A method for intelligent deployment of mechanical resources in a dry bulk cargo handling area based on ship scheduling

Abstract

This invention discloses an intelligent allocation method for mechanical resources in dry bulk cargo loading and unloading areas based on ship scheduling, belonging to the field of intelligent scheduling technology for port conveyor systems. Specifically, it includes: constructing an unloading task set based on ship berthing and departure plans and yard stacking information; selecting feasible paths from the port conveyor network and choosing the shortest path as the initial allocation path to form an initial scheduling scheme; automatically detecting time conflicts in the plan by simulating exclusive use of paths and verifying time window compliance, generating a conflict criterion set; collecting the working status of port machinery in real time and integrating it with conflict information to form a path adjustment parameter set; iteratively solving based on the path adjustment parameter set to generate path reallocation and time adjustment schemes; and finally updating the initial scheme based on the optimization results to generate a transportation equipment scheduling execution scheme that can dynamically adapt to changes in the field. This invention effectively improves the collaborative efficiency of port machinery and the robustness of the operation plan.

Description

Technical Field

[0001] This invention relates to the field of intelligent scheduling technology for port transport systems, specifically to an intelligent allocation method for mechanical resources in dry bulk cargo loading and unloading areas based on ship scheduling. Background Technology

[0002] In large ports, dry bulk cargo unloading is a complex process involving multiple stages and equipment coordination. Its core efficiency depends on the precise scheduling and coordination of various mechanical resources such as gantry cranes, ship unloaders, stacker-reclaimers, and belt conveyors. Currently, port production scheduling mainly relies on a combination of manual experience and static planning.

[0003] Specifically, dispatchers typically allocate unloading machinery (such as gantry cranes or ship unloaders) and fixed conveyor belt transport routes to each ship in advance based on the ship's predicted berthing and departure schedule (i.e., scheduling) and yard stacking information, and estimate the operation time to form a static operation plan. However, this traditional method has significant limitations: First, static plans cannot anticipate and resolve dynamic conflicts that inevitably arise during operations. For example, when the operation plans of multiple ships require the use of the same critical conveyor belt route, simply scheduling according to fixed times can easily lead to overlapping equipment occupancy times, resulting in work line interruptions. Second, once the plan is formulated, it lacks flexibility. When unexpected situations occur on-site, such as the failure of a critical unloading machine, temporary maintenance of a conveyor belt, or a ship's operation progress being significantly ahead of schedule or behind schedule, the original static plan quickly becomes invalid. Dispatchers can only rely on experience to make emergency adjustments, which are often local and suboptimal, easily triggering chain reactions, leading to longer waiting times for subsequent ships and reduced equipment utilization. Furthermore, manual scheduling makes it difficult to comprehensively consider all constraints (such as equipment capacity, route exclusivity, and ship time windows) and perform global optimization in a short period of time, often resulting in neglecting some aspects while focusing on others.

[0004] Although existing research has attempted to optimize port scheduling using operations research models or information systems, most solutions still focus on pre-emptive static planning or localized optimization of single equipment (such as conveyor belts). There is a lack of a comprehensive technical solution capable of real-time linkage between ship scheduling, yard stacking, and the status of various types of machinery, as well as continuous status monitoring, dynamic conflict resolution, and global rescheduling during operations. Therefore, significant technical bottlenecks remain in port dry bulk unloading operations regarding efficiency improvement, cost control, and handling of uncertainties. Summary of the Invention

[0005] The purpose of this invention is to provide an intelligent allocation method for mechanical resources in dry bulk cargo loading and unloading areas based on ship scheduling, and to solve the following technical problems: Existing dry bulk unloading machinery scheduling methods based on human experience and static planning are difficult to respond to dynamic changes such as equipment conflicts and mechanical failures in real time during operation, resulting in longer ship waiting times and low overall operation efficiency.

[0006] The objective of this invention can be achieved through the following technical solutions: A method for intelligent allocation of mechanical resources in dry bulk cargo loading and unloading areas based on ship scheduling includes the following steps: S1. Based on ship scheduling information and port stacking information, establish a set of unloading tasks that includes unloading tasks, task time windows, and berth-stacking position correspondence. S2, based on the berth and stacking positions of each unloading task in the unloading task set, select feasible physical transmission paths from the port conveyor network, initially allocate material handling machinery to each unloading task and plan the corresponding physical transmission path and operation time, forming an initial scheduling scheme for task-path. S3. Obtain the planned time interval of each transmission path from the initial scheduling scheme, detect equipment occupation conflicts of different unloading tasks on the same transmission path, extract the tasks with time overlap conflicts and their corresponding transmission paths, and generate a conflict criterion set. S4. Real-time acquisition of the working status of port material handling machinery and belt conveyor and analysis of the time overlap relationship of each transmission path in the conflict criterion set. Combined with the working status of the machinery, construct a path adjustment parameter set including time conflict structure and available equipment resource status. S5, map the path adjustment parameter set to the input parameters of the optimization model, and iteratively solve the optimization model to re-determine the allocation of material handling machinery and physical transport paths for conflicting unloading tasks, and generate path redistribution schemes and time adjustment schemes. S6. Based on the path redistribution scheme and time adjustment scheme, update the physical transmission path and planned time interval corresponding to each unloading task in the initial scheduling scheme to generate a transportation equipment scheduling execution scheme to guide port site operations.

[0007] As a further aspect of the present invention: in S1, the process of generating the unloading task set is as follows: Obtain the scheduling information of vessels scheduled to berth, and extract the berthing time and departure time of each vessel from the scheduling information; generate an unloading task for each vessel based on the berthing time and departure time, and determine the time window of the task as the time period from the berthing time to the departure time; obtain the stacking position allocation information output by the port stacking system, which includes the planned stacking position of the cargo carried by each vessel; combine each unloading task, its corresponding time window, and its associated berth position and planned stacking position to establish an unloading task set that includes the unloading task, the time window, and the berth-stacking position correspondence.

[0008] As a further aspect of the present invention: in step S2, the specific process for generating the initial scheduling scheme is as follows: For each unloading task in the unloading task set, based on its associated berth and stack location, obtain all feasible physical transmission paths that can connect the berth and stack location and their transmission efficiency parameters from the port conveyor network. Obtain the cargo volume for the unloading task and, based on the transmission efficiency parameters of each feasible physical transmission path, calculate the planned occupancy time required to complete the cargo transportation along each feasible physical transmission path. Select the feasible physical transmission path with the shortest planned occupancy time from all feasible physical transmission paths as the initial allocation path for the unloading task; summarize the initial allocation paths of all unloading tasks and their corresponding planned occupancy times to form the initial scheduling scheme of task-path.

[0009] As a further aspect of the present invention: in S3, the specific process for generating the conflict criterion set is as follows: For each physical transmission path in the port conveyor network, all unloading tasks that were selected as the initial allocation path in the initial scheduling scheme are collected; based on the berthing time of the ship corresponding to each unloading task, the execution order of each unloading task on the physical transmission path is determined in chronological order. According to the execution order of each unloading task, the available time of the physical transmission path is allocated to each task in sequence to form its planned time interval. The starting point of the interval for the first unloading task is the starting point of its task time window, and the starting point of the interval for subsequent unloading tasks is immediately after the ending point of the interval for the previous task. The length of the interval for each unloading task corresponds to the planned time of the unloading task. The planned time interval allocated for each unloading task is checked for inclusion with the task time window corresponding to that unloading task. If the check fails, the unloading task and the physical transmission path involved are marked as conflict entries. After traversing all physical transmission paths, all conflict entries are summarized to generate a conflict criterion set.

[0010] As a further aspect of the present invention: the specific process for generating the path adjustment parameter set in S4 is as follows: Obtain the conflict criterion set, and parse the physical transmission path identifier, associated unloading task identifier, and conflict relationship description between the planned time interval and the task time window contained in each conflict record. The current availability status of the physical transmission path related to the conflict record in the port equipment monitoring system is collected in real time, as well as the current working status of the material handling machinery corresponding to the associated unloading task. The conflict relationship description corresponding to each conflict record is associated and encapsulated with the collected corresponding path and the real-time status of the machinery to generate a decision parameter entry containing specific conflict constraints and real-time resource status; the decision parameter entries corresponding to all conflict records are collected to construct the path adjustment parameter set.

[0011] As a further aspect of the present invention: the specific process of generating the path redistribution scheme and the time adjustment scheme in step S5 is as follows: From the unloading task set, extract the cargo quantity corresponding to each unloading task recorded in the conflict criterion set, and calculate its remaining task time window based on the current time and the task time window; The path adjustment parameter set, the extracted cargo volume, and the calculated remaining task time window are mapped together as input parameters for an optimization model. Based on the input parameters, the simulated annealing algorithm is used to iteratively solve the optimization model. In each iteration, a new solution is generated by adjusting the material handling machinery assignment and physical transport path selection of the unloading task in the conflict record. The new solution is accepted based on the change of the model objective function value and the current temperature. At the same time, the temperature is gradually reduced according to a preset rule to control the search process. When the simulated annealing algorithm reaches the preset termination condition, the final solution is converted into the material handling machine identifier, the assigned physical transmission path identifier, and the planned time interval replanned on the path corresponding to the unloading task in each conflict record, thus obtaining the path redistribution scheme and the time adjustment scheme.

[0012] As a further aspect of the present invention: the specific process for generating the remaining task time window is as follows: Obtain the start and end times of the task time window corresponding to the unloading task from the unloading task set; obtain the current time and compare it with the start time of the task time window. If the current time is later than the start time, use the current time as the start time of the remaining task time window; otherwise, use the start time of the task time window as the start time of the remaining task time window; use the end time of the task time window directly as the end time of the remaining task time window to obtain the remaining task time window.

[0013] As a further aspect of the present invention: in step S6, the specific process for generating the transportation equipment scheduling execution plan is as follows: Based on the path redistribution scheme, the physical transmission paths corresponding to each unloading task in the initial scheduling scheme are updated; based on the time adjustment scheme, the planned time intervals occupied by each unloading task on the physical transmission paths in the initial scheduling scheme are updated, thereby generating a transportation equipment scheduling execution scheme.

[0014] The beneficial effects of this invention are: 1) This invention integrates ship berthing and unberthing plans with yard stacking allocation information to establish a structured set of unloading tasks. This clarifies the operational time window and spatial location relationship for each task, laying a data foundation for precise scheduling. Based on this, and using cargo volume and equipment transmission parameters, the invention automatically selects and plans the theoretically shortest operational path and time interval from all feasible paths for each task, embedding efficiency optimization guidance from the initial stage of plan generation. Furthermore, an automated conflict look-ahead detection mechanism is introduced. By simulating the sequential occupancy of equipment and rigorously verifying the conformity between planned time and the inherent time window of each task, it can proactively and automatically identify potential time conflicts arising from multiple tasks competing for the same critical path, making the risk of operational interruption explicit before execution. This process enables the scheduling scheme to have self-examination and conflict early warning capabilities, changing the reactive, post-event handling mode that relies on manual experience, and significantly improving the reliability, feasibility, and ability to rehearse complex operational scenarios.

[0015] 2) This invention constructs a decision parameter system that integrates real-time equipment status and identified conflict structures, enabling the system to dynamically perceive "where the planned conflicts are" and "what the current resource status is." Based on this parameter system, an intelligent optimization algorithm is driven to perform a global iterative search within a complex solution space that includes multiple constraints such as remaining time windows, available paths, and equipment status. This allows for a systematic exploration of various rescheduling possibilities, including reassigning work machinery, switching transmission paths, and adjusting work sequences, and automatic evaluation and comparison, ultimately generating a globally optimized adjustment scheme that coordinates multiple factors. This mechanism endows the scheduling system with powerful online response and dynamic decision-making capabilities, enabling it to effectively cope with sudden interference such as on-site machinery failures and work progress fluctuations, rather than relying on localized and delayed manual emergency handling, thereby ensuring the flexibility, adaptability, and overall robustness of the work plan in the face of uncertainty.

[0016] 3) This invention achieves deep collaboration and intelligent allocation of various mechanical resources, such as gantry cranes, ship unloaders, and belt conveyors, across time and space by constructing a complete technical process encompassing intelligent equipment task planning, automated conflict early warning, real-time data-driven optimization, and closed-loop solution updates. This method ensures that the task allocation and scheduling of each piece of equipment are always within a continuously self-verified and dynamically iteratively optimized global framework, guaranteeing a more rational and balanced allocation of core port resources such as berth operation time and critical transportation routes. Consequently, it drives the entire dry bulk unloading system to form a more collaborative and fluid operational rhythm, specifically resulting in more efficient utilization of vessel time in port, effectively reducing idle and waiting periods for critical equipment resources, and systematically improving the overall continuity, stability, and comprehensive operational efficiency of the operational process. Attached Figure Description

[0017] The invention will now be further described with reference to the accompanying drawings.

[0018] Figure 1 This is a schematic diagram of a method for intelligent allocation of mechanical resources in dry bulk cargo loading and unloading areas based on ship scheduling, according to the present invention. Detailed Implementation

[0019] 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.

[0020] Please see Figure 1 As shown, this invention is an intelligent allocation method for mechanical resources in dry bulk cargo loading and unloading areas based on ship scheduling, comprising the following steps: S1. Based on ship scheduling information and port stacking information, establish a set of unloading tasks that includes unloading tasks, task time windows, and berth-stacking position correspondence. First, the scheduling information of all vessels scheduled to berth during the planning period is obtained from the port's production management system. This information includes at least the planned berthing time, contracted departure time, assigned berth, and type and quantity of cargo to be unloaded for each vessel. Simultaneously, the location coordinates of the specific stacking location assigned to each vessel's cargo are obtained from the port's yard stacking system. Based on this information, a unique unloading task is created for each vessel. The core attributes of this task include: task identifier, associated vessel identifier, operating berth location, target stacking location, and a rigid task time window defined by the vessel's berthing and departure times. Finally, all created unloading tasks are structured, organized, and stored to form a global, standardized unloading task set, serving as the unified data source and factual basis for all subsequent scheduling decisions.

[0021] S2, based on the berth and stacking positions of each unloading task in the unloading task set, select feasible physical transmission paths from the port conveyor network, initially allocate material handling machinery to each unloading task and plan the corresponding physical transmission path and operation time, forming an initial scheduling scheme for task-path. After obtaining the set of unloading tasks, for each task, path reachability analysis is performed in a network model representing the topology and connectivity of the port conveyor system, based on the physical locations of its starting point (berth) and ending point (stacking location). All possible path combinations are automatically traversed, and all feasible physical transport paths connecting the berth and the stall are selected. For each feasible path, its preset transport capacity parameters (such as rated capacity and current health status) are invoked. Next, considering the total cargo volume of the unloading task, the shortest theoretically required pure operation time for completing the transport using this path is calculated. Then, following the efficiency-first principle, the path with the shortest calculated operation time is selected from all feasible paths as the "initial allocation path" for that task. Based on the starting point of the task time window, a preliminary continuous planned occupancy time interval is planned. After performing this process for all tasks, a preliminary "task-path-time" correspondence table is obtained, which constitutes the initial scheduling scheme. This step aims to find a locally optimal operation path for each task, providing an optimized starting point for global coordination.

[0022] S3. Obtain the planned time interval of each transmission path from the initial scheduling scheme, detect equipment occupation conflicts of different unloading tasks on the same transmission path, extract the tasks with time overlap conflicts and their corresponding transmission paths, and generate a conflict criterion set. The initial scheme was generated based on the assumption of single-task independence and did not consider the exclusivity of key port resources (such as conveyor belt paths). Therefore, this step performs global conflict detection. Using each physical transmission path as an analysis unit, all unloading tasks selected as the initial allocation path in the initial scheme are aggregated. Then, based on the berthing time sequence of the vessels associated with these tasks, their theoretical operation order on that path is determined. Next, a resource exclusivity scenario is simulated: starting from the planned start time of the first task, a continuous operation time period is allocated to it, the length of which is equal to its required operation time; after completion, the next task is immediately allocated a time period. After allocation, the "planned occupied time interval" obtained by each task in this simulated exclusivity scenario is compared and verified with its own inherent "task time window". If the planned occupied time interval of a task does not fall completely within its own task time window (for example, the simulation end time exceeds the vessel's departure time), then a "plan and window conflict" is determined to exist. Such conflicts, along with the involved unloading task identifier, physical transmission path identifier, and specific conflict type (such as timeout), are recorded to form clear conflict records. The collection of all conflict records constitutes the "conflict criterion set," which precisely points out the crux of why the initial plan is infeasible.

[0023] S4. Real-time acquisition of the working status of port material handling machinery and belt conveyor and analysis of the time overlap relationship of each transmission path in the conflict criterion set. Combined with the working status of the machinery, construct a path adjustment parameter set including time conflict structure and available equipment resource status. The conflict criterion set reflects the inherent contradictions of the static plan, while actual scheduling also needs to deal with dynamic changes. This step aims to provide a comprehensive input for dynamic decision-making that integrates static conflicts and real-time status. On the one hand, the data stream of port equipment monitoring is continuously accessed to obtain the real-time working status of key machinery (including material handling machinery such as gantry cranes / ship unloaders, and belt conveyor drive units, etc.), such as whether it is online, whether it is faulty, current load, and estimated idle time. On the other hand, the conflict criterion set generated by S3 is parsed to understand the specific constraints of each conflict (e.g., "Task A needs to operate on path P until time X, but time X has exceeded its time window"). Subsequently, for each conflict record, the static "conflict structure" description contained therein is associated and encapsulated with the real-time "working status" of the machinery directly related to the conflict (such as the belt conveyor corresponding to the conflict path, the material handling machinery specified by the conflict task) collected at the current time. For example, the encapsulated information may be: "Conflict path P1, current status is idle; associated task TA, its specified gantry crane M1 is currently faulty, estimated recovery time is T1; alternative gantry crane M2 is currently idle." In this way, each conflict record is transformed into a "parameter entry" rich in decision-making information, and all entries are collected to form a "path adjustment parameter set." This parameter set is the precise input for subsequent optimization solutions.

[0024] S5, map the path adjustment parameter set to the input parameters of the optimization model, and iteratively solve the optimization model to re-determine the allocation of material handling machinery and physical transport paths for conflicting unloading tasks, and generate path redistribution schemes and time adjustment schemes. The path adjustment parameter set constructed by S4, along with information such as the relevant task cargo volume extracted from the unloading task set and the remaining task time window recalculated based on the current time, are standardized into a set of input parameters for an optimization model. This optimization model aims to minimize total operation delay and resource switching costs, and its constraints cover task time windows, path transmission capacity, equipment status, and path exclusivity. Since this scheduling problem is a complex combinatorial optimization problem, this invention employs the simulated annealing algorithm for solution. The algorithm starts with a random or current-state-based initial solution that defines the machinery and path assignments for all conflicting tasks. During iteration, the path assignments of some tasks are randomly changed or their operation times are fine-tuned (generating "neighborhood solutions"), and the objective function value of the new solution is calculated. Based on the core mechanism of simulated annealing, not only are better new solutions accepted, but also temporarily worse solutions are accepted with a certain probability, which decreases as the "temperature" parameter gradually decreases. This strategy helps the algorithm escape local optima and search towards the global optimum. When the termination condition is met (such as low temperature or sufficient iterations), the algorithm stops and decodes the final solution. The decoding result clearly indicates the number of material handling machines to be used, the physical transport path to be used, and the precise start and end times of the adjusted operation for each original conflicting task, thus outputting the "path reallocation scheme" and "time adjustment scheme" respectively.

[0025] S6. Based on the path redistribution scheme and time adjustment scheme, update the physical transmission path and planned time interval corresponding to each unloading task in the initial scheduling scheme to generate a transportation equipment scheduling execution scheme to guide port site operations.

[0026] According to the "path reallocation scheme," the corresponding unloading tasks in the initial scheduling scheme are identified, and their "physical transmission path" attribute is updated to the optimized new path. Simultaneously, according to the "time adjustment scheme," the "planned time interval" attribute of these tasks on the corresponding paths is updated to the optimized, re-planned time period. For tasks not in the conflict criterion set and not affected by this optimization, their arrangements in the initial scheme remain unchanged. After this update operation, a new, globally consistent "transportation equipment scheduling execution scheme" is generated. This scheme incorporates the efficiency optimization starting point of the initial scheme and overcomes resource conflicts and adapts to real-time equipment status through conflict resolution and dynamic adjustment, possessing high executability. Finally, this scheme is sent to port production execution to guide the actual operation of equipment such as gantry cranes, ship unloaders, and belt conveyors, thereby achieving intelligent, dynamic, and globally optimized allocation of dry bulk cargo unloading machinery resources.

[0027] It is worth noting that the "task modeling-conflict detection-state fusion-dynamic optimization" intelligent scheduling logic constructed in this solution is essentially a general resource collaborative optimization framework independent of logistics direction. Therefore, this solution is also applicable to bulk cargo loading operations. Its core principle is that loading operations can be viewed as the reverse process of unloading operations. Both essentially involve clearly defined "start-end" transmission tasks, ship operation plans constrained by time windows, a physical network composed of conveyor belts and various loading and unloading machinery, and conflicts arising from multiple tasks competing for limited resources. In the loading scenario, the system constructs a loading task set using the ship loading plan and the stacker reclaimer positions in the yard as input; based on the flow direction from the stacker positions to the loading berth, it filters transmission paths within the same conveyor belt network and forms an initial loading scheduling scheme; it identifies path preemption conflicts through the same time overlap conflict detection mechanism; and it integrates the status of equipment such as loading machines and stacker reclaimers in real time, driving the optimization algorithm to reschedule the path and machinery allocation, ultimately generating an equipment execution scheme for the loading operation. By integrating unloading and loading into a unified model for optimization, this solution achieves system-level coordination in resource allocation, time coordination, and conflict resolution for two-way bulk cargo logistics at ports, thereby improving the overall system throughput efficiency and scheduling intelligence throughout the entire loading and unloading process.

[0028] In another preferred embodiment of the present invention, the process of generating the unloading task set in S1 is as follows: Obtain the scheduling information of vessels scheduled to berth, and extract the berthing time and departure time of each vessel from the scheduling information; generate an unloading task for each vessel based on the berthing time and departure time, and determine the time window of the task as the time period from the berthing time to the departure time; obtain the stacking position allocation information output by the port stacking system, which includes the planned stacking position of the cargo carried by each vessel; combine each unloading task, its corresponding time window, and its associated berth position and planned stacking position to establish an unloading task set that includes the unloading task, the time window, and the berth-stacking position correspondence.

[0029] In another preferred embodiment of the present invention, the specific process of generating the initial scheduling scheme in step S2 is as follows: For each unloading task in the unloading task set, based on its associated berth and stack location, obtain all feasible physical transmission paths that can connect the berth and stack location and their transmission efficiency parameters from the port conveyor network. Obtain the cargo volume for the unloading task and, based on the transmission efficiency parameters of each feasible physical transmission path, calculate the planned occupancy time required to complete the cargo transportation along each feasible physical transmission path. Select the feasible physical transmission path with the shortest planned occupancy time from all feasible physical transmission paths as the initial allocation path for the unloading task; summarize the initial allocation paths of all unloading tasks and their corresponding planned occupancy times to form the initial scheduling scheme of task-path.

[0030] For each unloading task in the unloading task set, based on its associated berth and stack location, all feasible physical transmission paths connecting the berth and stack, along with their transmission efficiency parameters, are obtained from the port's conveyor network. This step is achieved by comparing the task's specified start and end points with the port's pre-set equipment topology map; only physically connected routes with permissible equipment conditions are listed as feasible paths. Next, the cargo volume for the unloading task is obtained, and based on the transmission efficiency parameters of each feasible physical transmission path, the planned occupancy time required to complete cargo transportation along each feasible physical transmission path is calculated. The principle is to estimate the pure operation time by dividing the total cargo volume by the transmission efficiency of the path. The transmission efficiency reflects the amount of cargo that the path can stably transport per unit time. Therefore, dividing the two yields the theoretical shortest time. Then, the feasible physical transmission path with the shortest planned occupancy time is selected from all feasible physical transmission paths as the initial allocation path for the unloading task. This is to implement the principle of efficiency priority in a local scope, aiming to find the theoretically fastest way to complete each independent task. Finally, the initial allocation paths of all unloading tasks and their corresponding planned occupancy times are summarized to form the initial task-path scheduling scheme.

[0031] First, by independently selecting and binding each task with a theoretically shortest path, efficiency optimization is instilled in every operation from the outset, laying the foundation for minimizing overall operation time. This changes the arbitrariness or habituality of path selection that may exist in manual experience-based allocation. Second, generating such a structured initial scheduling scheme provides a clear and specific target for subsequent collaborative analysis and conflict detection. It avoids complex global optimization in a vacuum, instead establishing a locally optimal but potentially contradictory "blueprint." This provides a clear basis and target for subsequent detection of equipment conflicts, time window violations, and other issues. Finally, while considering computational complexity, it provides a high-quality optimization starting point for the entire intelligent dispatching process. It is neither the final solution nor far superior to random or empirical arrangements, allowing subsequent conflict resolution and dynamic adjustments to proceed on a better baseline. This leads to a more efficient convergence to a globally feasible high-quality scheduling scheme, playing a crucial foundational and guiding role in achieving the overall goals of reducing ship waiting time and improving machinery utilization.

[0032] In another preferred embodiment of the present invention, the specific process of generating the conflict criterion set in step S3 is as follows: For each physical transmission path in the port conveyor network, all unloading tasks that were selected as the initial allocation path in the initial scheduling scheme are collected; based on the berthing time of the ship corresponding to each unloading task, the execution order of each unloading task on the physical transmission path is determined in chronological order. According to the execution order of each unloading task, the available time of the physical transmission path is allocated to each task in sequence to form its planned time interval. The starting point of the interval for the first unloading task is the starting point of its task time window, and the starting point of the interval for subsequent unloading tasks is immediately after the ending point of the interval for the previous task. The length of the interval for each unloading task corresponds to the planned time of the unloading task. The planned time interval allocated for each unloading task is checked for inclusion with the task time window corresponding to that unloading task. If the check fails, the unloading task and the physical transmission path involved are marked as conflict entries. After traversing all physical transmission paths, all conflict entries are summarized to generate a conflict criterion set.

[0033] For each physical transmission path in the port conveyor network, the first step is to identify all unloading tasks planned to use that path in the initial scheduling scheme. This is achieved by traversing the path identifiers assigned to each task in the initial scheduling scheme and grouping tasks with the same path identifier together. For example, if path P1 is selected as the initial allocation path by both task A and task B, A and B will be grouped together as the pending tasks for path P1. Next, based on the berthing time of the vessel corresponding to each unloading task, the execution order of each unloading task on that physical transmission path is determined in chronological order. The principle is that the vessel berthing time represents the earliest possible start time of the operation, and sorting according to the basic logic of first-come, first-served can simulate the most natural task execution sequence, thus providing a basis for subsequent time allocation. Then, according to the execution order of each unloading task, the available time of the physical transmission path is allocated to each task in sequence to form its planned time interval. The starting point of the interval for the first unloading task is the starting point of its task time window, because the first task can start using the path the earliest. The starting point of the interval for subsequent unloading tasks is immediately after the ending point of the interval for the previous task. This simulates the physical reality that the path, as an exclusive resource, can only serve one task at a time. The interval length of each unloading task corresponds to the planned time occupied by the unloading task, that is, the theoretical pure operation time required by the task on the selected path extracted from the initial scheduling scheme. After allocation, the planned time interval for each unloading task is checked against the corresponding task time window. This involves verifying that the start of the planned interval is no earlier than the start of the time window and that the end of the planned interval is no later than the end of the time window. If the check fails, it indicates that the task cannot be completed within the vessel's allowed time frame under the exclusive queuing mode, and the unloading task and its associated physical transport path are marked as conflict entries. For example, if the planned end time of task B exceeds the vessel's departure time, then B and path P1 are recorded as a conflict. Finally, all physical transport paths are traversed, and the above steps of aggregation, sorting, allocation, and verification are repeated for each path. All conflict entries are then aggregated to generate a conflict criterion set, resulting in a comprehensive list of all plans that are infeasible due to path contention or time window limitations.

[0034] First, by simulating task queuing and time allocation along each path, the system realistically reflects the resource competition issues that may arise from the exclusivity of key port equipment during actual operations. This makes the global conflicts hidden in locally optimal plans explicit, avoiding the lag in traditional manual scheduling where conflicts only surface during execution. Second, based on the mechanism of berthing time sorting and continuous time allocation, the system follows the common first-come, first-served principle in operational sites, while strictly verifying the compliance of time windows. This ensures that the identified conflicts are rigid contradictions that will inevitably occur in reality, rather than theoretical assumptions, providing highly credible targets for subsequent adjustments. Finally, the core purpose of generating the conflict criterion set is to provide precise optimization guidance for the entire intelligent dispatching system. Like a "diagnostic report," it clearly identifies weaknesses in the plan, enabling subsequent dynamic adjustments to be targeted and focused on resolving real and urgent conflicts. This significantly improves the executability and overall efficiency of the dispatching scheme, playing a crucial early warning and support role in achieving the ultimate goal of reducing ship waiting time and efficient coordination of mechanical equipment.

[0035] It is worth noting that, in actual implementation, the task execution order determination mechanism in this solution is not limited to the simple order of ship berthing times. Multi-dimensional decision factors can also be introduced for comprehensive priority assessment. For example, multiple factors can be considered, such as the urgency of the cargo owner's contract, the type of cargo (e.g., whether it is easily oxidized or a high-value mineral), the volume of cargo, yard turnover needs, and ship demurrage risks, to calculate a dynamic priority weight for each unloading task. In practical applications, these factors can be quantified into unified comparable indicators through preset or configurable weight rules, thereby globally ranking multiple unloading tasks converging on the same physical transmission path, thus forming an operational execution sequence that better aligns with actual production and business needs. This achieves intelligent operational scheduling based on a balance of value, efficiency, and risk, further improving the precision of port operation scheduling and overall operational efficiency. Its specific implementation can be adapted and deployed according to actual business needs, and will not be elaborated upon here.

[0036] In another preferred embodiment of the present invention, the specific process of generating the path adjustment parameter set in step S4 is as follows: Obtain the conflict criterion set, and parse the physical transmission path identifier, associated unloading task identifier, and conflict relationship description between the planned time interval and the task time window contained in each conflict record. The current availability status of the physical transmission path related to the conflict record in the port equipment monitoring system is collected in real time, as well as the current working status of the material handling machinery corresponding to the associated unloading task. The conflict relationship description corresponding to each conflict record is associated and encapsulated with the collected corresponding path and the real-time status of the machinery to generate a decision parameter entry containing specific conflict constraints and real-time resource status; the decision parameter entries corresponding to all conflict records are collected to construct the path adjustment parameter set.

[0037] First, obtain the conflict criterion set generated in the previous step. This is a structured list recording all identified scheduling conflicts. Read each record in this list; each record clearly records three types of key information: a unique identifier of the physical transmission path where the conflict occurred (e.g., "conveyor P1"), a unique identifier of the unloading task related to the conflict (e.g., "task A"), and a specific description of the conflict relationship—usually manifested as details of the inconsistency between the planned time interval allocated to the task and the task's own allowed time window (e.g., "planned operation on P1 until 15:00, but task A's ship departure time is 14:00"). This allows for an accurate understanding of the essence of each conflict: which equipment resource, which task, and what time conflict made the plan infeasible. Next, dynamic, real-world information needs to be injected into these static conflict descriptions. This involves connecting in real-time to the port equipment monitoring data interface to proactively query the latest status of equipment directly related to the conflict record currently being processed. For example, for "belt conveyor P1" mentioned in the record, it will determine whether it is currently in normal operation, under maintenance, or scheduled maintenance. For "task A" mentioned in the record, it will determine the current status of the designated or potentially used material handling equipment (such as gantry crane M1), whether it is idle, busy, or offline. The principle behind this step is that a conflict resolution is highly dependent on the actual availability of relevant resources at any given moment. If P1 is faulty, the solution must consider alternative paths; if M1 is busy, it is necessary to assess when it will be idle or whether there are alternative machines available. Then, the static "conflict relationship description" parsed from each conflict record is associated and encapsulated with the "real-time status" of the corresponding path and machinery, generating a new and richer decision parameter entry. This association process essentially binds the problem (conflict) to the available real-world conditions (resource status) for solving the problem. For example, an entry might be generated like this: "Conflict: Task A will time out on path P1; Status: P1 is currently normal but has been booked until 13:00 when it becomes available. Task A's original gantry crane M1 is malfunctioning and expected to be repaired at 14:00. The alternative gantry crane M2 is currently available and can connect to path P2." Finally, the above parsing, collection, association, and encapsulation operations are repeated for all conflict records. By gathering all the generated decision parameter entries together, a "path adjustment parameter set" is constructed. This set is no longer just a list of problems, but a comprehensive information database that clearly defines each problem to be solved and provides a complete picture of the currently available resource constraints for solving the problem.

[0038] First, it creatively integrates the static conflicts discovered through offline planning analysis with the dynamic equipment status feedback from online production. This ensures that subsequent optimization decisions are no longer based on idealized or outdated assumptions, but rather on real snapshots of port operations, greatly improving the reliability and timeliness of the decision-making basis. Second, by constructing independent and fully informed decision parameter entries for each conflict, it effectively decomposes the complex global scheduling problem into a series of clearly defined sub-decision units. This lays a structured data foundation for efficient solutions using optimization algorithms, enabling the algorithms to accurately understand the specific constraints and options faced by each adjustment action. Ultimately, the core purpose of this step is to create a high-quality "input material library" for generating feasible and realistic dynamic adjustment solutions. It ensures that the subsequent intelligent optimization process does not occur in a vacuum, but closely revolves around real conflicts and utilizes real available resources to find answers. This directly and powerfully supports the ultimate goal of the invention—to quickly generate efficient and executable scheduling solutions when facing planning conflicts and sudden changes in the field, thereby achieving the fundamental purpose of reducing non-operational waiting time for ships and improving the comprehensive utilization rate of key port machinery.

[0039] In another preferred embodiment of the present invention, the specific process of generating the path redistribution scheme and the time adjustment scheme in step S5 is as follows: From the unloading task set, extract the cargo quantity corresponding to each unloading task recorded in the conflict criterion set, and calculate its remaining task time window based on the current time and the task time window; The path adjustment parameter set, the extracted cargo volume, and the calculated remaining task time window are mapped together as input parameters of an optimization model. The input parameters include the conflict relationship between the planned occupied time interval and the task time window corresponding to each conflict record in the parameter set, the feasible physical transmission paths and material handling machinery status that are available other than the currently occupied path. Based on the input parameters, the simulated annealing algorithm is used to iteratively solve the optimization model. In each iteration, a new solution is generated by adjusting the material handling machinery assignment and physical transport path selection of the unloading task in the conflict record. The new solution is accepted based on the change of the model objective function value and the current temperature. At the same time, the temperature is gradually reduced according to a preset rule to control the search process. When the simulated annealing algorithm reaches the preset termination condition, the final solution is converted into the material handling machine identifier, the assigned physical transmission path identifier, and the planned time interval replanned on the path corresponding to the unloading task in each conflict record, thus obtaining the path redistribution scheme and the time adjustment scheme.

[0040] In another preferred embodiment of the present invention, the specific process for generating the remaining task time window is as follows: Obtain the start and end times of the task time window corresponding to the unloading task from the unloading task set; obtain the current time and compare it with the start time of the task time window. If the current time is later than the start time, use the current time as the start time of the remaining task time window; otherwise, use the start time of the task time window as the start time of the remaining task time window; use the end time of the task time window directly as the end time of the remaining task time window to obtain the remaining task time window.

[0041] First, the start and end times of the task time window corresponding to the unloading task are obtained from the unloading task set. This is because the unloading task set serves as the core data source for the entire scheduling process. Each task is associated with an allowable operating period determined by the ship's berthing and departure times. The earliest start time and latest end time for the task can be found using its unique identifier. For example, for the task "Unloading the Bohai Vessel," the task time window starts at 08:00 and ends at 18:00. Next, the current time is obtained. This time is usually derived from a server clock synchronized with port production to ensure a consistent and accurate time base; for example, the current time is 10:00. Then, the current time is compared with the start time of the task time window. The principle of this comparison is that the actual operable time of the task cannot be earlier than the planned start time. However, if the current time is later than the start time, it means that the interval from the start time to the current time has been consumed or wasted due to various reasons (such as equipment failure, waiting for resources, etc.). The task cannot be backdated to an earlier start time, so the remaining available time can only be calculated from the current moment. If the current time is later than the start time, the current time is used as the start time of the remaining task time window. If the current time is not later than the start time, it means the task has not yet reached the permitted start time, and the original start time is used as the start time of the remaining window. For example, if the current time is 10:00, which is later than the start time of 08:00, the start time of the remaining window is adjusted to 10:00; if the current time is 07:00, the start time remains unchanged at 08:00. Finally, the end time of the task time window is directly used as the end time of the remaining task time window. This is because the ship's departure time is a rigid constraint that cannot be postponed, and the operation must be completed before this time. Therefore, the end time of the remaining window is consistent with the original window. For example, the task end time of 18:00 remains unchanged. Through the above steps, a continuous time period from the adjusted start time to the original end time is finally obtained, which is the remaining task time window, such as 10:00 to 18:00 in the example above. This window accurately represents the maximum available working time that the task actually has from the current time until the deadline.

[0042] First, by introducing the dynamic variable of current time, it transforms the static planned time window into a remaining time window synchronized with the actual progress of operations. This makes the measurement of time resources more closely aligned with the rapidly changing realities on-site, avoiding misjudgments of time resources caused by using fixed windows. Second, by comparing the current time with the start time and taking the later one as the starting point of the remaining window, it automatically deducts elapsed invalid or waiting time, ensuring that the time constraints on which subsequent scheduling decisions are based are real, available, and urgent. This provides an accurate basis for assessing the urgency of tasks and optimizing job sequencing. Finally, the core purpose of this step is to provide key and accurate time boundary parameters for the dynamic rescheduling process. This ensures that the optimization algorithm, when reallocating paths and equipment, can fully consider the real time pressure faced by each task, thereby generating feasible solutions that both meet the hard deadline for ship departure and make the most efficient use of remaining time. This plays a crucial supporting role in achieving the ultimate goal of reducing ship waiting time in port, improving the continuity of operational processes, and enhancing overall efficiency. It enables scheduling to respond to the challenges brought by the passage of time in a timely manner and make smarter, more appropriate decisions.

[0043] In another preferred embodiment of the present invention, the specific process of generating the transportation equipment scheduling execution plan in step S6 is as follows: Based on the path redistribution scheme, the physical transmission paths corresponding to each unloading task in the initial scheduling scheme are updated; based on the time adjustment scheme, the planned time intervals occupied by each unloading task on the physical transmission paths in the initial scheduling scheme are updated, thereby generating a transportation equipment scheduling execution scheme.

[0044] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A method for intelligent allocation of mechanical resources in dry bulk cargo loading and unloading areas based on ship scheduling, characterized in that, Includes the following steps: S1. Based on ship scheduling information and port stacking information, establish a set of unloading tasks that includes unloading tasks, task time windows, and berth-stacking position correspondence. S2, based on the berth and stacking positions of each unloading task in the unloading task set, select feasible physical transmission paths from the port conveyor network, initially allocate material handling machinery to each unloading task and plan the corresponding physical transmission path and operation time, forming an initial scheduling scheme for task-path. S3. Obtain the planned time interval of each transmission path from the initial scheduling scheme, detect equipment occupation conflicts of different unloading tasks on the same transmission path, extract the tasks with time overlap conflicts and their corresponding transmission paths, and generate a conflict criterion set. S4. Real-time acquisition of the working status of port material handling machinery and belt conveyor and analysis of the time overlap relationship of each transmission path in the conflict criterion set. Combined with the working status of the machinery, construct a path adjustment parameter set including time conflict structure and available equipment resource status. S5, map the path adjustment parameter set to the input parameters of the optimization model, and iteratively solve the optimization model to re-determine the allocation of material handling machinery and physical transport paths for conflicting unloading tasks, and generate path redistribution schemes and time adjustment schemes. S6. Based on the path redistribution scheme and time adjustment scheme, update the physical transmission path and planned time interval corresponding to each unloading task in the initial scheduling scheme to generate a transportation equipment scheduling execution scheme to guide port site operations.

2. The intelligent allocation method for mechanical resources in dry bulk cargo loading and unloading areas based on ship scheduling as described in claim 1, characterized in that, In S1, the process of generating the unloading task set is as follows: Obtain the scheduling information of vessels scheduled to berth, and extract the berthing time and departure time of each vessel from the scheduling information; generate an unloading task for each vessel based on the berthing time and departure time, and determine the time window of the task as the time period from the berthing time to the departure time; obtain the stacking position allocation information output by the port stacking system, which includes the planned stacking position of the cargo carried by each vessel; combine each unloading task, its corresponding time window, and its associated berth position and planned stacking position to establish an unloading task set that includes the unloading task, the time window, and the berth-stacking position correspondence.

3. The intelligent allocation method for mechanical resources in dry bulk cargo loading and unloading areas based on ship scheduling according to claim 1, characterized in that, In S2, the specific process of generating the initial scheduling scheme is as follows: For each unloading task in the unloading task set, based on its associated berth and stack location, obtain all feasible physical transmission paths that can connect the berth and stack location and their transmission efficiency parameters from the port conveyor network. Obtain the cargo volume for the unloading task and, based on the transmission efficiency parameters of each feasible physical transmission path, calculate the planned occupancy time required to complete the cargo transportation along each feasible physical transmission path. Select the feasible physical transmission path with the shortest planned time from all feasible physical transmission paths as the initial allocation path for this unloading task. The initial allocation paths of all unloading tasks and their corresponding planned occupancy times are summarized to form the initial scheduling scheme of tasks and paths.

4. The intelligent allocation method for mechanical resources in dry bulk cargo loading and unloading areas based on ship scheduling according to claim 1, characterized in that, In S3, the specific process for generating the conflict criterion set is as follows: For each physical transmission path in the port conveyor network, all unloading tasks that were selected as the initial allocation path in the initial scheduling scheme are collected; based on the berthing time of the ship corresponding to each unloading task, the execution order of each unloading task on the physical transmission path is determined in chronological order. According to the execution order of each unloading task, the available time of the physical transmission path is allocated to each task in sequence to form its planned time interval. The starting point of the interval for the first unloading task is the starting point of its task time window, and the starting point of the interval for subsequent unloading tasks is immediately after the ending point of the interval for the previous task. The length of the interval for each unloading task corresponds to the planned time of the unloading task. The planned time interval allocated for each unloading task is checked for inclusion with the task time window corresponding to that unloading task. If the verification fails, the unloading task and the physical transmission path involved are marked as conflict entries; after traversing all physical transmission paths, all conflict entries are summarized to generate a conflict criterion set.

5. The intelligent allocation method for mechanical resources in dry bulk cargo loading and unloading areas based on ship scheduling according to claim 1, characterized in that, In step S4, the specific process for generating the path adjustment parameter set is as follows: Obtain the conflict criterion set, and parse the physical transmission path identifier, associated unloading task identifier, and conflict relationship description between the planned time interval and the task time window contained in each conflict record. The current availability status of the physical transmission path related to the conflict record in the port equipment monitoring system is collected in real time, as well as the current working status of the material handling machinery corresponding to the associated unloading task. The conflict relationship description corresponding to each conflict record is associated and encapsulated with the collected corresponding path and the real-time status of the machinery to generate a decision parameter entry containing specific conflict constraints and real-time resource status; the decision parameter entries corresponding to all conflict records are collected to construct the path adjustment parameter set.

6. The intelligent allocation method for mechanical resources in dry bulk cargo loading and unloading areas based on ship scheduling according to claim 1, characterized in that, In step S5, the specific process of generating the path redistribution scheme and the time adjustment scheme is as follows: From the unloading task set, extract the cargo quantity corresponding to each unloading task recorded in the conflict criterion set, and calculate its remaining task time window based on the current time and the task time window; The path adjustment parameter set, the extracted cargo volume, and the calculated remaining task time window are mapped together as input parameters of an optimization model. Based on the input parameters, the simulated annealing algorithm is used to iteratively solve the optimization model. In each iteration, a new solution is generated by adjusting the material handling machinery assignment and physical transport path selection of the unloading task in the conflict record. The new solution is accepted based on the change of the model objective function value and the current temperature. At the same time, the temperature is gradually reduced according to a preset rule to control the search process. When the simulated annealing algorithm reaches the preset termination condition, the final solution is converted into the material handling machine identifier, the assigned physical transmission path identifier, and the planned time interval replanned on the path corresponding to the unloading task in each conflict record, thus obtaining the path redistribution scheme and the time adjustment scheme.

7. The intelligent allocation method for mechanical resources in dry bulk cargo loading and unloading areas based on ship scheduling according to claim 6, characterized in that, The specific process for generating the remaining task time window is as follows: Obtain the start and end times of the task time window corresponding to the unloading task from the unloading task set; obtain the current time and compare it with the start time of the task time window. If the current time is later than the start time, use the current time as the start time of the remaining task time window; otherwise, use the start time of the task time window as the start time of the remaining task time window; use the end time of the task time window directly as the end time of the remaining task time window to obtain the remaining task time window.

8. The intelligent allocation method for mechanical resources in dry bulk cargo loading and unloading areas based on ship scheduling according to claim 1, characterized in that, In step S6, the specific process for generating the transportation equipment scheduling execution plan is as follows: Based on the path redistribution scheme, the physical transmission paths corresponding to each unloading task in the initial scheduling scheme are updated; based on the time adjustment scheme, the planned time intervals occupied by each unloading task on the physical transmission paths in the initial scheduling scheme are updated, thereby generating a transportation equipment scheduling execution scheme.

Images