form a partial feasible solution. Then, the unfrozen task that caused the conflict is re-inserted using the earliest feasible insertion rule or the lowest cost insertion rule, and hierarchical repair is performed in the order of "channel-berth-time consistency".

[0012] In some embodiments of the present invention, ranking individuals in the population based on evaluation values ​​and performing elite retention includes: When the difference in waiting time between two individuals is less than the difference threshold, compare the baseline plan modification costs of the two individuals in the previous cycle, and retain the individual with the smaller modification cost.

[0013] In some embodiments of the present invention, optimizing the baseline plan of the previous cycle using a global reordering strategy includes: The frozen tasks remain unchanged, while the unfrozen tasks are generated using greedy construction and random perturbation to produce diverse initial solutions.

[0014] In some embodiments of the present invention, after obtaining port resource information, vessel information, and the baseline plan of the previous cycle, the method further includes: Based on port resource information, ship information, and the baseline plan of the previous period, an initial population is generated, and a hierarchical collaborative encoder is used to pair and map time-series individuals and berth individuals into time-space scheduling plans.

[0015] The port rolling collaborative scheduling method provided by this invention, which considers task uncertainty, optimizes the order of ship entry and exit and berth allocation within a rolling time window. It introduces a driving discrimination mechanism based on the severity of interference and a plan release rule of "freezing executed plans and updating unexecuted plans" to quickly output executable scheduling schemes that meet berth constraints and navigation sequence constraints. This effectively avoids the problems of increased ship waiting and large-scale failure of port tasks caused by changes in port tasks in traditional static planning. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is one of the flowcharts of the port rolling collaborative scheduling method considering task uncertainty provided in the embodiments of the present invention; Figure 2 This is the second flowchart of the port rolling collaborative scheduling method considering task uncertainty provided in the embodiments of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0020] "A and / or B" includes the following three combinations: A only, B only, and a combination of A and B.

[0021] The use of "applies to" or "configured to" in this invention implies an open and inclusive language, which does not exclude the applicability to or configuration to devices performing additional tasks or steps. Additionally, the use of "based on" implies openness and inclusivity, because processes, steps, calculations, or other actions "based on" one or more conditions or values ​​may in practice be based on additional conditions or values ​​beyond those conditions.

[0022] In this invention, the term "exemplary" is used to mean "serving as an example, illustration, or description." Any embodiment described as "exemplary" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0023] The following describes, with reference to the accompanying drawings, a port rolling collaborative scheduling method considering task uncertainty provided by an embodiment of the present invention.

[0024] like Figure 1 As shown, this embodiment of the invention provides a port rolling collaborative scheduling method considering task uncertainty, the method comprising the following steps: S101: Obtain port resource information, vessel information, and the baseline plan of the previous cycle, and construct the objective function, berth constraints, and navigation sequence constraints for port collaborative scheduling optimization based on the port resource information and vessel information.

[0025] In some examples, port resource information includes channel information, berth resources, anchorage resources, tidal window information, and wharf operation resources. Vessel information includes vessel type, vessel draft, estimated time of arrival, cargo quantity, and length.

[0026] S102, under berth constraints and navigation sequence constraints, performs conflict identification on the baseline plan of the previous cycle to obtain the severity of interference from mission uncertainty disturbances.

[0027] S103. Determine the optimization strategy of the baseline plan of the previous cycle based on the severity of the interference, and under the determined optimization strategy, determine the evaluation value according to the objective function, and obtain the collaborative scheduling plan based on the evaluation value.

[0028] The port rolling collaborative scheduling method considering task uncertainty provided in this invention optimizes the order of ship entry and exit and berth allocation within a rolling time window. It introduces a driving discrimination mechanism based on the severity of interference and a plan release rule of "freezing executed and not executed" to quickly output an executable scheduling scheme that meets berth constraints and navigation sequence constraints. This effectively avoids the problem of increased ship waiting and large-scale failure of port tasks caused by changes in port tasks in traditional static planning.

[0029] In some embodiments of the present invention, determining the optimization strategy for the baseline plan of the previous cycle based on the severity of interference includes: Interference severity Less than or equal to the interference threshold In this case, the baseline plan of the previous cycle is optimized by using a local interference strategy, and the search of the population is restricted to be adjusted within the preset range of unfrozen tasks, which are ship tasks that are in an unexecuted or interruptible state.

[0030] If the severity of the interference exceeds the interference threshold, a global reordering strategy is used to optimize the baseline plan of the previous cycle, and the order, berth, and resource variables of unfrozen tasks are released for reordering.

[0031] In some examples, if port task demand fluctuates slightly, such as minor adjustments to the expected arrival time of individual vessels, short-term changes in operation duration, or time shifts of a few tasks, only short-term demand fluctuations occur, but the consequences do not affect the substantial changes in resources such as vessel sequence and berth allocation in the baseline plan. In this case, the baseline plan is maintained, and information is only updated for the affected vessels.

[0032] If congestion at nearby ports or waterway control measures cause unplanned vessels to request temporary operations or cancel scheduled tasks, resulting in conflicts for only a few vessels regarding berths and waterway resources during the rolling period, an interference management strategy can be adopted. Interference management mainly involves inserting or exchanging local arrival and departure sequences, fine-tuning berthing and departure times and resource allocation, and eliminating local conflicts while maintaining the baseline plan as much as possible.

[0033] If the changes in mission requirements are extremely drastic during a certain period, resulting in the cancellation or concentrated arrival of a large number of established plans, forced priority of urgent tasks, or large-scale restructuring of mission operation requirements, the baseline plan will become invalid and serious conflicts of berth and channel resources will occur among most ships, then a global rescheduling strategy will be initiated.

[0034] If no disruptions occur due to changes in task requirements within the current rolling decision cycle, the baseline plan will be executed as scheduled. After rolling forward one cycle, the port management receives a task change notification and assesses that operational requirements will change in future cycles. If continuing to follow the baseline plan would lead to conflicts in channel and berth resources for a few vessels, a disruption management strategy will be applied to the affected vessels, while other vessels will continue to follow the baseline plan. Rolling forward another cycle, as the task changes intensify, the port task time window is dynamically restructured, and severe conflicts in berth and channel resources for most vessels are identified. This moment is designated as the rescheduling decision point. At this node, the system will regenerate an optimal integrated plan for coordinated scheduling of vessels and berths / tugboats based on the updated task requirements.

[0035] In some embodiments of the present invention, under berth constraints and navigation sequence constraints, conflict identification is performed on the baseline plan of the previous cycle to obtain the severity of disturbances caused by mission uncertainty, including: Calculate the channel conflict index under berth constraints and navigation sequence constraints respectively. Berth conflict indicators Resource Exceeding Targets and planned stability loss indicators .

[0036] Among them, waterway conflict indicators The result is calculated as follows: For all vessels entering the port in the current baseline plan, check whether the difference between the start times of two vessels' entry into the port is less than the prescribed safe navigation distance; the same applies to departure. The sum of the number of channel conflicts constitutes this indicator, calculated as follows: ; In the formula, and These are the port entry start times for vessels i and j, respectively. To ensure safe distances for ships entering and leaving port, and These are the departure start times for vessels i and j, respectively.

[0037] Berth Conflict Indicators The berthing time of any two vessels assigned to the same berth overlaps. The amount of overlap constitutes this indicator, calculated as follows:

[0038] In the formula, For port berth assembly, For the order of operations at the same berth, refer to the variable representing the sequence of operations. and These are the start berthing times for vessels i and j, respectively. For ships Homework time.

[0039] Resource Exceeding Limits This indicator is calculated as follows: It checks whether the total demand for dockside resources such as tugboats and pilots exceeds the capacity limit during the current time period. The sum of demands exceeding the capacity limit constitutes this indicator, calculated as follows:

[0040] in, It is a ship The required amount of a certain type of operational resource; It refers to the quantity of a certain type of operational resource at the target port; if ships In time period Start the assignment ,but =1, otherwise It is 0.

[0041] Planned stability loss index The calculation is as follows: By comparing the new task set with the baseline plan, the number of ships with time deviations exceeding a threshold or the cumulative deviation is counted. The calculation method is as follows:

[0042] In the formula, For ships Start Task time, To execute the planned vessels Start time of the task.

[0043] The severity of the disturbance is obtained by weighting and summing the indicators of channel conflict, berth conflict, resource overrun, and planned stability loss. (Illustratively, the severity of the disturbance is...) In the formula, .

[0044] In some embodiments of the present invention, an objective function for port collaborative scheduling optimization, along with berth constraints and navigation sequence constraints, are constructed based on port resource information and vessel information, including: The ship waiting time is defined as the difference between the ship's arrival start time and its estimated arrival time. An objective function is constructed to minimize this waiting time, and is expressed as follows: ; In the formula, This is the start time of the ship's arrival in port. This is the estimated arrival time of vessel i. It is the effective set of ships at the k-th decision time.

[0045] Based on the compatibility between berth length and ship length, the order of operations of different ships at the same berth, and the non-overlapping berth occupancy time, berth constraints are constructed.

[0046] In some examples, berth constraints include: ; ; ; ; In the formula, B represents the set of port berths; It is a 0 / 1 variable, if the ship Assigned to berth If the value is 1, then the value is 1; otherwise, the value is 0. It is a variable representing the order of operations at the same berth, if the ship At berth Prior to ships If the assignment is correct, the value is 1; otherwise, it is 0. It is a ship length; It is a berth length; , They are ships Commencement time, vessel Departure time.

[0047] Based on the temporal relationship between the safe interval between ships entering and leaving the port, tidal windows, estimated arrival time, port entry sailing time, berthing operation time and departure time, a navigation sequence constraint is constructed.

[0048] In some examples, flight sequence constraints include: ; ; ; ; ; ; ; ; ; ; ; In the formula, For ship operation task types, , This indicates the anchoring and waiting phase. Indicates the port entry and navigation phase. Indicates the berthing phase. Indicates the departure phase. This indicates that the ship's schedule cannot be modified. This indicates that the ship plan can be modified; For ships Is it necessary to enter or leave the port with the tide? It is a ship Current task type; It is a 0 / 1 variable, if the ship In time period Start the assignment If the value is 1, then the value is 1; otherwise, the value is 0. It's a 15-minute interval. ; It is the time of high tide for both arrival and departure from the port. Large ships are not allowed to enter or leave the port. ; , They are ships In time period Commence port entry and departure procedures; It is a ship Start Task time; It refers to the safe distance for ships entering and leaving the port. It is a 0 / 1 variable, if the ship On the ship If the port was previously entered or left, the value is 1; otherwise, the value is 0. It is the estimated sailing time for a ship from entering the port to reaching its berth; It is a ship Homework time; , They are ships Start time of entry into port, vessel Commencement berthing time.

[0049] The port rolling collaborative scheduling method considering task uncertainty provided in this invention aims to minimize ship waiting time. It proactively resolves berth occupancy conflicts and channel conflicts through conflict location and repair mechanisms, improves the executability and continuity of scheduling instructions, significantly reduces ship waiting time in port, and enhances the robustness and real-time response capability of the port under task-changing scenarios.

[0050] In some embodiments of the present invention, under a defined optimization strategy, determining the evaluation value based on the objective function includes: Under the determined optimization strategy, the objective function value of each candidate scheduling scheme is calculated using the objective function. That is to say, = .

[0051] Feasibility checks are performed on berth constraints and navigation sequence constraints, and penalty values ​​are determined based on the check results. In some examples, feasibility checks are performed on constraints such as tidal windows, safety intervals, berth physical adaptation, resource capacity, and task timing. When a constraint is violated, a penalty item should be generated. .

[0052] Specifically, the tidal window constraint is as follows: ; ; The safety interval constraint is:

[0053] ; The berth physical adaptation constraints are: ; ; Resource capacity and task timing constraints are as follows: ; ; ; ; ; .

[0054] Penalty items The calculation method is as follows:

[0055] in, The amount of violation of the above constraint c (such as time exceeding the limit, length exceeding the limit, etc.). To constrain the weight corresponding to c, this weight can be selected based on the target port capacity or the needs and preferences of management personnel. If the target is to reduce time, then a relatively high weight can be set for the time constraint.

[0056] The evaluation value is determined based on the objective function value and the penalty value. Illustratively, the evaluation value... The feasible solution satisfies , This is a penalty function used during the search process to prioritize eliminating infeasible individuals without changing the model's definition, which only uses waiting time as the objective function. In some embodiments of the present invention, obtaining a cooperative scheduling plan based on an evaluation value includes: Individuals in the population are ranked based on their evaluation values, and elite retention is performed.

[0057] Based on fitness values, a selection operation is performed on the retained pairs to form a parent set. The selection operation employs either tournament selection or roulette wheel selection, prioritizing the retention of high-performing individuals with lower objective function values. An elite retention mechanism is used to directly copy the currently best feasible individual into the next generation. When objective function values ​​are similar, the individual with the lowest plan modification cost is selected first to maximize the plan's consistency and stability in rolling scenarios.

[0058] Based on the parent set, and without strictly altering the genes corresponding to frozen tasks, crossover and mutation operations are performed on unfrozen tasks to obtain the offspring set. Specifically, for the time-series group, conflict-guided sequential crossover is performed to adjust the order of port entry and exit operations; for the resource group, block crossover or redistribution mutation is performed to adjust berth assignments. If conflicts arise regarding channel safety intervals, berth occupancy, or resource overruns, the unfrozen task causing the conflict is identified and removed, forming a partially feasible solution. Then, the earliest feasible insertion or lowest cost insertion rule is used to re-insert the conflicting unfrozen task, and hierarchical repair is performed according to the "channel-berth-time consistency" order to ensure that offspring individuals meet stringent physical and time consistency constraints.

[0059] The fitness value of the child set is calculated and constraints are verified. Based on the fitness value and constraint verification results, it is determined whether the current child set meets preset conditions. If the preset conditions are met, a collaborative scheduling plan is determined from the final child set. These preset conditions include reaching the maximum number of iterations or achieving stable convergence of the optimal objective function value. The collaborative scheduling plan is the optimal collaborative scheduling scheme for the current cycle, including ship sequence and berth allocation. Rolling cycle scheduling task instructions are generated according to the rule of "freezing the executed parts unchanged and only updating the unexecuted parts".

[0060] In addition, after obtaining the collaborative scheduling plan, the collaborative scheduling plan is used as the baseline plan for the next rolling cycle to achieve continuous rolling optimization.

[0061] In some embodiments of the present invention, ranking individuals in the population based on evaluation values ​​and performing elite retention includes: When the difference in waiting time between two individuals is less than the difference threshold, the cost of modifying the baseline plan of the previous period for the two individuals is compared, and the individual with the smaller modification cost is retained to ensure the stability of the plan in the rolling scenario.

[0062] Indicatively, modify the cost It is an indicator that measures the extent of change of the new plan relative to the baseline plan of the previous cycle, and is calculated as follows:

[0063] in, The start time of the mission for vessel i during the execution process. This corresponds to the time in the baseline plan; The berth allocated to vessel i Assign berths in the baseline plan; This is an indicator function; it returns 1 if the berth location changes. The relative weights of time deviation and berth change.

[0064] In some embodiments of the present invention, optimizing the baseline plan of the previous cycle using a global reordering strategy includes: The frozen tasks remain unchanged, while the unfrozen tasks are generated using greedy construction and random perturbation to produce diverse initial solutions.

[0065] In some embodiments of the present invention, after obtaining port resource information, vessel information, and the baseline plan of the previous cycle, the method further includes: Based on port resource information, ship information, and the baseline plan of the previous period, an initial population is generated, and a hierarchical collaborative encoder is used to pair and map time-series individuals and berth individuals into time-space scheduling plans.

[0066] In some examples, the set of active ships is obtained at the rolling decision time t. Berth assembly Port tidal window set and the previous cycle benchmark plan and determine the freeze indicator. ,in This indicates that the corresponding ship mission is in an executed or uninterruptible state and remains unchanged in this round of optimization. Based on this, an initial population is generated. A hierarchical co-encoder is used to pair and map time-series individuals and berth individuals to specific time-space scheduling plans. The time is extrapolated according to the time sequence and resources such as waterways and berths are verified simultaneously. The baseline plan of the previous cycle is inherited first. Multiple variants are generated by making minimal changes to newly added, canceled, and deviated tasks; when a global reordering is triggered, frozen tasks remain unchanged and diverse initial solutions are generated for unfrozen tasks using greedy construction and random perturbation.

[0067] In some embodiments of the present invention, such as Figure 2 As shown, a port rolling collaborative scheduling method considering task uncertainty includes the following steps: S201, rolling decision time t, obtain ship, berth, tidal window and baseline plan, and determine the freeze task.

[0068] Specifically, at the rolling decision time t, the system acquires vessel information, berth resource information, tidal window information, and the baseline plan from the previous period within the current rolling time window. Vessel information may include vessel type, draft, estimated arrival time, cargo quantity, length, and current task type; berth resource information may include berth set, berth length, and berth occupancy status; tidal window information constrains the arrival and departure times of vessels requiring high tide; and the baseline plan represents the vessel arrival and departure sequence, berth allocation, and task start time established in the previous rolling period. Based on this, and depending on whether a vessel's task has already been executed or is uninterruptible, tasks are frozen to ensure they remain unchanged during subsequent optimization processes.

[0069] S202, perform hierarchical collaborative coding to generate an initial population that includes resource verification.

[0070] Specifically, ship time-series individuals are paired and mapped to berth allocation individuals to corresponding time-space scheduling plans. When generating individuals, resource checks such as channel safety distance, berth physical adaptation, tidal window, task timing, and resource capacity are performed simultaneously, so that the initial population can represent multiple candidate collaborative scheduling schemes.

[0071] S203, determine whether the maximum number of iterations or the optimal value convergence is satisfied.

[0072] Specifically, determine whether the current optimization process has reached the maximum number of iterations or whether the optimal fitness value has converged stably. If the iteration stopping condition is met, execute S214; if the iteration stopping condition is not met, execute S204 to continue the population iterative optimization.

[0073] S204, perform fitness selection, and retain the best feasible solution to copy into the next generation.

[0074] Specifically, selection is based on the fitness values ​​of individuals in the current population, prioritizing feasible individuals with better fitness as subsequent iteration targets. Simultaneously, the current optimal feasible solution is directly copied into the next generation to prevent the loss of high-quality scheduling schemes during iteration.

[0075] S205, calculate the degree of interference, identify constraint conflicts, and calculate the impact of interference and the loss of plan stability respectively.

[0076] Specifically, the baseline plan is constrained under the current rolling cycle to identify conflict issues, including conflicts related to safe channel clearance, berth occupancy, resource overruns, and plan stability loss indicators.

[0077] S206 calculates the waiting time for ships and generates penalty terms for violations of constraints such as tides, safe distances, and timing.

[0078] Specifically, the waiting time for each individual vessel in the current population is calculated. The waiting time can be determined based on the difference between the actual arrival time or start time of the vessel and the expected arrival time. Simultaneously, constraints such as tidal window, safety distance, berth adaptation, resource capacity, and task timing are verified. If any constraint is violated, a penalty term is generated for the corresponding violation, which is used in subsequent evaluation value calculations and individual selection.

[0079] S207, perform scheduling strategy selection, and compare the impact of interference with the threshold.

[0080] Specifically, the severity of the interference obtained from S205 With interference threshold A comparison is made to determine the scheduling strategy used in the current rolling cycle. If the severity of the interference... Not greater than the interference threshold If the interference severity is high, then execute S208; Greater than the interference threshold If so, then S209 will be executed.

[0081] S208 employs a local interference strategy to restrict minor adjustments to unfrozen tasks.

[0082] Specifically, in terms of interference severity Not greater than the interference threshold In such cases, a local interference strategy is adopted. In this case, the frozen tasks remain unchanged, and the search range is limited to small adjustments to the unfrozen tasks. By locally inserting, exchanging, or fine-tuning the arrival and departure sequence, berthing and departure times, or berth allocation of the unfrozen tasks, a small number of local conflicts are eliminated, while maintaining the stability of the baseline plan as much as possible.

[0083] S209 employs a global rearrangement strategy, releasing unfrozen tasks and rearranging all variables.

[0084] Specifically, in terms of interference severity Greater than the interference threshold In this case, a global rescheduling strategy is adopted. The frozen tasks remain unchanged, while the ship sequence variables, berth allocation variables, and related resource variables corresponding to the unfrozen tasks are released. The unfrozen tasks are then fully rescheduled to generate new candidate scheduling schemes that meet the current task requirements and resource constraints.

[0085] S210, under the determined strategy, sorts and retains elites based on evaluation values.

[0086] Specifically, under either a local disturbance strategy or a global reordering strategy, evaluation values ​​are generated based on factors such as ship waiting time, constraint violation penalties, and plan stability losses. Individuals in the population are then ranked according to these evaluation values. Individuals with superior evaluation values ​​are retained as elites. When multiple individuals have similar ship waiting times, those with smaller modifications to the baseline plan are prioritized for retention to improve the continuity of the rolling scheduling plan.

[0087] S211, generate the parent set.

[0088] Specifically, based on the S210 sorting and the population individuals after elite retention, a selection operation is performed to generate the parent set. The selection operation can employ tournament selection, roulette wheel selection, or other fitness selection methods to ensure that individuals with higher evaluation values ​​have a higher probability of entering the parent set.

[0089] S212 involves crossover and mutation of unfrozen tasks while strictly preserving frozen genes.

[0090] Specifically, crossover and mutation operations are performed on individuals in the parent set. During the crossover and mutation process, the genes corresponding to frozen tasks remain strictly unchanged, and only the genes corresponding to unfrozen tasks are processed. Sequential crossover or sequential mutation can be performed on the ship time sequence genes of unfrozen tasks, and block crossover or redistribution mutation can be performed on the berth allocation genes to generate the offspring set.

[0091] S213, perform conflict level repair, inserting repairs in the order of "channel-berth-sequence consistency".

[0092] Specifically, conflict checks are performed on the scheduling individuals in the child set. If conflicts exist regarding channel safety clearance, berth occupancy, or task timing, the unfrozen tasks causing the conflicts are located, and hierarchical repairs are performed in the order of "channel—berth—timing consistency." Specifically, the channel passage sequence and safety clearance are checked and repaired first, then berth occupancy and berth adaptation relationships are checked and repaired, and finally the timing consistency between arrival, berthing, operation, and departure is checked and repaired, ensuring that the repaired child individuals meet the scheduling constraints.

[0093] S214, recalculate offspring fitness values ​​and check constraints.

[0094] Specifically, after the conflict level is repaired, the fitness value of the offspring individuals is recalculated, and constraints such as tidal window, safety distance, berth resources, task timing, and resource capacity are checked again. The offspring individuals that have been recalculated and checked are used as candidates for the next iteration, and the process returns to S203 to continue to determine whether the maximum number of iterations or the optimal value convergence condition is met.

[0095] S215 outputs the optimal cooperative scheduling scheme.

[0096] Specifically, when S203 determines that the maximum number of iterations or the optimal value convergence condition is met, it outputs the currently obtained optimal collaborative scheduling scheme. The optimal collaborative scheduling scheme includes the ship arrival and departure sequence, berth allocation results, berthing time, departure time, and related resource usage plan.

[0097] S216 generates rolling cycle scheduling task instructions and updates the plan for continuous optimization.

[0098] Specifically, based on the optimal collaborative scheduling scheme output in S215, scheduling task instructions for the current rolling cycle are generated, and the scheduling plan is updated. For frozen tasks that have already been executed or cannot be interrupted, the original plan remains unchanged; for tasks that have not been executed, they are updated according to the optimal collaborative scheduling scheme for this cycle. The updated plan serves as the baseline plan for the next rolling cycle, so that S201 to S216 can continue to be executed at the new rolling decision point, thereby achieving continuous rolling optimization of the port scheduling plan.

[0099] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the parts that contribute to the prior art, can be embodied in the form of software products. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0100] The above provides a detailed description of a port rolling collaborative scheduling method considering task uncertainty provided by the embodiments of the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A port rolling collaborative scheduling method considering task uncertainty, characterized in that, The method includes: Obtain port resource information, vessel information, and the baseline plan of the previous cycle, and construct an objective function for port collaborative scheduling optimization, as well as berth constraints and navigation sequence constraints based on the port resource information and vessel information; Under the constraints of berth and navigation sequence, conflict identification is performed on the baseline plan of the previous cycle to obtain the severity of disturbances caused by mission uncertainty. The optimization strategy for the baseline plan of the previous cycle is determined based on the severity of the interference. Under the determined optimization strategy, the evaluation value is determined according to the objective function, and the collaborative scheduling plan is obtained based on the evaluation value.

2. The method according to claim 1, characterized in that, The optimization strategy for determining the baseline plan of the previous cycle based on the severity of the interference includes: If the severity of the interference is less than or equal to the interference threshold, it is determined that the baseline plan of the previous cycle will be optimized using a local interference strategy, and the search of the population will be restricted to a preset range within the unfrozen tasks, wherein the unfrozen tasks are ship tasks that are in an unexecuted or interruptible state. If the severity of the interference is greater than the interference threshold, it is determined that the baseline plan of the previous cycle will be optimized using a global reordering strategy, and the order, berth, and resource variables of unfrozen tasks will be released for reordering.

3. The method according to claim 1, characterized in that, Under the constraints of berth and navigation sequence, conflict identification is performed on the baseline plan of the previous cycle to obtain the severity of disturbances to mission uncertainty, including: Under the aforementioned berth constraints and navigation sequence constraints, the channel conflict index, berth conflict index, resource overrun index, and plan stability loss index are calculated respectively. The severity of the disturbance is obtained by weighted summation of the channel conflict index, berth conflict index, resource overrun index, and plan stability loss index.

4. The method according to claim 1, characterized in that, Based on the port resource information and vessel information, an objective function for port collaborative scheduling optimization, along with berth constraints and navigation sequence constraints, is constructed, including: The vessel waiting time is defined as the difference between the vessel's arrival start time and its estimated arrival time. An objective function is constructed to minimize this waiting time, and the objective function is characterized as follows: ； In the formula, This is the start time of the ship's arrival in port. This is the estimated arrival time of vessel i. It is the effective set of ships at the k-th decision time; The berth constraints are constructed based on the compatibility between berth length and ship length, the order of operations of different ships on the same berth, and the non-overlapping berth occupancy time. The navigation sequence constraints are constructed based on the temporal relationship between the safe interval between ships entering and leaving the port, the tidal window, the estimated arrival time, the port entry sailing time, the berthing operation time and the departure time.

5. The method according to claim 4, characterized in that, Under a defined optimization strategy, the evaluation value is determined based on the objective function, including: Under the determined optimization strategy, the objective function value of each candidate scheduling scheme is calculated using the objective function. The feasibility of the berth constraints and navigation sequence constraints is verified, and the penalty value is determined based on the verification results. The evaluation value is determined based on the objective function value and the penalty value.

6. The method according to claim 1, characterized in that, The process of obtaining a collaborative scheduling plan based on the evaluation value includes: The population individuals are ranked based on their evaluation values, and elite retention is performed. Based on fitness values, a selection operation is performed on the retained pairs to form a parent set; Based on the parent set, crossover and mutation operations are performed on the unfrozen tasks to obtain the child set; The fitness value of the child set is calculated and the constraints are checked. Based on the fitness value and the constraint check results, it is determined whether the current child set meets the preset conditions. If the preset conditions are met, the collaborative scheduling plan is determined from the final child set.

7. The method according to claim 6, characterized in that, The process of performing crossover and mutation operations on unfrozen tasks includes: For time-series groups, conflict-guided sequential crossovers are implemented to adjust the order of inbound and outbound operations; For resource groups, perform block crossover or redistribution mutations to adjust berth assignments; If a conflict occurs in the channel safety interval, berth occupancy, or resource overrun, the unfrozen task that caused the conflict is located and removed to form a partial feasible solution. Then, the unfrozen task that caused the conflict is re-inserted using the earliest feasible insertion rule or the lowest cost insertion rule, and hierarchical repair is performed in the order of "channel-berth-time consistency".

8. The method according to claim 6, characterized in that, The process of ranking individuals in the population based on evaluation values ​​and performing elite retention includes: When the difference in waiting time between two individuals is less than the difference threshold, compare the baseline plan modification costs of the two individuals in the previous cycle, and retain the individual with the smaller modification cost.

9. The method according to claim 2, characterized in that, The optimization of the previous cycle's baseline plan using a global rearrangement strategy includes: The frozen tasks remain unchanged, while the unfrozen tasks are generated using greedy construction and random perturbation to produce diverse initial solutions.

10. The method according to claim 1, characterized in that, After obtaining port resource information, vessel information, and the baseline plan for the previous period, the method further includes: Based on the port resource information, ship information, and the baseline plan of the previous period, an initial population is generated, and a hierarchical collaborative encoder is used to pair and map time-series individuals and berth individuals into time-space scheduling plans.

Images