Scheduling method and device for concrete mixing plant and storage medium

Abstract

The application discloses a scheduling method and device for a concrete mixing station and a storage medium, and relates to the technical field of material scheduling. The method comprises the following steps: acquiring multi-source state data in real time, wherein the multi-source state data comprises concrete mixing station production state data, concrete mixer truck transportation state data and construction site demand state data; determining a plan deviation degree according to the multi-source state data and a pre-scheduling scheme, and identifying and judging whether a disturbance event occurs according to the plan deviation degree; in the case of determining that the disturbance event occurs, performing a construction site material interruption risk assessment on the disturbance event to determine a material interruption risk index corresponding to a construction site affected by the disturbance event; determining a target hierarchical response strategy corresponding to the disturbance event according to the plan deviation degree and the material interruption risk index; and determining a final scheduling scheme of the concrete mixing station according to the pre-scheduling scheme and the target hierarchical response strategy, so as to execute the final scheduling scheme. The application can improve scheduling efficiency.

Description

Technical Field

[0001] This application relates to the field of logistics scheduling technology, specifically to a scheduling method, device, and storage medium for concrete mixing plants. Background Technology

[0002] Production scheduling at a concrete batching plant is a complex and dynamic process. Even the most comprehensive pre-planning can encounter various unforeseen circumstances during actual execution, such as: sudden vehicle breakdowns: transport vehicles experience mechanical failures en route, preventing them from arriving at the construction site or returning to the batching plant on time; traffic congestion: unexpected traffic conditions (such as traffic accidents or temporary traffic control) significantly extend vehicle transport time; changes in construction site demand: the construction site temporarily adjusts the pouring speed, suspends pouring, or increases the amount of concrete required; production line malfunctions: temporary malfunctions occur in the batching plant's production line, affecting concrete production capacity; sudden weather changes: severe weather (such as heavy rain or fog) affects transportation safety and speed, or causes construction to be suspended at the site.

[0003] Currently, in response to these emergencies, real-time dispatching at concrete mixing plants mainly relies on the dispatcher's experience for reactive, "firefighting" handling. Dispatchers obtain on-site information via telephone, walkie-talkie, etc., and then adjust vehicle assignments, notify construction sites to wait, or coordinate production lines based on their experience. However, this existing dispatching method suffers from low efficiency. Summary of the Invention

[0004] The purpose of this application is to provide a scheduling method, device, and storage medium for concrete mixing plants to solve the problem of low scheduling efficiency in the prior art.

[0005] To achieve the above objectives, the first aspect of this application provides a scheduling method for a concrete mixing plant, the scheduling method comprising:

[0006] After determining the pre-scheduling plan for the concrete batching plant, multi-source status data is acquired in real time, including concrete batching plant production status data, concrete mixer truck transportation status data, and construction site demand status data. Based on multi-source state data and pre-scheduling schemes, determine the plan deviation and identify whether a disturbance event has occurred based on the plan deviation. In the event of a disturbance, a risk assessment of material shortage at the construction site is conducted to determine the corresponding material shortage risk index for the affected construction site. Based on the deviation from the plan and the material shortage risk index, determine the target-level response strategy corresponding to the disturbance event; Based on the pre-scheduling plan and the target hierarchical response strategy, the final scheduling plan for the concrete mixing plant is determined and executed.

[0007] In this application embodiment, the plan deviation includes at least one of the following: time deviation, determined based on the difference between the actual or expected time of the task node and the planned time in the pre-scheduling scheme; resource deviation, determined based on the change in the availability status of key resources of the concrete mixing plant; and demand deviation, determined based on the difference between the actual demand at the construction site and the planned demand at the construction site in the pre-scheduling scheme.

[0008] In this embodiment of the application, the material shortage risk index is determined based on the estimated arrival time of the next concrete mixer truck at the construction site, the estimated unloading end time of the concrete mixer truck currently unloading at the construction site, and the maximum allowable material shortage interval at the construction site.

[0009] In this embodiment, the target hierarchical response strategy corresponding to the disturbance event is determined based on the plan deviation and the material shortage risk index. This includes: when the plan deviation is within a low plan deviation range and the material shortage risk index is within a low material shortage risk index range, the target hierarchical response strategy for the disturbance event is determined to be non-intervention, monitoring only; when the plan deviation is within a low plan deviation range and the material shortage risk index is within a high material shortage risk index range, the target hierarchical response strategy for the disturbance event is determined to be local fine-tuning; when the plan deviation is within a high plan deviation range and the material shortage risk index is within a low material shortage risk index range, the target hierarchical response strategy for the disturbance event is determined to be local rearrangement with rolling window optimization; and when the plan deviation is within a high plan deviation range and the material shortage risk index is within a high material shortage risk index range, the target hierarchical response strategy for the disturbance event is determined to be global rearrangement.

[0010] In this application embodiment, local fine-tuning includes at least one of the following operations: shifting task time within the time buffer reserved in the pre-scheduling scheme, performing single-point task exchange, and switching production lines within the concrete mixing plant.

[0011] In this embodiment, the local rearrangement step of the rolling window optimization includes: determining the rolling optimization window based on a preset window duration, starting from the current time; rearranging and optimizing the tasks within the rolling optimization window with the goal of minimizing the total deviation cost within the rolling optimization window; wherein, the total deviation cost includes task delay cost, vehicle usage cost, material shortage risk penalty, and penalty for the vehicle status at the tail of the rolling optimization window being inconsistent with the original plan; the rearrangement optimization adopts a fast genetic algorithm with an elite retention strategy.

[0012] In this embodiment of the application, the final scheduling scheme for the concrete mixing plant is determined based on the pre-scheduling scheme and the target hierarchical response strategy, including: creating a shadow copy of the pre-scheduling scheme; running the target hierarchical response strategy on the shadow copy to generate a shadow plan; performing a difference quantification analysis on the shadow plan and the shadow copy, providing conflict warning prompts for the shadow plan, and allowing the scheduler to adjust or confirm the shadow plan; and in response to the scheduler's confirmation instruction regarding the shadow plan, determining the shadow plan confirmed by the scheduler as the final scheduling scheme.

[0013] In this embodiment of the application, after executing the final scheduling scheme, the method further includes: performing feedback learning based on the execution results to optimize the parameters of subsequent pre-scheduling schemes or hierarchical response strategies. The feedback learning includes: recording disturbance events, target hierarchical response strategies, scheduler intervention behaviors, and execution results, and using the recorded data to dynamically optimize the time buffer parameters, material shortage risk index threshold, or cost function weights of the rearrangement algorithm in the pre-scheduling scheme through reinforcement learning algorithms or online learning algorithms.

[0014] A second aspect of this application provides a scheduling device for a concrete batching plant, comprising: a memory configured to store instructions; and a processor configured to retrieve instructions from the memory and, when executing the instructions, to implement the scheduling method for the concrete batching plant described above.

[0015] A third aspect of this application provides a machine-readable storage medium storing instructions for causing a machine to perform the scheduling method for a concrete mixing plant as described above.

[0016] The aforementioned technical solution, after determining the pre-scheduling plan for the concrete mixing plant, acquires multi-source status data in real time. By comparing the multi-source status data with the pre-scheduling plan, the plan deviation is obtained, which can accurately identify and quantify various sudden disturbance events. For disturbance events of different impact levels, different levels of graded response strategies are pre-set. By determining the material shortage risk index corresponding to the affected construction site, and based on the plan deviation and the material shortage risk index, the corresponding target graded response strategy is selected to minimize the impact of disturbance events on the overall plan, avoiding unnecessary resource waste and unreasonable resource use. This achieves proactive intervention in disturbance events and rapid optimization of the pre-scheduling plan, enabling the rapid generation of a high-quality final scheduling plan. Adopting different strategies according to the degree of disturbance impact ensures rapid handling of minor disturbances while providing systematic rearrangement capabilities for complex disturbances, improving scheduling efficiency and system robustness. By efficiently responding to emergencies, it reduces vehicle idling and waiting time, lowers operating costs, and improves the operational efficiency and risk resistance of the concrete mixing plant. Simultaneously, it ensures timely concrete supply to construction sites, improving site satisfaction and trust.

[0017] Other features and advantages of the embodiments of this application will be described in detail in the following detailed description section. Attached Figure Description

[0018] The accompanying drawings are provided to further illustrate the embodiments of this application and form part of the specification. They are used together with the following detailed description to explain the embodiments of this application, but do not constitute a limitation on the embodiments of this application. In the drawings: Figure 1 The illustration shows a schematic flowchart of a scheduling method for a concrete mixing plant according to an embodiment of this application. Figure 2 The schematic diagram illustrates a flow chart of a scheduling method for a concrete mixing plant according to another embodiment of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only for illustration and explanation of the embodiments of this application and are not intended to limit the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0020] It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of this application all comply with relevant laws and regulations. In the embodiments of this application, certain existing industry solutions such as software, components, and models may be mentioned. These should be considered exemplary, intended only to illustrate the feasibility of implementing the technical solution of this application, and do not imply that the applicant has already used or necessarily used such solutions.

[0021] It should be noted that if the embodiments of this application involve directional indicators (such as up, down, left, right, front, back, etc.), the directional indicators are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indicators will also change accordingly.

[0022] Furthermore, if the embodiments of this application involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, features defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the technical solutions of various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. If the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed in this application.

[0023] Figure 1 The illustration schematically shows a flow chart of a scheduling method for a concrete mixing plant according to an embodiment of this application. Figure 1 As shown in the figure, this application provides a scheduling method for a concrete mixing plant. Taking the application of this scheduling method to a processor as an example, the scheduling method may include the following steps.

[0024] Step S102: After determining the pre-scheduling plan for the concrete batching plant, multi-source status data is acquired in real time. The multi-source status data includes the production status data of the concrete batching plant, the transportation status data of the concrete mixer trucks, and the demand status data of the construction site.

[0025] Step S104: Determine the plan deviation based on the multi-source state data and the pre-scheduling scheme, and identify and judge whether a disturbance event has occurred based on the plan deviation.

[0026] Step S106: If a disturbance event is determined to have occurred, conduct a site material shortage risk assessment to determine the corresponding material shortage risk index for the site affected by the disturbance event.

[0027] Step S108: Determine the target-level response strategy corresponding to the disturbance event based on the plan deviation and the material shortage risk index.

[0028] Step S110: Determine the final scheduling plan for the concrete batching plant based on the pre-scheduling plan and the target hierarchical response strategy, and execute the final scheduling plan.

[0029] It can be understood that the pre-scheduling scheme for a concrete batching plant is a predetermined scheduling plan. Multi-source status data refers to status information from multiple sources, including but not limited to concrete mixer trucks, construction sites, and the concrete production lines of the batching plant. Concrete batching plant production status data refers to status information related to the production of the concrete batching plant, including but not limited to the production status information of the concrete production line, such as production line malfunctions. Concrete mixer truck transportation status data refers to status information related to the transportation of concrete mixer trucks, such as sudden truck malfunctions or traffic congestion. Construction site demand status data refers to status information related to the construction site, such as changes in construction site demand. Plan deviation describes the degree of difference between the current state and the original plan; that is, it is used to quantify the deviation between real-time multi-source status data and the predetermined pre-scheduling scheme, and may include time deviation and / or demand deviation. Since the existence of plan deviation does not necessarily indicate the occurrence of a disturbance event, the plan deviation can be compared with a preset plan deviation threshold to identify whether a disturbance event has occurred. Understandably, the construction site material shortage risk assessment is an assessment of whether the construction site can continuously pump materials. If a material shortage occurs, continuous pumping cannot be achieved. The material shortage risk index is used to assess the degree of risk of a material shortage at a construction site affected by a disturbance event. The higher the value of the material shortage risk index, the higher the risk of material shortage at that site. The target graded response strategy is the graded response strategy corresponding to the disturbance event. It can be understood that the graded response strategy can be predetermined or pre-set, that is, there is a correspondence between the plan deviation and the material shortage risk index and the graded response strategy. Based on the plan deviation and the material shortage risk index, the target graded response strategy corresponding to the disturbance event can be determined from the preset graded response strategies. The final scheduling scheme is the final executable scheduling scheme for the concrete mixing plant generated based on the pre-scheduling scheme and the target graded response strategy.

[0030] Specifically, after determining the pre-scheduling plan for the concrete batching plant, the processor can acquire multi-source status data in real time. This multi-source status data includes the production status data of the concrete batching plant, the transportation status data of the concrete mixer trucks, and the demand status data of the construction site. Based on the multi-source status data and the pre-scheduling plan, the processor determines the plan deviation between the two and identifies whether a disturbance event has occurred based on the plan deviation. If a disturbance event is determined, a construction site material shortage risk assessment is performed to determine the material shortage risk index corresponding to the construction site affected by the disturbance event. Based on the plan deviation and the material shortage risk index, the processor determines the target hierarchical response strategy corresponding to the disturbance event. Finally, based on the pre-scheduling plan and the target hierarchical response strategy, the processor determines the final scheduling plan for the concrete batching plant, so that the concrete batching plant executes the scheduling according to the final scheduling plan.

[0031] The aforementioned technical solution, after determining the pre-scheduling plan for the concrete mixing plant, acquires multi-source status data in real time. By comparing the multi-source status data with the pre-scheduling plan, the plan deviation is obtained, which can accurately identify and quantify various sudden disturbance events. For disturbance events of different impact levels, different levels of graded response strategies are pre-set. By determining the material shortage risk index corresponding to the affected construction site, and based on the plan deviation and the material shortage risk index, the corresponding target graded response strategy is selected to minimize the impact of disturbance events on the overall plan, avoiding unnecessary resource waste and unreasonable resource use. This achieves proactive intervention in disturbance events and rapid optimization of the pre-scheduling plan, enabling the rapid generation of a high-quality final scheduling plan. Adopting different strategies according to the degree of disturbance impact ensures rapid handling of minor disturbances while providing systematic rearrangement capabilities for complex disturbances, improving scheduling efficiency and system robustness. By efficiently responding to emergencies, it reduces vehicle idling and waiting time, lowers operating costs, and improves the operational efficiency and risk resistance of the concrete mixing plant. Simultaneously, it ensures timely concrete supply to construction sites, improving site satisfaction and trust.

[0032] In one embodiment, the plan deviation includes at least one of the following: time deviation, determined based on the difference between the actual or expected time of the task node and the planned time in the pre-scheduling scheme; resource deviation, determined based on changes in the availability status of key resources of the concrete mixing plant; and demand deviation, determined based on the difference between the actual demand at the construction site and the planned demand at the construction site in the pre-scheduling scheme.

[0033] As can be understood, a task node is a node in a concrete transportation task, such as the arrival of a concrete mixer truck at the construction site. The actual or estimated time of a task node is the time determined based on the actual site conditions, while the planned time in the pre-scheduling plan is the time of the task node in the pre-arranged scheduling plan based on the site's concrete order demand. Time deviation can be the difference between the actual or estimated time of a task node and the planned time in the pre-scheduling plan. Critical resources can be concrete mixer trucks or concrete production lines, etc. Resource deviation is used to monitor changes in the availability status of critical resources. For example, taking a concrete mixer truck as a critical resource, if the concrete mixer truck malfunctions, the resource deviation value can be 1; if the concrete mixer truck is functioning normally, the resource deviation value can be 0. Similarly, taking a concrete production line as a critical resource, if the concrete production line malfunctions, the resource deviation value can be 1; if the concrete production line is functioning normally, the resource deviation value can be 0. The actual demand at the construction site is the concrete demand determined based on the actual site conditions. The planned demand in the pre-scheduling plan is the concrete demand of the construction site in the pre-scheduled plan based on the concrete order demand of the construction site. In addition, the planned demand can also be directly determined from the concrete order demand of the construction site. The demand deviation can be the difference between the actual demand and the planned demand.

[0034] Specifically, the deviation from the plan can include any one of time deviation, resource deviation, and demand deviation, or any two of time deviation, resource deviation, and demand deviation, or all three of them.

[0035] In this embodiment of the application, the plan deviation of different dimensions is further refined in order to prepare for the identification of disturbance events and improve the accuracy of disturbance event identification.

[0036] In one embodiment, the material shortage risk index is determined based on the estimated arrival time of the next concrete mixer truck at the construction site, the estimated end time of unloading of the concrete mixer truck currently unloading at the construction site, and the maximum allowable material shortage interval at the construction site.

[0037] In one embodiment, a material shortage risk index is determined based on the estimated arrival time of the next concrete mixer truck at the construction site, the estimated unloading end time of the concrete mixer truck currently unloading at the site, and the maximum allowable material shortage interval at the site, including determination according to the following formula: Material shortage risk index = max{0, [Estimated arrival time at construction site - (Estimated unloading end time + Maximum material shortage interval)] / Maximum material shortage interval} For example, suppose that C01 at a construction site is continuously pumping, and the maximum allowable material interruption interval at the site is 15 minutes.

[0038] Scenario 1: Normal situation The current vehicle V01 is expected to finish unloading at 10:30, and the next vehicle V02 is expected to arrive at 10:40. Therefore, the material shortage risk index = max {0, [10:40 - (10:30 + 15 minutes)] / 15 minutes} = max {0, [10:40 - 10:45] / 15 minutes} = max {0, (-5 minutes) / 15 minutes} = max {0, -0.33} = 0. Understandably, a material shortage risk index of 0 indicates no risk of material shortage.

[0039] Scenario 2: Minor delay The current vehicle V01 is expected to finish unloading at 10:30. The next vehicle, V02, is expected to arrive at 10:50 due to slight congestion. Therefore, the material shortage risk index = max {0, [10:50 - (10:30 + 15 minutes)] / 15 minutes} = max {0, [10:50 - 10:45] / 15 minutes} = max {0, (5 minutes) / 15 minutes} = max {0, 0.33} = 0.33. Understandably, a material shortage risk index of 0.33 indicates a slight risk of material shortage, requiring attention and potentially triggering a Level 1 response.

[0040] Scenario 3: Severe Delay The current vehicle V01 is expected to finish unloading at 10:30. The next vehicle, V02, is expected to arrive at 11:00 due to severe congestion or malfunction. Therefore, the material shortage risk index = max {0, [11:00 - (10:30 + 15 minutes)] / 15 minutes} = max {0, [11:00 - 10:45] / 15 minutes} = max {0, (15 minutes) / 15 minutes} = max {0, 1} = 1. Understandably, a material shortage risk index of 1 indicates an extremely high risk of material shortage, meaning the shortage time is equal to the maximum interruption interval, necessitating the activation of a Level 2 or Level 3 response.

[0041] In this embodiment of the application, the material shortage risk index determined based on the above parameters can clearly characterize whether there is a risk of material shortage at the construction site and the degree of such risk, thereby improving the accuracy of the judgment of the risk of material shortage at the construction site.

[0042] In one embodiment, the target hierarchical response strategy corresponding to the disturbance event is determined based on the plan deviation and the material shortage risk index, including: when the plan deviation is within the low plan deviation range and the material shortage risk index is within the low material shortage risk index range, the target hierarchical response strategy corresponding to the disturbance event is determined to be non-intervention and only monitoring; when the plan deviation is within the low plan deviation range and the material shortage risk index is within the high material shortage risk index range, the target hierarchical response strategy corresponding to the disturbance event is determined to be local fine-tuning; when the plan deviation is within the high plan deviation range and the material shortage risk index is within the low material shortage risk index range, the target hierarchical response strategy corresponding to the disturbance event is determined to be local rearrangement with rolling window optimization; when the plan deviation is within the high plan deviation range and the material shortage risk index is within the high material shortage risk index range, the target hierarchical response strategy corresponding to the disturbance event is determined to be global rearrangement.

[0043] As we can understand, the plan deviation describes the degree of difference between the current state and the original plan. For example, a plan deviation of 20 minutes is simply a factual statement, indicating that a vehicle is 20 minutes late. The material shortage risk index, on the other hand, builds upon this by predicting the most severe consequences that the disturbance might cause, i.e., whether it will lead to a disruption in continuous pumping at the construction site. Plan deviation and the material shortage risk index, as two core dimensions driving tiered response decisions, work together to ensure the comprehensiveness and rationality of the decisions. The low plan deviation range is a pre-set range below a preset plan deviation threshold. The high plan deviation range is a pre-set range above a preset plan deviation threshold. The low material shortage risk index range is a pre-set range below a preset material shortage risk index threshold. The high material shortage risk index range is a pre-set range above a preset material shortage risk index threshold.

[0044] Specifically, when the plan deviation is within the low plan deviation range and the material shortage risk index is within the low material shortage risk index range, it indicates that both disturbances and risks are within acceptable limits, falling within the observation zone. In this case, no intervention is needed, only monitoring. When the plan deviation is within the low plan deviation range and the material shortage risk index is within the high material shortage risk index range, it indicates that the plan deviation is not significant, but signs of risk have emerged. A Level 1 response should be adopted, where local fine-tuning can be implemented, such as using buffer zones or small-scale task exchanges, to quickly eliminate risks while maintaining the original plan to the greatest extent possible. When the plan deviation is within the high plan deviation range and the material shortage risk index is within the low material shortage risk index range... In cases where the plan deviation is within a certain range, it indicates a significant deviation from the plan. Although there is no immediate risk of material shortage, without intervention, subsequent tasks will be significantly delayed, increasing future risks. A Level 2 response is adopted, in which case a local rescheduling with rolling window optimization can be initiated. This local rescheduling aims to minimize costs while ensuring safety. When the plan deviation is within the high plan deviation range and the material shortage risk index is within the high material shortage risk index range, it indicates a serious plan deviation and an imminent risk of material shortage. A Level 3 response is adopted, where "ensuring safety" is the top priority. A global rescheduling can be initiated, even at the expense of some economic objectives (such as increasing transportation costs), to ensure continuous supply to key construction sites.

[0045] In this embodiment, deviation is the "trigger" for decision-making, while the material shortage risk index acts as both the "amplifier" and the "steering wheel." This application does not consider only one indicator but combines the plan deviation and the material shortage risk index to form a two-dimensional decision space. This achieves a balance between maintaining plan stability and ensuring production safety. The two work together to ensure the comprehensiveness and rationality of the decision. This collaborative mechanism enables scheduling decisions to be both forward-looking and able to grasp the main contradictions, ensuring the system's intelligence and robustness.

[0046] In one embodiment, local fine-tuning includes at least one of the following operations: shifting task time within a time buffer reserved in the pre-scheduling scheme, performing single-point task exchange, and switching production lines within the concrete mixing plant.

[0047] Local fine-tuning can be understood as making minor adjustments to affected local tasks without changing the overall pre-scheduling framework. Shifting task time within the time buffer reserved in the pre-scheduling plan (time shifting) can absorb delays and, in addition, can slightly adjust task start / end times without affecting critical constraints. Single-point task swapping involves finding a vehicle in the subsequent task chain of the affected vehicle that can be swapped with an idle vehicle; that is, only one task is swapped without changing the order of other tasks. Switching production lines within a concrete mixing plant is resource fine-tuning. For example, within the mixing plant, based on the production line load, a task about to be produced is switched from a busy production line to an idle or less loaded one.

[0048] In one embodiment, the local reordering step of rolling window optimization includes: determining a rolling optimization window based on a preset window duration, starting from the current time; reordering and optimizing the tasks within the rolling optimization window with the objective of minimizing the total deviation cost within the rolling optimization window; wherein, the total deviation cost includes task delay cost, vehicle usage cost, material shortage risk penalty, and penalty for the vehicle state at the tail of the rolling optimization window being inconsistent with the original plan; the reordering optimization adopts a fast genetic algorithm with an elite retention strategy.

[0049] It is understandable that when the impact of a disturbance exceeds the local fine-tuning capability, the system defines a rolling optimization window [t, t+H] (where H is the window duration, such as 2-4 hours) starting from the current time t, and performs rapid reordering optimization on the tasks within this window. During in-window reordering, the algorithm pays special attention to the vehicle's "task chain," meaning that after a vehicle completes its current task, its state (location, available time, load, etc.) directly affects its subsequent tasks. Therefore, rolling window optimization not only optimizes the tasks within the window but also ensures that the reordering results within the window smoothly connect with the subsequent pre-scheduled plans outside the window. That is, the vehicle state at the end of the window should be as consistent as possible with the "tail state" of the original pre-scheduled plan to avoid causing new chain disturbances to subsequent trips. This is achieved by adding a penalty term for tail state deviation to the objective function. Objective function: Within window H, the rearrangement is performed with the goal of minimizing the "total deviation cost". The total deviation cost includes: task delay cost: the delay time cost of all tasks; vehicle usage cost: the additional vehicle usage cost or waiting cost caused by the rearrangement; material shortage risk penalty: the penalty for the predicted material shortage risk; and penalty for the vehicle status at the tail of the rolling optimization window being inconsistent with the original plan.

[0050] Furthermore, the rearrangement optimization can employ a fast genetic algorithm with an elite retention strategy. This algorithm is optimized for real-time requirements in the following ways: Initial population: Based on the original plan within the current window, an initial population is generated through small-scale random mutations (such as swapping two tasks or assigning a task to another vehicle), ensuring the quality of the initial solution; Fast convergence: A smaller population size and number of iterations are used, and an "elite retention" strategy is introduced to ensure that the optimal solution in each generation is not lost, thus accelerating convergence; Constraint handling: Key constraints (such as continuous pumping and vehicle task chains) are treated as hard constraints, and any solution that violates the constraints is repaired or eliminated during generation.

[0051] In one embodiment, regarding the hierarchical response strategy for global rescheduling, when the disturbance has a wide impact and a long duration, causing the pre-scheduling scheme to lose most of its reference value, the processor will trigger a global rescheduling. At this time, the processor can use all unfinished tasks and available resources after the current moment as new input, combined with the real-time resource status, to rerun a simplified or fast-converging version of the pre-scheduling algorithm. Furthermore, to ensure response speed, the algorithm can relax some secondary optimization objectives (such as allowing a slight increase in total cost) and prioritize satisfying the core constraints.

[0052] In one embodiment, determining the final scheduling scheme for a concrete mixing plant based on a pre-scheduling scheme and a target-level response strategy includes: creating a shadow copy of the pre-scheduling scheme; running the target-level response strategy on the shadow copy to generate a shadow plan; performing a difference quantification analysis on the shadow plan and the shadow copy, providing conflict warnings for the shadow plan, and allowing the dispatcher to adjust or confirm the shadow plan; and, in response to the dispatcher's confirmation instruction regarding the shadow plan, determining the shadow plan confirmed by the dispatcher as the final scheduling scheme.

[0053] Understandably, to avoid impacting the currently executing schedule during optimization calculations, the processor will first create a shadow copy of the current schedule when triggering a level 2 or 3 response. All reordering optimizations are performed on this shadow schedule. After the calculation is complete, the processor will perform a "difference analysis between the old and new schedules," highlighting all changes, including: Quantitative analysis of differences: Clearly demonstrates the changes in key indicators such as total cost, total delay time, vehicle empty mileage, and material shortage risk index between the optimized plan and the original plan. For example, the new plan may reduce the material shortage risk by 0.5, but may increase the total transportation cost by 5%.

[0054] Conflict warning alert: Automatically identifies and highlights potential conflicts or risks that may exist in the shadow plan, such as a vehicle having too many tasks in a short period of time, or a construction site still having a high BRI after adjustments.

[0055] Dispatcher Intervention Interface: Provides an intuitive graphical interface that allows dispatchers to manually intervene and adjust shadow plans before confirmation. Dispatchers can manually adjust vehicle tasks, modify routes, and even temporarily add or reduce resources. The processor evaluates the impact of manual intervention in real time and provides "What-if" analysis to help dispatchers understand the consequences of their decisions. Each manual adjustment by the dispatcher is treated by the processor as a new constraint or objective preference and can trigger local re-optimization of the algorithm.

[0056] Once the scheduler confirms with a single click, the shadow plan will seamlessly switch to the new official execution plan and be immediately deployed. This mechanism ensures the stability and continuity of scheduling while fully leveraging both human experience and the computational advantages of machines.

[0057] In one embodiment, after executing the final scheduling scheme, the method further includes: performing feedback learning based on the execution results to optimize the parameters of subsequent pre-scheduling schemes or hierarchical response strategies. The feedback learning includes: recording disturbance events, target hierarchical response strategies, scheduler intervention behaviors, and execution results, and using the recorded data to dynamically optimize the time buffer parameters, material shortage risk index threshold, or cost function weights of the rearrangement algorithm in the pre-scheduling scheme through reinforcement learning algorithms or online learning algorithms.

[0058] It is understood that the embodiments of this application not only realize real-time scheduling, but also construct a closed-loop feedback learning machine of "disturbance-decision-result", enabling the entire scheduling system to have the ability to continuously adaptively optimize.

[0059] Data recording and knowledge accumulation: Real-time recording of detailed information for each disturbance event (type, time of occurrence, scope of impact); recording of automatically generated response plans (local fine-tuning, rolling window optimization, global reordering); recording of dispatcher's manual intervention (which tasks were modified, which parameters were adjusted); recording of the final execution results (actual delay time, actual material shortage, actual cost); this data is stored in a structured manner, forming a valuable "scheduling experience knowledge base".

[0060] Learning objectives and algorithms: Adaptive parameter adjustment: Utilizing historical data, key parameters are continuously optimized through reinforcement learning or online learning algorithms. For example: Pre-scheduling buffer optimization: Based on the frequency and impact of historical disturbances, the size of the pre-scheduling buffer for different construction sites and time periods is dynamically adjusted to improve the robustness of the pre-scheduling plan.

[0061] BRI Threshold Dynamic Adjustment: Based on the occurrence of historical material shortage events, the BRI thresholds for different construction sites and different concrete types are fine-tuned to more accurately reflect actual risks.

[0062] Cost function weight optimization: By learning the dispatcher's preferences for objective functions such as "time cost", "vehicle cost" and "material shortage penalty" in different situations, the weight of each cost item in the rearrangement algorithm is dynamically adjusted so that the algorithm-generated solution is more in line with actual operational needs.

[0063] Strategy library optimization: By analyzing the success rate of automated countermeasures and the effectiveness of human intervention in historical data, we can identify and suggest optimizations to the rules in the strategy library, such as adding more refined automated countermeasures for a specific type of congestion event.

[0064] Feedback loop: The output of the learning algorithm (optimized parameters, suggested policy rules) is periodically or in real-time fed back to the pre-scheduling system and the real-time scheduling system, thereby achieving continuous improvement and adaptive optimization of the entire scheduling process. This enables the system to learn from historical experience and continuously improve its prediction accuracy, response efficiency, and optimization capabilities.

[0065] Existing scheduling schemes suffer from the following drawbacks: Lack of effective integration with pre-scheduling: Most existing real-time scheduling systems operate independently of pre-scheduling, failing to fully utilize the optimization benchmarks provided by pre-scheduling. This leads to deviations from the overall optimal goal during real-time adjustments, and may even create new conflicts. Low efficiency and error-prone real-time decision-making: When faced with complex and unpredictable emergencies, manual scheduling struggles to process massive amounts of information and make optimal decisions quickly, easily resulting in scheduling chaos, resource waste, or construction delays. Inability to effectively handle complex disturbances: For chain-reaction emergencies involving multiple vehicles, construction sites, and production lines, existing technologies struggle to conduct systematic impact assessments and global optimization and rearrangement. Insufficient protection of key constraints: In emergency situations, manual scheduling or simple systems cannot ensure that core process constraints such as continuous concrete pumping are not broken, posing a risk of quality accidents. Lack of learning and feedback mechanisms: Experience and data generated during real-time scheduling are not effectively fed back to the pre-scheduling system, preventing the robustness of the pre-scheduling plan from continuously improving.

[0066] To address the above problems, in one specific embodiment, such as Figure 2 As shown, a scheduling method for concrete batching plants is provided. This scheduling method takes the pre-scheduled plan of the concrete batching plant as a baseline, and intelligently identifies sudden disturbances by monitoring the production, transportation and construction site status in real time. According to the type and degree of disturbance, a hierarchical response strategy is adopted to perform local fine-tuning, rolling window optimization or global rearrangement to minimize the impact of disturbances on the overall plan, ensure the satisfaction of key constraints, and continuously optimize scheduling efficiency.

[0067] The core algorithm logic of this application is divided into four levels: perturbation quantization, hierarchical response decision, reordering optimization, and feedback learning. A "shadow plan" mechanism is introduced to ensure a smooth transition of the scheduling scheme. The specific steps are as follows: Step 1: Real-time data acquisition and perturbation quantization model The system continuously collects real-time data from multiple sources and compares it with the pre-scheduled plan (baseline), calculating the "plan deviation" using a disturbance quantification model. Time Deviation (TD): For any task node (such as a vehicle arriving at the construction site), calculate the difference between its actual / estimated time (T_actual) and its planned time (T_planned).

[0068] TD = T_actual - T_planned Resource Deviation (RD): Monitors changes in the availability of critical resources (vehicles, production lines).

[0069] RD_vehicle = 1 (if the vehicle is faulty), 0 (if the vehicle is normal) RD_line = 1 (if production line malfunctions), 0 (normal) Demand Deviation (DD): Monitors changes in site demand, i.e., the difference between actual / expected demand (V_actual_demand) and planned demand (V_planned_demand).

[0070] DD = V_actual_demand - V_planned_demand The system triggers disturbance identification based on a preset deviation threshold. For example, a disturbance event is identified if TD > 15 minutes or RD_vehicle = 1.

[0071] Step Two: Disturbance Impact Assessment and Tiered Response Decision The system performs real-time impact assessments on identified disturbances, predicting their cascading effects on subsequent plans, particularly the potential threat to the "continuous pumping constraint at the construction site." The assessment model calculates a "Breakage Risk Index (BRI)."

[0072] Risk Index for Material Shortage (BRI) Calculation Formula: BRI=max(0,frac{T_{next_arrival}-(T_{current_unload_end}+T_{max_gap})}{T_{max_gap}}) The textual explanation is as follows: Material shortage risk index = max{0, [Estimated arrival time at the construction site - (Estimated unloading end time + Maximum material shortage interval)] / Maximum material shortage interval} in: T_{next_arrival}: The estimated arrival time of the next concrete mixer truck at the construction site.

[0073] T_{current_unload_end}: The estimated time when the unloading of the current mixer truck will be completed.

[0074] T_{max_gap}: The maximum time interval (e.g., 15 minutes) during which continuous pumping is allowed on the construction site.

[0075] This formula quantifies the extent to which the arrival time of the next vehicle exceeds the maximum permissible interval. If $BRI>0$, it indicates a risk of material shortage, and the higher the value, the greater the risk.

[0076] BRI calculation example: Suppose that a construction site C01 is continuously pumping, and the maximum allowable material interruption interval T_{max_gap} is 15 minutes.

[0077] Scenario 1: Normal situation Vehicle V01 is expected to finish unloading at 10:30 (T_{current_unload_end}).

[0078] The next train, V02, is expected to arrive at 10:40 (T_{next_arrival}).

[0079] BRI = \max(0, \frac{10:40 - (10:30 + 15 minutes)}{15 minutes}) = \max(0, \frac{10:40 - 10:45}{15 minutes}) = \max(0, -0.33) = 0.

[0080] Conclusion: BRI is 0, indicating no risk of material shortage.

[0081] Scenario 2: Minor delay Due to slight congestion, V02's estimated arrival time has changed to 10:50.

[0082] BRI = \max(0, \frac{10:50 - (10:30 + 15 minutes)}{15 minutes}) = \max(0, \frac{10:50 - 10:45}{15 minutes}) = \max(0, 0.33) = 0.33.

[0083] Conclusion: The BRI is 0.33, indicating a slight risk of material shortage. The system needs to pay attention and may need to initiate a Level 1 response.

[0084] Scenario 3: Severe Delay Due to severe congestion or malfunction, the estimated arrival time of V02 has been changed to 11:00.

[0085] BRI = \max(0, \frac{11:00 - (10:30 + 15 minutes)}{15 minutes}) = \max(0, \frac{11:00 - 10:45}{15 minutes}) = \max(0, 1) = 1.

[0086] Conclusion: With a BRI of 1, the risk of material shortage is extremely high, meaning the material shortage time is already equal to a maximum permissible interval. The system must initiate a level 2 or 3 response.

[0087] The collaborative decision-making mechanism of deviation and BRI: Plan deviation (TD, RD, DD) and the risk index of material shortage (BRI) are two core dimensions driving graded response decisions. They work together to ensure the comprehensiveness and rationality of the decision.

[0088] Deviation is the quantification of "disturbance," while BRI is the quantification of "consequences": Deviation describes the degree of difference between the current state and the original plan. For example, TD = 20 minutes is simply a factual statement indicating that a vehicle is 20 minutes late. BRI, however, goes beyond this and further predicts the most severe consequences that the disturbance might cause—that is, whether it will lead to a continuous pumping interruption at a critical site. Even with the same 20-minute delay, if the site has sparse subsequent tasks and sufficient buffer, the BRI may still be low; but if subsequent tasks are intensive, the BRI may be high.

[0089] Two-dimensional decision matrix: The system does not simply look at a single indicator, but combines the two to form a two-dimensional decision space in order to achieve a balance between the two objectives of "maintaining the stability of the plan" and "ensuring production safety".

[0090] Table 1 Two-dimensional decision matrix table

[0091] In summary, deviation is the "trigger" for decision-making, while BRI (Balance of Integrity) acts as both the "amplifier" and the "steering wheel." A high deviation accompanied by a low BRI will lead the system to make the most cost-effective adjustment; conversely, even a low deviation, if accompanied by a high BRI, will prompt the system to take the highest-priority intervention. This synergistic mechanism ensures that scheduling decisions are both forward-looking and address the core issues, guaranteeing the system's intelligence and robustness.

[0092] Step 3: Core Rearrangement Algorithm, "Shadow Project" Mechanism, and Human-Machine Collaborative Decision Making Local fine-tuning: Logic: Without altering the overall pre-scheduling framework, make minor adjustments to the affected local tasks. This includes: Time shifting: Utilize the time buffer reserved in the pre-schedule to absorb delays, or slightly adjust the start / end time of tasks without affecting critical constraints.

[0093] Mission Exchange: In the subsequent mission chain of the affected vehicle, find a vehicle that can be exchanged with an available one. "Single-point task swap" means swapping only one task without changing the order of other tasks.

[0094] Resource fine-tuning: For example, within a mixing plant, depending on the production line load, the task to be produced can be switched from a busy production line to another idle or less busy one.

[0095] Scrolling window optimization: Logic: This is one of the core algorithms of this invention. When the impact of a disturbance exceeds the local fine-tuning capability, the system defines a rolling optimization window [t, t+H] (where H is the window duration, such as 2-4 hours) starting from the current time t. Tasks within this window are then rapidly rearranged and optimized.

[0096] Vehicle Task Chain and Tail State Consistency: During in-window reordering, the algorithm pays special attention to the vehicle's "task chain." That is, once a vehicle completes its current task, its state (location, available time, load, etc.) directly affects its subsequent tasks. Therefore, rolling window optimization not only optimizes tasks within the window but also ensures a smooth transition between the in-window reordering result and the subsequent pre-ordering plan outside the window. In other words, the vehicle state at the end of the window should be as consistent as possible with the "tail state" of the original pre-ordering plan to avoid new cascading disturbances to subsequent trips. This is achieved by adding a penalty term for tail state deviation to the objective function.

[0097] Objective function: Within window H, the rearrangement aims to minimize the "total deviation cost". Total deviation cost includes: Task delay cost (Cost_delay): The delay time cost of all tasks.

[0098] Cost of vehicle usage: Additional vehicle usage costs or waiting costs incurred due to reordering.

[0099] Penalty_breakage: A penalty for the predicted risk of material breakage (a very large value).

[0100] Penalty_tail_discontinuity for vehicles whose status at the rear of the window does not match the original plan.

[0101] Algorithm: A "Fast Genetic Algorithm with Elitism" is employed. This algorithm is optimized for real-time requirements. Initial population: Based on the original plan within the current window, an initial population is generated through small-scale random mutations (such as swapping two tasks or assigning a task to another vehicle), ensuring the quality of the initial solution.

[0102] Fast convergence: A smaller population size and number of iterations are used, and an "elite retention" strategy is introduced to ensure that the optimal solution in each generation is not lost, thus accelerating convergence.

[0103] Constraint handling: Critical constraints (such as continuous pumping and vehicle task chains) are treated as hard constraints, and any solutions that violate the constraints are repaired or eliminated during generation.

[0104] Global Reordering: Logic: When a disturbance has a wide impact and a long duration, causing the pre-scheduling baseline to lose most of its reference value, the system will trigger a global rescheduling. At this time, the system will take all unfinished tasks and available resources after the current moment as new inputs, combine them with the real-time resource status, and rerun a simplified or fast-converging version of the pre-scheduling algorithm.

[0105] Simplification: To ensure response speed, the algorithm may relax some secondary optimization objectives (such as allowing a slight increase in total cost) and prioritize ensuring the satisfaction of core constraints.

[0106] The "Shadow Plan" mechanism and human-machine collaborative decision-making: To avoid affecting the currently executing schedule during optimization calculations, the system first creates a "shadow copy" of the current schedule when a level 2 or 3 response is triggered.

[0107] All rearrangement optimizations are performed on this shadow plan. After the calculation is complete, the system will perform a "difference analysis between the old and new plans," highlighting all changes, including: Quantitative analysis of differences: Clearly demonstrates the changes in key indicators such as total cost, total delay time, vehicle empty mileage, and material shortage risk index between the optimized plan and the original plan. For example, the new plan may reduce the material shortage risk by 0.5, but may increase the total transportation cost by 5%.

[0108] Conflict warning alert: Automatically identifies and highlights potential conflicts or risks that may exist in the shadow plan, such as a vehicle having too many tasks in a short period of time, or a construction site still having a high BRI after adjustments.

[0109] Dispatcher Intervention Interface: The system provides an intuitive graphical interface that allows dispatchers to manually intervene and adjust shadow plans before confirmation. Dispatchers can manually adjust vehicle tasks, modify routes, and even temporarily add or reduce resources. The system will evaluate the impact of manual intervention in real time and provide "What-if" analysis to help dispatchers understand the consequences of their decisions. Each manual adjustment by the dispatcher will be regarded by the system as a new constraint or objective preference, which may trigger local re-optimization of the algorithm.

[0110] Once the scheduler confirms with a single click, the shadow plan will seamlessly switch to the new official execution plan and be immediately deployed. This mechanism ensures the stability and continuity of scheduling while fully leveraging both human experience and the computational advantages of machines.

[0111] Step 4: Feedback Learning and Adaptive Optimization Mechanism This application not only achieves real-time scheduling, but also constructs a closed-loop feedback learning machine of "perturbation-decision-result", enabling the entire scheduling system to have the ability to continuously adapt and optimize.

[0112] Data recording and knowledge accumulation: The system records detailed information (type, time of occurrence, and scope of impact) for each disturbance event in real time.

[0113] Record the response schemes automatically generated by the system (local fine-tuning, scrolling window optimization, global reordering).

[0114] Record the scheduler's manual intervention actions (which tasks were modified and which parameters were adjusted).

[0115] Record the final execution results (actual delay time, actual material shortage, and actual cost).

[0116] This data is stored in a structured manner, forming a valuable "scheduling experience knowledge base".

[0117] Learning objectives and algorithms: Parameter adaptive adjustment: The system utilizes historical data and employs reinforcement learning or online learning algorithms to continuously optimize key parameters. For example: Pre-scheduling buffer optimization: Based on the frequency and impact of historical disturbances, the size of the pre-scheduling buffer for different construction sites and time periods is dynamically adjusted to improve the robustness of the pre-scheduling plan.

[0118] BRI Threshold Dynamic Adjustment: Based on the occurrence of historical material shortage events, the BRI thresholds for different construction sites and different concrete types are fine-tuned to more accurately reflect actual risks.

[0119] Cost function weight optimization: By learning the dispatcher's preferences for objective functions such as "time cost", "vehicle cost" and "material shortage penalty" in different situations, the weight of each cost item in the rearrangement algorithm is dynamically adjusted so that the algorithm-generated solution is more in line with actual operational needs.

[0120] Strategy library optimization: By analyzing the success rate of automated countermeasures and the effectiveness of human intervention in historical data, the system can identify and suggest optimizations to the rules in the strategy library, such as adding more refined automated countermeasures for a specific type of congestion event.

[0121] Feedback loop: The output of the learning algorithm (optimized parameters, suggested policy rules) is periodically or in real-time fed back to the pre-scheduling system and the real-time scheduling system, thereby achieving continuous improvement and adaptive optimization of the entire scheduling process. This enables the system to learn from historical experience and continuously improve its prediction accuracy, response efficiency, and optimization capabilities.

[0122] Emergency Response Strategy Library: This application establishes a structured "disturbance-impact-countermeasure" matrix strategy library, providing a standardized handling process for every foreseeable emergency.

[0123] Table 2 Emergency Response Strategy Library

[0124] Relationship between core algorithms and policy libraries The core rearrangement algorithm in this application (such as the fast genetic algorithm in rolling window optimization) and the contingency response strategy library are complementary rather than conflicting, together forming a complete real-time scheduling decision-making system. The strategy library can be figuratively compared to the "decision logic of the scheduling brain," while the core algorithm is the "computational engine of the scheduling brain."

[0125] The Policy Layer defines "what to do." It contains pre-defined response rules, triggering conditions, impact assessment criteria (such as the BRI threshold) for various contingencies, and optimization objectives and constraints to be prioritized at different response levels. Built upon domain knowledge and historical experience, the Policy Layer guides the system on which level of response (local fine-tuning, rolling window optimization, or global reordering) should be adopted when facing specific disturbances, and provides the algorithm with specific parameters and target weights.

[0126] The core algorithm (Execution Layer) defines "how to do it." It receives instructions and parameters from the policy library and, using optimization models and computational methods, quickly generates optimal or near-optimal scheduling schemes while satisfying all constraints (including hard constraints defined by the policy library). For example, when the policy library indicates the need for "rolling window optimization," the core algorithm initiates a fast genetic algorithm to calculate a new scheduling scheme in a short time based on the objective function set by the policy library (including cost, risk penalties, tail state consistency penalties, etc.) and constraints.

[0127] The strategy library acts as an "intelligent guide" for algorithms, selecting appropriate algorithm patterns and providing optimization directions and boundaries based on real-time conditions and preset rules. The algorithm, on the other hand, is the "efficient executor" of the strategy library, quickly and accurately calculating specific scheduling schemes under its guidance. Working together, they achieve closed-loop management from intelligent decision-making to efficient execution.

[0128] In summary, the technical solutions provided in this application have the following key technical points: 1) Real-time state deviation identification and classification technology based on "baseline-disturbance": Innovatively, the pre-scheduled plan is used as an "ideal baseline". By continuously comparing multi-source real-time data with the baseline, various sudden disturbances (such as time deviation, resource usage deviation, and demand deviation) are accurately identified and quantified, and intelligently classified to provide a basis for subsequent graded response.

[0129] 2) Hierarchical Response Real-Time Scheduling Strategy: To address perturbations of different types and degrees of impact, a three-tiered intelligent response mechanism is proposed, consisting of local fine-tuning, rolling window optimization, and global rearrangement. This hierarchical strategy ensures that the system maintains response speed while also considering optimization depth and computational efficiency, avoiding overkill or insufficient improvement.

[0130] 3) Human-machine collaborative decision support and feedback learning loop: The system not only automatically generates optimization plans but also provides an intuitive human-computer interaction interface, supporting dispatchers' manual intervention and decision-making. Simultaneously, the results of manual intervention and experience data from real-time scheduling are fed back to the pre-scheduling algorithm through machine learning and other methods, enabling continuous adaptive optimization and evolution of the entire scheduling system.

[0131] Therefore, the technical solution of this application has the following advantages: 1) From “passive firefighting” to “proactive intelligent intervention”: This invention liberates dispatchers from the heavy and error-prone manual “firefighting” through real-time monitoring and intelligent algorithms, transforming passive response into proactive prediction and optimization.

[0132] 2) Seamless integration of pre-scheduling and real-time scheduling: This invention fully utilizes the optimization results of pre-scheduling as a benchmark to ensure that real-time adjustments are made within the overall optimal framework, avoiding the problem of "treating the symptoms rather than the root cause".

[0133] 3) Tiered response, high efficiency and flexibility: Different strategies are adopted according to the degree of disturbance, which not only ensures the rapid processing of minor disturbances, but also provides a systematic rearrangement capability for complex disturbances, thereby improving scheduling efficiency and system robustness.

[0134] 4) Intelligent protection of key constraints: Especially for the core process requirement of continuous concrete pumping, this invention can predict risks in real time and take multiple hedging measures to minimize quality accidents.

[0135] 5) Continuous learning and optimization: The data and experience accumulated during real-time scheduling can feed back into the pre-scheduling algorithm, forming a closed-loop optimization, enabling the entire scheduling system to continuously evolve.

[0136] 6) Improve economic efficiency and customer satisfaction: By efficiently responding to emergencies, the time spent by vehicles running empty and waiting time was reduced, thus lowering operating costs; at the same time, the timely supply of concrete was ensured, improving the operating efficiency and risk resistance of the mixing plant, and enhancing the satisfaction and trust of the construction site.

[0137] In one embodiment, this application also provides a scheduling device for a concrete mixing plant, comprising: a memory configured to store instructions; and a processor configured to retrieve instructions from the memory and, when executing the instructions, to implement the scheduling method for a concrete mixing plant according to the above embodiments.

[0138] In one embodiment, this application also provides a concrete mixing plant, including: a scheduling device for a concrete mixing plant according to the above embodiments.

[0139] In one embodiment, this application also provides a machine-readable storage medium storing instructions for causing a machine to execute the scheduling method for a concrete mixing plant according to the above embodiments.

[0140] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0141] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0142] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0143] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0144] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0145] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0146] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0147] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0148] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A scheduling method for a concrete mixing plant, characterized in that, The scheduling method includes: After determining the pre-scheduling plan for the concrete batching plant, multi-source status data is acquired in real time, including concrete batching plant production status data, concrete mixer truck transportation status data, and construction site demand status data. Based on the multi-source state data and the pre-scheduling scheme, the plan deviation is determined, and the plan deviation is used to identify and determine whether a disturbance event has occurred. In the event of a disturbance, a material shortage risk assessment is conducted on the disturbance to determine the material shortage risk index corresponding to the construction site affected by the disturbance. Based on the planned deviation and the material shortage risk index, determine the target graded response strategy corresponding to the disturbance event; Based on the pre-scheduling scheme and the target hierarchical response strategy, the final scheduling scheme for the concrete mixing plant is determined and executed.

2. The scheduling method according to claim 1, characterized in that, The deviation from the plan includes at least one of the following: Time deviation is determined based on the difference between the actual or expected time of the task node and the planned time in the pre-scheduling scheme. Resource deviation is determined based on changes in the availability of key resources at the concrete mixing plant; Demand deviation is determined based on the difference between the actual demand at the construction site and the planned demand in the pre-scheduling scheme.

3. The scheduling method according to claim 1, characterized in that, The material shortage risk index is determined based on the estimated arrival time of the next concrete mixer truck at the construction site, the estimated unloading end time of the concrete mixer truck currently unloading at the construction site, and the maximum allowable material shortage interval at the construction site.

4. The scheduling method according to claim 1, characterized in that, Based on the planned deviation and the material shortage risk index, determine the target graded response strategy corresponding to the disturbance event, including: If the planned deviation is within the low planned deviation range and the material shortage risk index is within the low material shortage risk index range, the target graded response strategy corresponding to the disturbance event is determined to be non-intervention and only monitoring. If the planned deviation is within the low planned deviation range and the material shortage risk index is within the high material shortage risk index range, the target graded response strategy corresponding to the disturbance event is determined to be local fine-tuning. If the planned deviation is within the high planned deviation range and the material shortage risk index is within the low material shortage risk index range, the target hierarchical response strategy corresponding to the disturbance event is determined to be local rearrangement optimized by rolling window. If the planned deviation is within the high planned deviation range and the material shortage risk index is within the high material shortage risk index range, the target hierarchical response strategy corresponding to the disturbance event is determined to be global rearrangement.

5. The scheduling method according to claim 4, characterized in that, The local fine-tuning includes at least one of the following operations: shifting task time within the time buffer reserved in the pre-scheduling scheme, performing single-point task exchange, and switching production lines within the concrete mixing plant.

6. The scheduling method according to claim 4, characterized in that, The steps of the local rearrangement optimization of the scroll window include: Starting from the current moment, determine the scrolling optimization window based on the preset window duration; The task rearrangement optimization is performed within the rolling optimization window with the objective of minimizing the total deviation cost. The total deviation cost includes task delay cost, vehicle usage cost, material shortage risk penalty, and penalty for the vehicle status at the tail of the rolling optimization window being inconsistent with the original plan. The rearrangement optimization adopts a fast genetic algorithm with an elite retention strategy.

7. The scheduling method according to claim 1, characterized in that, Based on the pre-scheduling scheme and the target hierarchical response strategy, the final scheduling scheme for the concrete mixing plant is determined, including: Create a shadow copy of the pre-scheduled scheme; The target hierarchical response strategy is run on the shadow copy to generate a shadow plan; The shadow plan and the shadow copy are subjected to difference quantitative analysis, and conflict warnings are issued for the shadow plan, allowing the scheduler to adjust or confirm the shadow plan; In response to the dispatcher's confirmation instruction regarding the shadow plan, the shadow plan confirmed by the dispatcher is determined as the final scheduling scheme.

8. The scheduling method according to claim 1, characterized in that, After executing the final scheduling scheme, the process also includes: Feedback learning is performed based on the execution results to optimize the parameters of subsequent pre-scheduling schemes or hierarchical response strategies. The feedback learning includes recording disturbance events, target hierarchical response strategies, scheduler intervention behaviors, and execution results. The recorded data is then used to dynamically optimize the time buffer parameters, material shortage risk index threshold, or cost function weights of the rearrangement algorithm in the pre-scheduling scheme through reinforcement learning algorithms or online learning algorithms.

9. A dispatching device for a concrete mixing plant, characterized in that, include: The memory is configured to store instructions; as well as The processor is configured to retrieve the instructions from the memory and, when executing the instructions, to implement the scheduling method for a concrete mixing plant according to any one of claims 1 to 8.

10. A machine-readable storage medium, characterized in that, The machine-readable storage medium stores instructions for causing the machine to perform a scheduling method for a concrete mixing plant according to any one of claims 1 to 8.

Images