Material scheduling method, device, equipment, medium and product

Abstract

This invention discloses a material scheduling method, apparatus, equipment, medium, and product. The method includes: acquiring delivery equipment data, delivery task information destined for a target machine, and material operation data of the target machine at the current time; determining the latest call time for the target machine to initiate a material request based on the current time, delivery equipment data, delivery task information, and material operation data; when the current time is earlier than the latest call time, determining the delivery load attribute based on the task flow data destined for the target machine within a preset time window; and determining the target material request method based on the delivery load attribute, the latest call time, the current time, and the material operation data, and initiating a material request based on the target material request method. The technical solution provided by this invention improves the timeliness and reliability of material delivery, reduces the risk of machine downtime due to material shortages and material blockage at the production line, thereby ensuring the safety and smooth operation of the production site.

Description

Technical Field

[0001] This invention relates to the field of computer processing technology, and in particular to a material scheduling method, apparatus, equipment, medium, and product. Background Technology

[0002] Currently, in the cigarette material distribution scenario of the cigarette making and packaging workshop, the common operating mode is for the machine to initiate a material request to the material distribution system, and then the warehouse responds by preparing the goods, shipping them out of the warehouse, and transporting them to the machine. To ensure production continuity, higher requirements are placed on the timeliness and stability of material scheduling.

[0003] However, existing scheduling methods mainly rely on machine-side material requests at fixed intervals or on manual material requests based on experience. This method can easily lead to material delivery delays or premature material arrivals, causing machine shutdowns due to material shortages or material blockages around the machines, thus affecting the safety and smooth operation of the production site. Summary of the Invention

[0004] This invention provides a material scheduling method, apparatus, equipment, medium, and product to improve the timeliness and reliability of material delivery, reduce the risk of machine downtime due to material shortage and material blockage at the production line, thereby ensuring the safety and smooth operation of the production site.

[0005] According to one aspect of the present invention, a material scheduling method is provided, the method comprising: Obtain the current delivery equipment data, delivery task information to the target machine, and material operation data of the target machine from the dispatch system; Based on the current time, the delivery equipment data, the delivery task information, and the material operation data, determine the latest call time for the target machine to initiate a material request; When the current time is earlier than the latest call time, the delivery load attribute is determined based on the task flow data sent to the target machine within a preset duration window; Based on the delivery load attributes, the latest call time, the current time, and the material operation data, a target material request method is determined, and the material request requests of the target machine are scheduled based on the target material request method.

[0006] According to another aspect of the present invention, a material dispatching device is provided, the device comprising: The data acquisition module is used to acquire the delivery equipment data of the scheduling system at the current moment, the delivery task information sent to the target machine, and the material operation data of the target machine; The latest call time determination module is used to determine the latest call time for the target machine to initiate a material request based on the current time, the delivery equipment data, the delivery task information, and the material operation data; The delivery load attribute determination module is used to determine the delivery load attribute based on the task flow data sent to the target machine within a preset duration window when the current time is earlier than the latest call time. The target material request method determination module is used to determine the target material request method based on the delivery load attribute, the latest call time, the current time and the material operation data, so as to schedule the material request request of the target machine based on the target material request method.

[0007] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: At least one processor; and a memory communicatively connected to said at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the material scheduling method according to any embodiment of the present invention.

[0008] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the material scheduling method according to any embodiment of the present invention.

[0009] According to another aspect of the present invention, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the material scheduling method as described in any embodiment of the present invention.

[0010] The technical solution of this invention obtains the delivery equipment data of the scheduling system, the delivery task information sent to the target machine, and the material operation data of the target machine at the current time; based on the current time, delivery equipment data, delivery task information, and material operation data, it determines the latest call time for the target machine to initiate a material request; when the current time is earlier than the latest call time, it determines the delivery load attribute based on the task flow data sent to the target machine within a preset time window; based on the delivery load attribute, the latest call time, the current time, and the material operation data, it determines the target material request method, and initiates a material request based on the target material request method. This solves the problem in the prior art where material requests are made according to fixed cycles or manual experience, which easily leads to material delivery delays or premature material arrival, thereby affecting the safety and smoothness of the production site. It realizes that by combining the current time, the delivery equipment data of the scheduling system, the delivery task information sent to the target machine, and the material operation data of the target machine, it determines the latest call time for the target machine to initiate a material request, thereby avoiding the backlog of line-side inventory caused by calling too early and avoiding the production risks caused by calling too late. Furthermore, by analyzing the task flow data sent to the target machine when the current time is earlier than the latest call time, the delivery load attribute is determined. This allows for the ability to anticipate the delivery logistics capacity before issuing new material requests. Based on the delivery load attribute, the latest call time, the current time, and material operation data, the target material request method is determined. This ensures that the target material request method is precisely matched with current production needs and delivery supply capacity. Material requests are then initiated based on the target material request method, improving the on-time rate and reliability of material delivery, reducing the risk of machine downtime due to material shortages and material blockages at the production line, thereby ensuring the safety and smooth operation of the production site.

[0011] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a flowchart of a material scheduling method provided according to an embodiment of the present invention; Figure 2 This is a flowchart of a material scheduling method provided according to an embodiment of the present invention; Figure 3This is a flowchart of a material scheduling method provided according to an embodiment of the present invention; Figure 4 This is a schematic diagram of a method for characterizing the material scheduling system to perform material scheduling according to an embodiment of the present invention; Figure 5 This is a schematic diagram of the state transition for representing a delivery task, provided according to an embodiment of the present invention; Figure 6 This is a flowchart of a material scheduling method provided according to an embodiment of the present invention; Figure 7 This is a schematic diagram of a method for characterizing and determining delivery load attributes provided in an embodiment of the present invention; Figure 8 This is a schematic diagram of a method for characterizing and determining a target material according to an embodiment of the present invention; Figure 9 This is a schematic diagram of the structure of a material scheduling device provided according to an embodiment of the present invention; Figure 10 This is a schematic diagram of the structure of an electronic device that implements the material scheduling method of this invention. Detailed Implementation

[0014] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. 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 should fall within the scope of protection of the present invention.

[0015] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0016] It should be noted that the collection, gathering, updating, analysis, processing, use, transmission, and storage of user personal information involved in the technical solution disclosed herein all comply with relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. Necessary measures are taken to prevent unauthorized access to user personal information data and to maintain user personal information security and network security. It should also be noted that the collection, gathering, updating, analysis, processing, use, transmission, and storage of user personal information involved in the technical solution disclosed herein are all conducted with the user's knowledge and consent, and comply with relevant privacy protection regulations.

[0017] Before introducing this technical solution, we can first describe the application scenario. The technical solution provided by this invention can be applied to any scenario that requires material scheduling. The material can be cigarette tobacco or any other scheduling material.

[0018] In existing technologies, the distribution of cigarette materials in the cigarette making and packaging workshop (such as pallet distribution of cigarette materials) is usually executed by multiple machines in parallel. To ensure that the machines are not out of stock, the cigarette making and packaging workshop generally adopts the process of "machine-side material request (MES (Manufacturing Execution System) submission) → warehousing / stocking → AGV (Automated Guided Vehicle) outbound transportation → machine arrival confirmation". During material scheduling, material requests are usually submitted at the machine side when the line-side inventory is judged to be lower than a certain experience threshold based on manual experience, or materials are replenished at fixed time intervals. This method is prone to causing machine downtime due to material shortages, resulting in production capacity loss and quality fluctuations.

[0019] To solve the above problems, the technical solution provided by this invention can determine when or how to request materials, ensuring that there is no backlog of line-side inventory and space occupation due to calling too early, and no risk of production line shutdown due to calling too late. This improves the continuity of material supply and production stability, while ensuring production safety and efficiency.

[0020] For example, when it is desirable to determine whether and / or when to request materials from the scheduling system, the target material request method can be determined based on the technical solution provided in this embodiment, so as to execute the material request call of the target machine based on the target material request method.

[0021] Figure 1This is a flowchart of a material scheduling method provided by an embodiment of the present invention. This embodiment is applicable to any scenario requiring materials. The method can be executed by a material scheduling device, which can be implemented in hardware and / or software and can be configured in a computing device. Figure 1 As shown, the method includes: S110. Obtain the current delivery equipment data of the scheduling system, the delivery task information sent to the target machine, and the material operation data of the target machine.

[0022] The scheduling system can be a control system used to coordinate the flow of materials and direct the operation of delivery equipment (such as AGVs, conveyor lines, etc.). The target machine refers to the terminal production equipment on the production line that actually consumes cigarette materials (such as cigarette making or packaging machines) and generates specific material requirements. The current moment refers to the point in time when the target machine triggers a material request decision. For example, when it is necessary to comprehensively evaluate the real-time operating status and the material supply and demand relationship to determine whether to initiate a material request, this moment can be used as the current moment.

[0023] Delivery equipment data is used to reflect the real-time operation data of material delivery in the current time and space environment, including but not limited to: the physical status of delivery equipment (such as real-time location, speed, load, power, working and fault status), the execution progress of tasks in transit (such as departed, in transit, waiting to be unloaded or completed), the traffic conditions of logistics channels (such as congestion level, path occupancy rate), and the backlog of task queues.

[0024] Delivery task information can be data related to material delivery tasks. Specifically, it can include static attributes of the delivery task (such as task ID, type, origin and destination, material type, quantity and priority), as well as dynamic progress within the delivery task lifecycle (such as the current stage (such as creation, preparation, outbound, transportation, arrival, completion, exception), estimated arrival time, elapsed time, remaining route), the status of the logistics resources it depends on (such as assigned vehicle number, load rate), and external environmental constraints (such as route congestion index, station waiting time).

[0025] Material operation data is used to reflect the urgency of material consumption on the machine side (i.e., the target machine), and may include, but is not limited to, the remaining inventory at the machine line, the real-time material consumption rate, the current production speed, and the safety buffer inventory threshold.

[0026] In this embodiment, real-time detection or periodic polling can be used to dynamically determine whether a material requisition detection event is triggered. The triggering conditions for the material requisition detection event include, but are not limited to, at least one of the following: a change in the status of the delivery task at the target machine's associated site; an update to the task queue for the target machine in the scheduling system (such as adding, canceling, or adjusting priority); receiving a task arrival confirmation signal from the target machine; the material inventory at the target machine's line-side dropping to a preset warning threshold; fluctuations in the real-time material consumption rate (such as acceleration or deceleration exceeding a set ratio); a change in the target machine's production status (such as resuming operation from shutdown, switching production grades, or adjusting production speed); and temporary adjustments to the upper-level production plan.

[0027] When any of the above-mentioned triggering events is detected, three-dimensional key data at the current moment can be collected simultaneously: first, delivery equipment data reflecting the real-time logistics operating environment (including equipment location, road congestion, etc.); second, delivery task information covering the entire task lifecycle (including on-the-go task progress, queue backlog, etc.); and third, material operation data characterizing the urgency of production consumption (including line-side inventory, consumption rate, etc.). Based on the fusion analysis of these three types of data, a comprehensive assessment is made as to whether to formally initiate a material request, and the optimal material scheduling strategy is generated accordingly (such as determining the timing of the call, the amount of material required, and the task priority) to achieve precise coordination between logistics response and production needs.

[0028] S120. Based on the current time, delivery equipment data, delivery task information, and material operation data, determine the latest call time for the target machine to initiate a material request.

[0029] Among them, the latest call time can be the theoretical deadline for initiating a material request to ensure that the target machine does not experience a material shortage and stop.

[0030] In one implementation, real-time data from delivery equipment (including vehicle location, estimated arrival time, and current load) can be collected. Simultaneously, delivery task information destined for the target machine (covering the type, quantity, and priority of materials to be delivered) can be obtained, and material operation data from the target machine can be read (extracting current line-side inventory, real-time production cycle time, and material consumption rate per unit time). Based on this data, the remaining production time is calculated by dividing the current line-side inventory by the real-time consumption rate. A preset fixed safety buffer time (to handle sudden shutdowns or delivery delays) and the average delivery time estimated based on current road conditions and distance are then deducted to derive the theoretical latest call time.

[0031] In another implementation, machine learning algorithms can be used to train historical delivery status, task execution records, and fluctuations in line-side material consumption to construct dynamic delivery time prediction models and material consumption prediction models. After acquiring real-time 3D data, it is input into the aforementioned models. The delivery time prediction model predicts the dynamic delivery time intervals affected by factors such as queues within future periods; the material consumption prediction model predicts the material consumption rate curve that changes with the production rhythm within future periods. Furthermore, combined with the minimum allowable material shortage tolerance of the target machine, a rolling time-domain optimization algorithm is used to calculate the latest call time that fluctuates with time by integrating the dynamic time intervals and consumption curves. This time is no longer a fixed value, but a variable that can be dynamically adjusted according to operating conditions, production rhythm, and task queue length, thereby more accurately adapting to complex cigarette production scenarios.

[0032] In another implementation, a digital twin can be constructed in a virtual digital environment, encompassing the cigarette production logistics network, AGVs, target machines, and line-side warehouses. Real-time collected delivery status, task data, and material operation data are mapped to the twin model as the initial state. Starting from the current moment, multiple future simulation scenarios are simulated in parallel within the virtual data environment, covering material flow processes under different operating conditions such as normal delivery, vehicle malfunctions, and urgent order insertions. By detecting the critical time point when the line-side inventory of the target machine drops to zero in each simulation scenario, the latest trigger point without material shortages under different confidence levels (e.g., 95% guarantee rate) is statistically determined, and this time point is established as the latest call time, thereby achieving risk-controllable decision-making based on probability statistics.

[0033] S130. When the current time is earlier than the latest call time, determine the delivery load attribute based on the task flow data sent to the target machine within the preset duration window.

[0034] The preset duration window refers to an evaluation period (e.g., 1 minute, 3 minutes, or 5 minutes) extending from the current moment into historical timeframes, designed to quantify material delivery pressure in the short term. Task flow data characterizes the dynamic data of all delivery tasks dispatched by the scheduling system to the target machine. This data includes, but is not limited to: detailed task information for tasks already issued or planned (e.g., task quantity, material volume / weight, estimated departure and arrival times); real-time dynamic characteristics of tasks (including the inventory of tasks in the "pending execution" state (reflecting backlog levels) and the progress of tasks in the "in transit" state (reflecting logistics resource occupancy)); the incremental number of tasks added within the window period (reflecting demand trends); and the flow efficiency indicators of tasks at key nodes (e.g., machine receiving ports) (e.g., average waiting time, queuing time, and stall time). Delivery load attributes refer to quantitative indicators (e.g., load rate) or qualitative descriptions (e.g., low load, high load, congestion risk, or resource idleness) derived from a comprehensive evaluation of task flow data. These attributes visually reflect the current and future service pressure level of the logistics delivery system on the target machine, revealing the availability and smoothness of delivery resources.

[0035] In one implementation, all task flow data destined for the target machine within a preset time window can be traversed, the total number of tasks in the "pending execution" and "in execution" states can be counted, and this total number can be compared with the theoretical maximum service capacity of the delivery equipment (such as AGV carts) within that window period (i.e., the maximum number of tasks that can be completed per unit time). If the total number of tasks is close to or exceeds the theoretical maximum service capacity, the delivery load attribute is determined to be high load or congestion; if the total number of tasks is far below the theoretical maximum service capacity, it is determined to be low load or idle. This method can intuitively and quickly reflect the degree of accumulation in the task queue and clearly express the busy status of the logistics channel.

[0036] In another implementation, the total estimated equipment time (including travel, loading and unloading) of all delivery tasks destined for the target machine within a preset time window can be accumulated. This sum is then compared with the total effective working time of available delivery equipment within the window to calculate the capacity utilization rate. Simultaneously, the distribution density of these delivery tasks along the time axis can be analyzed to detect conflicts where multiple tasks compete for the same critical path or the same delivery machine at the same time. If the capacity utilization rate exceeds a preset warning threshold, or the number of time-conflicting tasks exceeds a limit, the delivery load attribute is marked as resource-scarce or at risk of conflict; otherwise, it is marked as having sufficient capacity. This method can more accurately capture hidden congestion problems caused by time overlap and resource competition.

[0037] In another implementation, task flow data within a preset time window can be transformed into multi-dimensional feature vectors. These features include, but are not limited to, the total number of tasks, average task weight / volume, task time distribution entropy, historical delivery success rate for the same period, and the current overall traffic flow index of the workshop. Furthermore, these multi-dimensional feature vectors can be input into a pre-trained load classification model. This model learns the non-linear mapping relationship between task flow features and actual delivery delays / efficiencies from massive historical data, outputting a comprehensive delivery load attribute label (e.g., extremely high risk, moderately busy, stable operation). This method can implicitly learn and incorporate complex interference factors that are difficult to quantify (e.g., equipment aging, personnel movement interference, bottleneck effects of specific routes), thereby providing more robust load assessment results and improving the accuracy and adaptability of the assessment.

[0038] Optionally, the task flow data includes task stock data, task increment data, and flow efficiency data: Based on the task flow data sent to the target machine within a preset time window, the delivery load attributes are determined, including: determining the cumulative load attribute based on the task stock data; determining the traffic growth load attribute based on the task increment data; determining the congestion load attribute based on the flow efficiency data; and determining the delivery load attribute based on the cumulative load attribute, traffic growth load attribute, and congestion load attribute.

[0039] The task backlog data refers to the set of tasks generated but not yet delivered at the current moment. Specifically, it includes the number of tasks in the pending execution state (ordered but not dispatched) and the en route state (dispatched but not delivered), reflecting the current task backlog. The cumulative load attribute assessed based on this data characterizes the current logistics pressure level. Task increment data refers to the number of newly generated or planned delivery tasks within a preset time window, reflecting short-term demand growth trends. The flow growth load attribute assessed based on this data characterizes the upward trend of dynamic pressure in the short term. Flow efficiency data reflects the speed of processing existing delivery tasks, including average task completion time, task throughput per unit time, or average equipment turnover rate. The congestion load attribute assessed based on this data is a qualitative or quantitative description of system operational smoothness and potential congestion risks (e.g., smooth flow, slightly slow, severely congested), revealing the current throughput capacity and congestion severity of material consumption stations.

[0040] In one implementation, the total number of tasks in the pending and en route states can be counted and compared with the rated concurrent processing capacity of the target machine to obtain a ratio. If the ratio exceeds a first threshold, the cumulative load attribute is determined to be high backlog. Alternatively, the ratio can be used as the cumulative load data, or the total number of tasks can be used as the cumulative load attribute. Simultaneously, the number of newly added delivery tasks in the task increment data can be determined as the flow growth load attribute. Alternatively, the relative growth rate of the number of newly added delivery tasks compared to the historical baseline number of tasks can be calculated; this growth rate characterizes the fluctuation trend of material demand. This growth rate is used as the flow growth load attribute. Alternatively, if the growth rate is positive and large, it indicates a surge in production demand and sudden load pressure on logistics; if it is negative, it means a decline in demand. If this growth rate exceeds a second threshold, the flow growth load attribute is determined to be a surge. Furthermore, the average processing time of the current task can be calculated based on the flow efficiency data. The average processing time can be determined as the congestion load attribute. Alternatively, if the average processing time is higher than the standard working hours or its third threshold, the congestion load attribute is determined to be inefficient congestion.

[0041] Furthermore, the three attributes—cumulative load, traffic growth load, and congestion load—can be assigned different weighting coefficients (e.g., cumulative load has the highest weight, followed by congestion load, and then traffic growth load), and the delivery load attribute can be calculated using a linear weighted formula. Alternatively, the delivery load attribute can be divided into levels such as idle, stable, busy, or severely overloaded, thereby quantifying the contribution of each dimension to the overall load.

[0042] For example, a preset duration window can be used. Calculate cumulative load attributes within (e.g., 1 minute or 5 minutes) Traffic growth load attributes (close (Number of new tasks in the window), congestion load attributes (Average waiting time at key nodes in the target machine, duration of occupancy or task standstill in key sections, etc.). Delivery load attributes. .in, , , All are weights. The cumulative load attribute is represented as: ;in, for The number of material delivery tasks in the pending execution status within the window; Represented as The number of material delivery tasks in transit within the window. The traffic growth load attribute is represented as: ;in, Represented as Create new tasks within the window.

[0043] In another implementation, it's possible to determine whether the task inventory data exceeds the limit. If it does, the delivery load attribute is determined to be congested. If it doesn't exceed the limit, the next level of judgment is initiated, analyzing trends based on incremental task data. If the growth rate of newly created tasks in the incremental task data is greater than a preset value and shows no downward trend, the delivery load attribute is marked as increasing delivery pressure. Furthermore, the delivery load attribute can be corrected by combining flow efficiency data. For example, if both task inventory and incremental data are within the normal range, but flow efficiency data shows an abnormally long average time (indicating route congestion or equipment failure), the delivery load attribute is updated to latent congestion. This approach can accurately capture complex operating conditions that cannot be reflected by a single dimension, improving identification accuracy.

[0044] In another implementation, the number of existing tasks, the rate of task increment, and the reciprocal of the turnover efficiency (representing the degree of obstruction) can be used as three fuzzy input variables, mapped to fuzzy sets of "low," "medium," and "high," respectively. Fuzzy reasoning and defuzzification are then performed using a pre-defined fuzzy rule base (e.g., "If the existing task quantity is high and the increment rate is high, then the delivery load attribute is extremely high," and "If the existing task quantity is medium but the reciprocal of the turnover efficiency is extremely low, then the delivery load attribute is high") to output continuous delivery load attribute values. This setup simulates expert decision-making, effectively handling the uncertainty and fuzziness in the data, and maintaining high accuracy in load assessment even when tasks fluctuate drastically.

[0045] The technical solution provided in this embodiment subdivides task flow data into three dimensions: task stock, task increment, and flow efficiency. It then derives the attributes of cumulative load, traffic growth load, and congestion load, ultimately synthesizing them into accurate delivery load attributes. This method overcomes the limitations of traditional single-indicator evaluation, simultaneously analyzing the current backlog, future growth trends, and real-time operational efficiency. This improves the accuracy of delivery load assessment and the rationality of material scheduling, ensuring production continuity and smooth logistics.

[0046] In this embodiment, the flow efficiency data includes congestion characteristic data of at least one material consumption station in the target machine: determining the congestion load attribute based on the flow efficiency data includes: determining the congestion load attribute based on the congestion characteristic data of at least one material consumption station.

[0047] Among them, congestion characteristic data is a set of statistical indicators used to describe delays in delivery tasks at material consumption stations. Specifically, it includes waiting time (idle time from task arrival at the station to the start of unloading), queuing time (time a task waits in the station's entrance queue for the completion of a preceding task), and standby time (total time a task stops moving due to abnormal reasons during unloading). Flow efficiency data refers to a set of quantitative indicators reflecting the speed at which logistics and delivery tasks process materials at one or more stations on the target machine. Material consumption stations refer to the physical locations or interfaces (such as receiving ports or temporary storage bins) on the target machine used for receiving, temporarily storing, and supplying materials to production units.

[0048] Specifically, the average waiting time, queuing time, and standby time of all delivery tasks destined for material-consuming stations at target machines within a preset time window can be extracted from the flow efficiency data, and a comprehensive delay index can be generated through weighted calculation. This comprehensive delay index can be used as a congestion load attribute, or compared with a pre-set multi-level threshold standard to determine the level: if the comprehensive delay index is lower than the first low threshold, it indicates that the station's processing is smooth, and the congestion load attribute is determined to be smooth; if it is between the first low threshold and the second high threshold, it indicates that there is intermittent congestion, and the congestion load attribute is determined to be slightly slow; if it exceeds the second high threshold, it indicates that the station's processing capacity is saturated or an anomaly has occurred, and the congestion load attribute is determined to be severe congestion. This method transforms complex time-dimensional data into clear congestion load level labels.

[0049] Alternatively, the average waiting time of all delivery tasks destined for the target machine's material consumption station within a preset time window can be calculated and used as the congestion load attribute. Or, the sum of the waiting times of all delivery tasks destined for the target machine's material consumption station within a preset time window can be calculated and used as the congestion load attribute.

[0050] For example, congestion load attributes can be represented as: Congestion load attributes can also be expressed as: ;in For the collection of material consumption sites, For the site in Waiting / queueing time inside the window.

[0051] Alternatively, the preset time window can be divided into multiple time slices, and the rate of change of average waiting time, queuing time, or standby time within each time slice can be calculated. If the current congestion characteristic data is found to be within limits, but the rate of change is consistently positive and steep, congestion is predicted to occur soon, and the congestion load attribute is marked as an increasing risk of congestion in advance; conversely, if the data is high but shows a downward trend, the congestion load attribute is marked as congestion mitigation. This dynamic trend-based assessment mechanism can capture early signals of congestion formation, enabling target machines to take preventative measures (such as task diversion and call rhythm adjustment) before actual congestion occurs, thereby eliminating congestion in its early stages and avoiding sudden blockages.

[0052] The technical solution provided in this embodiment determines congestion load attributes by analyzing congestion characteristic data (average waiting time, queuing time, and standby time) at the material consumption sites of target machines, accurately sensing the logistical pressure of delivery tasks and the actual throughput status of the machine's receiving port. Based on these attributes, precise peak-shaving and valley-filling strategies can be implemented: for example, when high congestion load is detected, the issuance of non-urgent tasks is delayed or guided to a backup buffer to avoid exacerbating the risk of downtime; when low congestion load is detected, task flow is accelerated. This not only improves the smoothness and safety of end-of-line logistics in the packing and packaging workshop, but also maximizes the utilization rate of machine lineside space, effectively ensuring production continuity and logistics efficiency.

[0053] S140. Based on the delivery load attributes, the latest call time, the current time, and the material operation data, determine the target material request method and initiate a material request based on the target material request method.

[0054] The target material request method refers to the specific execution strategy selected for this material request, which may include various modes such as immediate emergency delivery, regular queue delivery, combined order delivery, delayed off-peak delivery, and peak shaving and valley filling delivery.

[0055] In one implementation, the time difference between the current moment and the latest call time (i.e., time margin) can be calculated, and the urgency of the material shortage risk can be analyzed in conjunction with material operation data. If the time margin is extremely small or the line-side inventory reaches a preset warning threshold, regardless of the current delivery load attributes, the target material request method is forcibly locked as immediate emergency delivery, and the highest priority queue-jumping mechanism is triggered to allocate the nearest idle equipment directly to the target machine to ensure continuous production. If the time margin is sufficient and the line-side inventory is ample, further decisions can be made based on the delivery load attributes. For example, when the delivery load attributes are idle or stable, or less than the preset load, regular queuing delivery is selected; when the delivery load attributes are busy or congested, or greater than the preset load, flexible options are available for merging and consolidating deliveries (packaging the machine's task with other machine tasks to improve transport efficiency) or delaying off-peak delivery (postponing non-urgent tasks to the off-peak period). This hierarchical decision-making method can quickly match the optimal strategy according to the urgency and logistics situation, prioritizing production stability under extreme conditions and taking into account logistics efficiency under normal conditions.

[0056] In another implementation, a multi-objective optimization function can be pre-constructed, encompassing "material shortage risk cost," "logistics and delivery cost," and "line-side congestion cost." Different candidate material requisition methods (such as immediate delivery, 5-minute delay, peak shaving and valley filling, etc.) are input variables into this multi-objective optimization function. An optimization algorithm then searches for the method that minimizes the total cost and determines it as the target material requisition method. In this model, the material shortage risk cost is dynamically determined by the difference between the current time and the latest call time, as well as material operation data; the smaller the difference, the exponentially higher the cost. Meanwhile, logistics and delivery costs and line-side congestion costs are primarily driven by delivery load attributes; the higher the load, the greater the marginal cost of initiating a delivery task. This method achieves globally optimal scheduling decisions by quantifying and balancing various costs.

[0057] In another implementation, a deep reinforcement learning model can be interactively trained with massive amounts of historical scheduling data to learn which target material delivery method under different states will yield the maximum long-term reward (the reward function is defined as the comprehensive benefit of no downtime, low congestion, and high turnover). Delivery load attributes, time margin (the difference between the current time and the latest call time), and material operation data can be encoded as environmental state vectors and input into the model. The model then outputs the target material delivery method with the highest probability. This method can capture complex nonlinear relationships and dynamic coupling characteristics that are difficult for human experts to define explicitly. As the running time increases and data accumulates, its decision-making strategy will continuously iterate and optimize, improving scheduling accuracy and efficiency.

[0058] The technical solution provided in this embodiment obtains the current time's delivery equipment data, delivery task information to the target machine, and material operation data of the target machine from the scheduling system. Based on the current time, delivery equipment data, delivery task information, and material operation data, it determines the latest call time for the target machine to initiate a material request. When the current time is earlier than the latest call time, it determines the delivery load attribute based on the task flow data to the target machine within a preset time window. Based on the delivery load attribute, the latest call time, the current time, and the material operation data, it determines the target material request method and initiates a material request based on the target material request method. This solves the problem in the prior art where material requests are made according to fixed cycles or manual experience, which can easily lead to material delivery delays or premature material arrivals, thus affecting the safety and smoothness of the production site. By combining the current time, delivery equipment data from the scheduling system, delivery task information to the target machine, and material operation data of the target machine, it determines the latest call time for the target machine to initiate a material request, thereby avoiding line-side inventory backlog caused by calling too early and production risks caused by calling too late. Furthermore, by analyzing the task flow data sent to the target machine when the current time is earlier than the latest call time, the delivery load attribute is determined. This allows for the ability to anticipate the delivery logistics capacity before issuing new material requests. Based on the delivery load attribute, the latest call time, the current time, and material operation data, the target material request method is determined. This ensures that the target material request method is precisely matched with current production needs and delivery supply capacity. Material requests are then initiated based on the target material request method, improving the on-time rate and reliability of material delivery, reducing the risk of machine downtime due to material shortages and material blockages at the production line, thereby ensuring the safety and smooth operation of the production site.

[0059] Figure 2 This is a flowchart of a material scheduling method according to an embodiment of the present invention. Based on the foregoing embodiments, step S120 is further refined. Specific implementation details can be found in the technical solution of this embodiment. Technical terms that are the same as or corresponding to those in the above embodiments will not be repeated here.

[0060] like Figure 2 As shown, the method specifically includes the following steps: S210. Obtain the current delivery equipment data of the scheduling system, the delivery task information sent to the target machine, and the material operation data of the target machine.

[0061] S220. Based on delivery equipment data and delivery task information, determine the estimated delivery time of materials.

[0062] The estimated material delivery time can be understood as the predicted time from the issuance of a delivery instruction to the arrival of the material at the target machine.

[0063] In this embodiment, the methods for determining the estimated material delivery time include, but are not limited to, the following three: The first method involves the following: Delivery equipment data includes the real-time location, real-time operating speed, acceleration / deceleration performance parameters, and usage information of the delivery equipment; delivery task information includes the coordinates of the target machine. By obtaining the real-time location of the delivery equipment and the coordinates of the target machine, and combining this with dynamic road condition information from the electronic map of the cigarette workshop (such as other equipment occupancy, temporary obstacles, etc.), an optimal conflict-free driving path is generated using a path planning algorithm. The real-time operating speed and acceleration / deceleration performance parameters of the delivery equipment in the current road segment are extracted, and the expected average speed is corrected for special operating conditions such as full load or low battery. The total length of the planned path is divided by the corrected expected average speed, and the estimated waiting time at key stations is added to calculate the estimated material delivery time.

[0064] Another approach is to pre-train a high-precision time prediction model by accessing a large number of delivery records from similar scenarios in a historical database. Current delivery task information (such as task type, distance between origin and destination, and material weight) and delivery equipment data (such as equipment model, load status, and current traffic density) are then input into the pre-trained time prediction model to output the estimated material delivery time.

[0065] Another approach is to construct a digital twin synchronized with the physical cigarette manufacturing workshop, mapping real-time delivery equipment data and delivery task information into the digital twin. Within the digital twin, the logistics operation process over a future period is simulated, mimicking the entire lifecycle of equipment movement, obstacle avoidance, charging, and loading / unloading along virtual paths. Through multiple Monte Carlo simulations, the time distribution curve of materials arriving at the target machine is statistically analyzed, and its expected value or the upper limit of the confidence interval is taken as the estimated material delivery time.

[0066] For example, key fields for each historical delivery task can be pre-acquired, including static attributes such as task ID, task type, origin station, destination station, and load / quantity. Simultaneously, real-time status data of the delivery equipment during the execution of that historical delivery task (such as equipment model, battery level, and load rate) can be acquired as static and dynamic input features for the model. The lifecycle states of historical delivery tasks are standardized into seven categories: Created, Picking, Outbound, Transporting, Arrived, Finished, and Exception. For each status node, its timestamp is precisely extracted, and the waiting time of materials at key stations under the machine is recorded synchronously. This time-series data will be used to construct a dynamic feature vector reflecting the degree of logistics congestion and scheduling efficiency. The task delivery time is configured as a prediction label for each historical task (used to train the model to predict future estimated material delivery times). Task delivery time refers to the time required for materials to reach the target machine. It can be the total time from the issuance of the call request until the material arrives at the target machine; or it can be the execution time from the actual start of the delivery task until the material arrives at the target machine. For example, starting with the call request, the task delivery time can be expressed as: ; Indicates the time at which the call request was initiated; This represents the completion time of the delivery task generated based on the call request. Starting from the start of execution, the task delivery time can be expressed as: . This indicates the start time of the delivery task generated based on the call request.

[0067] The extracted task features, device status, and timing waiting information are integrated with the calculated task delivery time to construct a complete structured training sample for subsequent training and validation of the time prediction model.

[0068] The time prediction model can be a regression model (such as random forest regression). A regression model can be trained based on training samples to fit the task time, resulting in a well-trained time prediction model. The MAE / MAPE was evaluated using cross-validation or hold-out methods, and the time prediction model version, feature parameters, and missing value handling strategies were solidified.

[0069] When the MES system generates new delivery requests or updates the status of delivery tasks, the collaborative service can read the delivery equipment data from the scheduling system in real time at the current moment. and delivery task information to the target machine. The data is input into the time prediction model to obtain the estimated material delivery time. .

[0070] Based on the above technical solution, the estimated delivery time can also be determined based on the current time and the expected material delivery time. The estimated delivery time is expressed as: .

[0071] In cigarette production logistics, determining the Estimated Time of Arrival (ETA) offers several advantages. First, by comparing the ETA with the remaining material consumption time of the machine in real time, a tiered alarm can be triggered in advance or an emergency order scheduling can be automatically initiated if the predicted arrival time is later than the safety threshold, fundamentally preventing downtime due to material shortages. Second, ETA data enables dynamic coordination of production rhythm. When a slight delay in delivery is predicted but remains within a controllable range, the Manufacturing Execution System (MES) can guide the machine to appropriately reduce its speed to match the material arrival time, or utilize this time window to schedule auxiliary operations such as equipment cleaning and label changing, maximizing production efficiency. Third, accurate ETA can optimize multi-vehicle coordination and path planning. The scheduling system can dynamically adjust the execution priority and travel path of AGVs based on the urgency of different tasks (i.e., the proximity of the ETA), avoiding congestion and improving overall logistics throughput. Fourth, visualized ETA information allows operators to monitor material arrival status in real time to rationally arrange processes. Management can continuously evaluate logistics service quality and optimize scheduling algorithm parameters based on the deviation analysis between historical ETAs and actual arrival times.

[0072] S230. Based on the current time, the estimated material delivery time, and material operation data, determine the latest call time.

[0073] In this embodiment, the real-time remaining inventory and current average consumption rate of the target machine can be extracted from the material operation data to calculate the theoretical duration for which the remaining materials can sustain production under the current production rhythm. A preset safety buffer time can be subtracted from this theoretical duration to obtain the sustainable production duration, in order to cope with instantaneous fluctuations in the consumption rate or the time spent on material receiving operations. The current moment is added to the sustainable production duration, and then the estimated material delivery time is subtracted to obtain the latest call time.

[0074] Time series analysis algorithms can also be used to analyze material operation data of the target machine over historical periods, predicting the material consumption trend curve within the expected material delivery time window. The predicted future cumulative consumption is determined by the material consumption trend curve, and then compared with the target machine's current inventory and minimum safety stock threshold to determine the critical stockout time point. This critical stockout time point can be backtracked by one expected material delivery time, dynamically determining the latest call time. This method can sensitively respond to changes in machine production speed, automatically calling earlier when production accelerates and appropriately delaying calling when production slows down, effectively avoiding the risk of line congestion caused by calling too early or material shortage downtime caused by calling too late.

[0075] The uncertainty distribution of the current time and the expected material delivery time (such as the probability of delivery delay) and the random fluctuations of material operation data (such as the variance of consumption rate) can also be simultaneously input into the risk assessment model. The risk assessment model calculates the probability of a material shortage accident occurring at different call times and determines the latest call time as the latest time when the probability of material shortage is lower than a preset tolerance (such as 0.1%).

[0076] Optionally, the material operation data includes currently available material data and material consumption rate: Based on the current time, the estimated material delivery time, and the material operation data, the latest call time is determined, including: based on the currently available material data and material consumption rate, determining the remaining support time; based on the current time and the remaining support time, determining the theoretical material shortage time; and based on the theoretical material shortage time, the estimated material delivery time, and the preset safety margin time, determining the latest call time.

[0077] The currently available material data refers to the actual remaining quantity of cigarette material in the target machine's lineside warehouse or temporary storage area at the current moment, usually measured in sticks, trays, or cartons. Material consumption rate refers to the quantitative indicator of material consumed per unit time (e.g., per minute) at the current production speed, reflecting the real-time production load. Remaining support time refers to the maximum time the existing inventory can sustain the machine's continuous operation without replenishing new materials. The theoretical material shortage moment represents the point at which the target machine will stop due to material depletion without intervention. The preset safety margin time is a buffer period reserved to cope with fluctuations in consumption rate, unexpected delivery delays, or time-consuming material receiving operations.

[0078] Specifically, the remaining available material time can be calculated by quotienting the current available material data and the material consumption rate. Further, the current time and the remaining available time are summed to obtain the theoretical material shortage time. Based on this, two subtraction operations are performed: a preset safety margin time is subtracted from the theoretical material shortage time to obtain the absolute safety cutoff point, designed to provide a buffer for unforeseen circumstances; then, the estimated material delivery time is subtracted from this absolute safety cutoff point to obtain the latest call time.

[0079] For example, the target machine can be calculated according to formula (1). Remaining available duration Formula (1) can be expressed as: ; This refers to the existing inventory at the production line (i.e., currently available material data, which can be converted into "supportable quantity"). Let be the consumption rate (units / minute). The latest call time that can be safely waited for can be determined based on formula (2). Formula (2) can be expressed as: .in This is expressed as the preset safety margin duration (in minutes). Represented as the current time; This indicates the estimated delivery time of the materials.

[0080] The advantages of the above settings are that they can ensure the continuity of material replenishment and production consumption in the time dimension, effectively eliminating the risk of material shortage and machine downtime caused by human experience and judgment errors; at the same time, by introducing safety margins and accurate duration calculations, they can avoid material redundancy at the line and blockage of logistics channels caused by premature calls, thereby improving the material turnover efficiency of the coiling and packaging workshop and the overall operational stability of the production system.

[0081] S240. When the current time is earlier than the latest call time, determine the delivery load attribute based on the task flow data sent to the target machine within the preset duration window.

[0082] S250: Based on the delivery load attributes, the latest call time, the current time, and material operation data, determine the target material request method and initiate a material request based on the target material request method.

[0083] The technical solution provided in this embodiment calculates the estimated material delivery time by integrating real-time equipment status and specific task characteristics. This not only improves the accuracy of material scheduling and effectively avoids the risks of machine downtime due to material shortages or material accumulation and blockage caused by prediction deviations, but also enables continuous optimization of logistics rhythm in complex and ever-changing workshop environments, ensuring the continuous, stable, and efficient operation of the cigarette production line. Furthermore, by comprehensively considering the current moment, the accurately predicted delivery time, and real-time material operation data to determine the latest call time, it achieves precise matching between material requests and production consumption, accurately avoiding the risk of machine downtime due to material delivery delays, while preventing material accumulation and congestion at the line side caused by premature calls. This ensures that new materials are accurately delivered within a safe buffer period before existing inventory is depleted, thereby improving production continuity and logistics efficiency.

[0084] Figure 3 This is a flowchart of a material scheduling method according to an embodiment of the present invention. Based on the foregoing embodiments, step S140 is further refined. Specific implementation details can be found in the technical solution of this embodiment. Technical terms that are the same as or corresponding to those in the above embodiments will not be repeated here.

[0085] like Figure 3 As shown, the method specifically includes the following steps: S310: Obtain the current delivery equipment data, delivery task information to the target machine, and material operation data of the target machine from the dispatch system.

[0086] S320: Based on the current time, delivery equipment data, delivery task information, and material operation data, determine the latest call time for the target machine to initiate a material request.

[0087] S330. When the current time is earlier than the latest call time, determine the delivery load attribute based on the task flow data sent to the target machine within the preset duration window.

[0088] S340. Determine the earliest call time based on the current time and material operation data.

[0089] The earliest call time is the earliest time point at which the target machine is allowed to initiate a material request, calculated based on the current time and material operation data. If a material request is made earlier than this earliest call time, it may cause the new material to arrive too early, leading to line congestion or exceeding the temporary storage capacity.

[0090] In this embodiment, the longest duration that the existing inventory can support after deducting the safety buffer can be calculated based on the material operation data. This time can be added to the current time and subtracted by a maximum allowable lead time to determine the earliest call time.

[0091] Optionally, the material operation data includes currently available material data, the upper limit of material quantity at the line edge, the material consumption rate, and the amount of material replenished per cycle. Based on the current time and the material operation data, the earliest call time is determined, including: determining the expected overflow amount based on the currently available material data, the upper limit of material quantity at the line edge, and the amount of material replenished per cycle; determining the buffer waiting time based on the material consumption rate and the expected overflow amount; and determining the earliest call time based on the current time, the buffer waiting time, and the preset zero lower limit.

[0092] The currently available material data refers to the actual amount of material remaining in the line-side warehouse at the current moment. The line-side material upper limit is the maximum amount of material allowed to be stored in the line-side warehouse by physical space or process specifications; exceeding this amount will lead to unloading difficulties or safety hazards. The material consumption rate is a quantitative indicator of the amount of material consumed by the target machine per unit time. Single replenishment quantity is the standard quantity of material transported by the delivery system each time a task is executed. The expected overflow quantity refers to the portion of the material exceeding the line-side material upper limit after immediate replenishment, reflecting potential congestion risks. The buffer waiting time is the time required to consume this expected overflow quantity and reduce the line-side inventory to a safe level to receive new materials. The preset zero lower limit is a very small positive number or zero value, representing the minimum safety stock threshold allowed when a call is initiated or the theoretical earliest time point baseline.

[0093] In practice, the current available material data can be added to the single replenishment quantity to obtain a cumulative value; the cumulative value is then subtracted from the upper limit of the material supply at the line edge to obtain the expected overflow. Alternatively, the current available material data can be added to the single replenishment quantity to obtain a cumulative value; the cumulative value is then subtracted from the upper limit of the material supply at the line edge to obtain a difference. If the difference is greater than zero, it is used as the expected overflow; if the difference is less than or equal to zero, the expected overflow is zero.

[0094] Next, the expected overflow can be divided by the real-time material consumption rate to obtain the precise buffer waiting time (if the overflow is zero, the waiting time is zero). Finally, this buffer waiting time is added to the current time, and a preset zero lower limit is introduced as a verification benchmark (i.e., ensuring that the calculated time is not earlier than the logical baseline of "current time + zero lower limit" to prevent negative time or logical paradoxes). The final determined time is taken as the earliest call time. Alternatively, the buffer waiting time and the preset zero lower limit can be compared, and the maximum value between the buffer waiting time and the preset zero lower limit can be taken. This maximum value is then added to the current time to obtain the earliest call time. The advantage of setting the earliest call time is that it prevents premature replenishment from causing line-side inventory to overflow (exceeding the upper limit inventory limit) while ensuring continuous material supply.

[0095] For example, the earliest call time can be represented as: ;in, Indicated as the earliest call time; This is the upper limit of the line-side inventory (i.e., the upper limit of the line-side materials). It is the amount of material replenished at one time at the line edge; This is the data for currently available materials. Material consumption rate (units / minute); optionally, 0 is a preset zero lower limit; This indicates taking the maximum value.

[0096] Furthermore, after calculating the expected overflow, it can be divided by a modified consumption rate (e.g., the real-time consumption rate multiplied by a safety factor less than 1, or the lowest consumption rate over a past period) to obtain a longer, more robust buffer waiting time, ensuring that overflow does not occur even when the machine is running at low speed. When determining the earliest call time, in addition to the current time and buffer waiting time, a preset zero lower limit is set to a small operational preparation time (such as the time required for equipment startup or communication delays), ensuring that the calculated earliest call time is feasible in actual operation. This approach, by introducing safety redundancy, effectively addresses uncertainties in the production process and improves the safety of material scheduling.

[0097] Alternatively, starting from the current moment, simulations can be performed backward in extremely small time steps (e.g., 1 second). Within each step, material consumption (based on the material consumption rate) and assumed replenishment arrival (based on the single replenishment amount) are simulated. The sum of the current inventory and the single replenishment amount is continuously checked to ensure it does not exceed the upper limit of the material supply at the line. The simulation identifies a moment when the expected overflow drops to zero or falls below the small tolerance range represented by a preset zero lower limit; this moment can be marked as the end of the buffer waiting time. The endpoint derived from this can then be used as the earliest call time.

[0098] By comprehensively utilizing current available material data, the upper limit of material quantity at the production line, the amount of material replenished at one time, and the material consumption rate, the expected overflow amount and buffer waiting time are derived sequentially. Finally, by combining the current time with the preset zero lower limit, the earliest call time is determined. This accurately calculates the time window required to digest the conflict between existing inventory and new materials, ensuring that there is always room in the production line warehouse when new materials arrive. This not only eliminates on-site safety hazards and equipment downtime risks caused by material overflow, but also improves the utilization rate of production line space and the cleanliness of the logistics site, achieving refined and safe management of cigarette production logistics and ensuring the efficiency and continuity of cigarette production.

[0099] S350: Generate the target call time based on the latest call time and the earliest call time.

[0100] Among them, the target call time is an effective call time window composed of the earliest call time and the latest call time, which represents the optimal execution time after comprehensively balancing production demand and logistics status.

[0101] In one implementation, the minimum of the latest and earliest call times can be used as the left threshold for the target call time, and the latest call time can be used as the right threshold. Alternatively, a point in time between the latest and earliest call times can be randomly selected, and this point in time can be combined with either the latest or earliest call time to form the target call time.

[0102] In another implementation, a comprehensive cost function can be predefined, including the line-side inventory holding cost (the earlier the call, the longer the inventory backlog time, and the higher the cost) and the logistics delivery delay risk cost (the later the call, the higher the risk of lateness due to delivery load, and the higher the cost). The weighting coefficient of the delivery delay risk cost is dynamically determined directly by the delivery load attribute: the higher the load, the greater the risk weight. Within the closed interval between the earliest and latest call times, the time point that minimizes the comprehensive cost function is traversed or searched, and this point is determined as the optimal target call time.

[0103] In another implementation, the target call time can be dynamically determined based on delivery load attributes within a time window consisting of the earliest and latest call times. For example, if the delivery load attributes indicate low load, the target call time can be set close to the earliest call time to make full use of idle capacity and complete the task as early as possible; if the delivery load attributes indicate high load (i.e., the delivery load attributes are greater than a preset high load threshold), the target call time can be shifted towards the latest call time to avoid the current peak by appropriately delaying the process.

[0104] For example, the target call time can be represented as [ ]; ; The selection method can be: ;in This refers to the delivery load attribute. If all delivery load attributes within the preset time window are greater than the preset high load threshold, then... Equal to the latest call time The system will display a high-load reason code and provide a prompt. If the load delivery attribute within the preset time window is not greater than the preset high-load threshold, then... Equal to the earliest call time .

[0105] Furthermore, the time difference between the current time and the earliest call time, the time difference between the current time and the latest call time, and the level of delivery load attributes can be used as fuzzy input variables. Through a preset rule base (such as "if the load is low and far from the latest call time, the target time is biased towards the earliest call time"; "if the load is high and close to the latest call time, the target time is close to the latest call time but needs to be expedited"), the precise target call time can be output.

[0106] S360: Based on the delivery load attributes, generate a target material request method that includes the target call time.

[0107] Among them, the target material request method may include a defined target call time, and may also include an execution strategy determined based on delivery load attributes (such as immediate execution, delayed execution, peak shaving and valley filling, priority route planning, etc.).

[0108] In this embodiment, different delivery load attributes can be associated with different material request methods. When determining the delivery load attribute at the current moment, the material request method associated with that delivery load attribute can be obtained and determined as the target material request method. For example, when the delivery load attribute indicates high load, the generated target material request method includes a delayed material request instruction; if the delivery load attribute indicates low load, the target material request method includes an early material request instruction.

[0109] Optionally, based on the delivery load attributes, a target material request method including the target call time is generated, including: when the delivery load attributes are greater than a preset high load threshold, a recommended waiting instruction including the target call time is generated; when the delivery load attributes are not greater than the preset high load threshold, a material request advance instruction including the target call time is generated.

[0110] The preset high-load threshold can be a critical value used to characterize the overall material delivery task corresponding to the target machine as being under high load. The recommended waiting instruction instructs the target machine to further postpone issuing material requests beyond the target call time to avoid the current logistics peak. It can also include suggested delay durations or new execution times. The material request advance instruction instructs the target machine to issue material requests earlier than the target call time, prioritizing delivery during the current relatively idle logistics window to prevent delays caused by future load increases.

[0111] In this embodiment, the delivery load attribute can be compared with a preset high load threshold. When the delivery load attribute is greater than the preset high load threshold, it can be determined that the current logistics channel is congested, and a recommended waiting instruction is generated. This instruction is used to indicate that the original target call time be postponed by a peak avoidance period (e.g., 10 or 15 minutes), and it can be marked as off-peak execution. Conversely, when the delivery load attribute is not greater than the preset high load threshold, it can be determined that the current transportation capacity is sufficient, and a material demand advance instruction is generated. This instruction is used to indicate that the original target call time be advanced by a rush period (e.g., 5 minutes) to ensure that materials can enter the transportation process as early as possible, and it is marked as priority execution.

[0112] For example, if the delivery load attribute is greater than a preset high load threshold and the current time is less than the latest call time, then for non-urgent material requests, a suggestion to wait / not to call is provided, along with a recommended target call time. When the delivery load attribute is not greater than the preset high load threshold, normal calls are allowed within the target call time, or it is suggested that requests that can be moved forward to fill the valley. If there are requests that can be executed in advance, a request forwarding suggestion instruction is generated, guiding the adjustment of requests to the current time period to achieve load balancing (valley filling).

[0113] Furthermore, when the delivery load attribute exceeds a preset high load threshold, the system calculates the proportion exceeding the threshold and maps it to a dynamic waiting coefficient. The higher the delivery load attribute, the larger the waiting coefficient. The generated recommended waiting instruction will suggest postponing the target call time for a longer period, potentially triggering a batch delivery strategy to completely avoid congestion peaks. When the delivery load attribute is not greater than the preset high load threshold, the system calculates the space where the delivery load attribute is below the threshold and maps it to a forward shift coefficient. The lower the delivery load attribute (i.e., the more idle capacity), the larger the forward shift coefficient. The generated material demand forward shift instruction will suggest advancing the target call time to fully utilize idle capacity for pre-delivery.

[0114] Furthermore, historical data can be used to predict load change trends in the near future (i.e., within the recommended waiting or forwarding time window). When the delivery load attribute exceeds a preset high load threshold, it can simulate whether the future load will decrease to a safe range if the recommended waiting instruction is executed. If the prediction shows that congestion will still occur after waiting, the instruction can include a strategy of waiting and merging other tasks to maximize the efficiency of a single transport. If the current delivery load attribute is not greater than the preset high load threshold, it simulates whether executing the material demand forwarding instruction will preempt resources from more urgent future tasks. If the prediction shows no conflict, a material demand forwarding instruction is generated.

[0115] This solution flexibly generates recommended waiting instructions or material demand advance instructions based on the comparison between delivery load attributes and preset high load thresholds. During high load periods, recommended waiting proactively avoids congestion, preventing new tasks from exacerbating system bottlenecks and ensuring logistics network stability. During low load periods, advance material demand proactively seizes idle resources, shortening material transit time and improving response speed. This not only effectively smooths out peak and trough periods in the packaging workshop's logistics system, reducing the probability of delivery delays due to congestion, but also improves the utilization rate of delivery equipment, thereby increasing production efficiency and reducing logistics costs.

[0116] In this embodiment, when the current time is not earlier than the latest call time, a target material request method including an immediate call instruction is generated. The immediate call instruction is a highest-priority scheduling command that requires skipping all regular queuing and immediately triggering the issuance of the material request.

[0117] Specifically, it can compare the current time with the latest call time in real time. When the current time is detected to be later than or equal to the latest call time, a target material request method containing an immediate call instruction can be generated, and based on this, a highest-priority material request can be sent to the scheduling system. This enables logistics and delivery tasks to trigger a queue-jumping mechanism, forcibly allocating the nearest available transportation capacity directly to the target machine, thereby completing material replenishment within the last window before inventory depletion, eliminating the risk of production line downtime caused by call delays, and ensuring the continuity and stability of production operations.

[0118] For example, if the latest call time is less than or equal to the current time, the material request is marked as an urgent need and an immediate call is prompted.

[0119] Furthermore, when generating a target material request method that includes an immediate call instruction, the system can simultaneously scan the pool of currently available delivery equipment to identify the delivery equipment closest to the target machine or in optimal operating condition for executing the emergency material request. This operation effectively shortens response time and improves the delivery success rate in emergency situations.

[0120] The technical solution of this embodiment constructs a safe time window by defining the earliest and latest call times to obtain the target call time. Then, based on the delivery load attributes, it accurately generates the target material request method within the target call time, effectively solving the problem of premature congestion or late material shortage caused by traditional fixed-cycle calls. The earliest call time prevents ineffective accumulation of materials and waste of space at the line, while the latest call time ensures production continuity. By using the delivery load attributes to find the best balance between the two, it achieves peak shaving and valley filling of logistics resources, thereby improving the material turnover efficiency of the packing workshop, reducing line-side inventory costs, and maximizing the safety of the production environment in a complex logistics environment.

[0121] As an optional embodiment of the above embodiments, specific application scenario examples are provided to enable those skilled in the art to further understand the technical solutions of the embodiments of the present invention. Specifically, please refer to the following detailed content.

[0122] The technical solution provided in this implementation can be implemented by a material scheduling system. The material scheduling system may include a Material Execution System (MES), a collaborative service (or an ETA and load assessment service), an AGV scheduling and execution system (i.e., an AGV system), a machine terminal (maintaining one or more target machines), and a data storage and training module. The MES system generates material requirements and receives recommendation information, presenting task progress, delivery load attributes, and estimated material delivery time to the machine terminal. The AGV system handles inventory preparation, outbound delivery, transportation, and arrival confirmation, outputting task status logs and queue information. The collaborative service, acting as an intermediary between the MES and AGV systems, receives task and status data, performs feature calculations, ETA inference, load calculations, and recommendation generation, and sends this information back to the MES system / machine terminal via an interface. The data storage and training module stores historical delivery task information and delivery equipment data, and periodically trains / updates the time prediction model and load threshold parameters.

[0123] See Figure 4Specifically, the implementation can be as follows: the machine terminal sends the material request (which may carry the urgent priority information of the material request) of the target machine to the MES system; the MES system creates a delivery task based on the received material request or submits the material request to the collaborative service. After receiving the delivery task or original material request from the MES system, the collaborative service can call the pre-trained time prediction model and load assessment algorithm in the data storage and training module, and combine it with real-time AGV location, power consumption, path congestion status and other delivery equipment data and delivery task information to dynamically calculate the estimated material delivery time for each delivery task, and assess the current overall system load level (i.e., delivery load attributes); based on the progress priority and load balancing recommendation strategy, it sorts and matches the delivery tasks in the task queue with resources. For example, for material requests marked as "urgent," their scheduling priority is increased and the nearest available delivery equipment is assigned. For routine tasks, the optimal dispatch plan is generated by comprehensively considering AGV idleness, path conflict probability, and energy efficiency. The target material request method, including the target machine ID, material type, target call time, recommended AGV number, and estimated material delivery time, is sent to the AGV system. Simultaneously, the task status, target material request method, estimated material delivery time, and delivery load attributes are synchronously transmitted back to the MES system and the machine terminal, achieving transparent management and control throughout the entire process. In addition, the collaborative service can continuously collect closed-loop data such as the actual AGV running trajectory, delivery time, and machine material receiving feedback, and periodically trigger incremental updates of the data storage and training modules to ensure the accuracy of the estimated material delivery time prediction and the adaptive optimization of the scheduling strategy as the production environment changes. This maximizes the utilization of logistics resources and suppresses line-side inventory backlog while ensuring production line continuity. The machine terminal displays the target material request method, delivery task progress, ETA prediction value, and load assessment results to facilitate determining when to initiate a material request based on the target material request method.

[0124] Below is an example of the lifecycle status (i.e., progress status) of a delivery task.

[0125] See Figure 5The lifecycle of a delivery task can be divided into several key nodes, forming a closed-loop state flow logic: The delivery task starts in the "creation" state, where the MES system or collaborative service generates an initial delivery instruction based on the machine's material request; it enters the "stocking" state, indicating that the warehouse or line-side warehouse is preparing the corresponding materials. If insufficient inventory or anomalies are found during the stocking process, the "out of stock / fault" branch is triggered, and the task enters the "abnormal / cancelled" state, and the upstream system is notified for rescheduling or manual intervention; after the stocking is completed, the task progresses to the "outbound" state, where the AGV has picked up the materials and started to execute the outbound action. If equipment failure, path blockage, or operational error occurs at this stage, the task jumps to the "abnormal / cancelled" state through the "fault / reassignment" mechanism, and the cause of the failure is recorded for subsequent analysis. After successful outbound delivery, the delivery task enters the "Transporting" phase. The AGV travels to the target machine along the planned route. If unforeseen circumstances arise en route (such as insufficient power or traffic congestion), the "Block / Fault" mechanism can be used to reassess the situation and may revert to "Abnormal / Cancelled" or dynamically adjust the route. When the AGV arrives at the target machine and completes the physical handover, "Arrival Confirmation" is triggered, and the task status is updated to "Arrived." The "Completion" phase then begins, containing two sub-processes: "Receipt / Completion" signifies that the machine operator has confirmed receipt and closed the task, while "Rejection / Abnormal" handles situations such as incorrect, damaged, or mismatched materials, which also lead to "Abnormal / Cancelled." Finally, all tasks that are normally completed or abnormally terminated are archived in the data storage module. Their entire lifecycle trajectory (including the duration of each status, the reason for the transition, and the responsible party) is fully recorded to support the training of time prediction models and the optimization of scheduling strategies. This state transition configuration ensures process standardization and fault tolerance, guaranteeing that every delivery task is traceable, controllable, and optimizable.

[0126] Based on the above technical solutions, a time prediction model can be pre-trained. See also Figure 6 This system can obtain historical delivery task information and equipment status data. The historical delivery task information and equipment status data are cleaned (e.g., denoised or completed) and aligned to obtain preprocessed historical delivery task information and equipment status data. Delivery times corresponding to the historical delivery task information and equipment status data are configured to construct training samples. The historical delivery task information and equipment status data from the training samples are input into a regression model for feature extraction, obtaining historical delivery task features and equipment status features. Based on these features, the delivery times are predicted. The predicted delivery times are compared with the delivery times in the training samples using a loss function to obtain an error value. The model parameters of the regression model are then adjusted based on this error value to obtain a trained time prediction model.

[0127] In practical applications, real-time delivery task information (such as delivery origin and destination, task type, quantity of delivered materials, etc.) and real-time delivery equipment data (such as queue backlog or data quality, material resource capacity or delivery cost, congestion level, availability / accuracy / timeliness of delivery equipment) can be obtained. The real-time delivery task information and data are input into a trained time prediction model, which outputs the estimated material delivery time. Then, based on the current time, the estimated material delivery time, and the target machine's material operation data, the latest call time for the target machine to send a material request to the scheduling system is determined. If the current time is earlier than the latest call time, the delivery load attribute is determined based on the task flow data sent to the target machine by the scheduling system within a preset time window. Based on the delivery load attribute, the latest call time, the current time, and the material operation data, the target material request method is determined, and scheduling is performed on the target machine's material request based on this method.

[0128] Simultaneously, it can record each recommended target material request method, the actual material request sent, the actual delivery time of the delivery task corresponding to the material request, prediction error, and reasons for anomalies, etc., for periodic training of the time prediction model, calibration of load thresholds (such as preset high load thresholds) and recommended material request methods, so as to achieve continuous optimization.

[0129] The following is an example of determining delivery load attributes.

[0130] See Figure 7 The system can calculate cumulative load attributes, traffic growth load attributes, and congestion load attributes separately. Then, it normalizes and weights these attributes to obtain the delivery load attributes. The relationship between the delivery load attributes and different load thresholds is then determined. If the delivery load attribute is greater than a preset high load threshold, the load level is high, and a recommended waiting instruction including the target call time can be generated. If the delivery load attribute is not greater than the preset high load threshold, the load level is medium or low, and a material demand advance instruction including the target call time can be generated. This allows normal calls within the target call time or suggests advancing the demand to fill the valley. In other words, if there are material demands that can be executed in advance, a demand advance suggestion instruction can be generated to guide the adjustment of demand to the current time period to achieve load balancing (valley filling).

[0131] The following is an example of how to determine the target material requirements.

[0132] See Figure 8It can obtain current available material data, preset safety margin duration, and material consumption rate. Based on the current available material data and material consumption rate, it determines the remaining support duration. Based on the current time and remaining support duration, it determines the theoretical material shortage time; based on the theoretical material shortage time, estimated material delivery time, and preset safety margin duration, it determines the latest call time. Simultaneously, it can determine the earliest call time based on current available material data, line-side material limit, material consumption rate, single replenishment quantity, and the current time. Based on the latest and earliest call times, it generates a target call time. According to delivery load attributes, it generates a target material requisition method including the target call time (such as delivery load attributes for delivery load level, reason, and estimated arrival countdown), and outputs the target material requisition method.

[0133] For example, the remaining support time of cigarette material at the current line edge of a certain machine. =45min, preset safety margin duration ==15min; Collaborative service predicts the estimated delivery time of materials for the current call. =12min, then the latest call time = current time + (45-15)-12 = current time + 18min. If the delivery load attribute is greater than the preset high load threshold, then for non-urgent requests, the message "High load does not recommend immediate calling. You can wait for the load to change to a level where the delivery load attribute is not greater than the preset high load threshold within 18 minutes before submitting" will be displayed. If you wait until the delivery load attribute is not greater than the preset high load threshold before submitting, you can reduce queue peaks and congestion spread, while still meeting the uninterrupted supply constraint.

[0134] The technical solution provided by this invention effectively reduces blind, repeated calls and ineffective waiting by providing transparent task stages and accurate estimated material delivery times to the machine side, thereby reducing delivery time and improving on-time delivery (OTD) rate. Furthermore, by utilizing delivery load grading and recommended target call times, some non-urgent demands are intelligently migrated from high-load periods to low-load periods, achieving peak shaving and valley filling, reducing the peak and average ratio of pending and en route tasks, and alleviating queuing congestion and road segment occupation problems at key stations. Simultaneously, by combining remaining inventory (i.e., currently available material data), material consumption rate, and estimated material delivery time to construct the latest call constraint time, it ensures that while utilizing off-peak periods to fill valleys, the safety boundary of continuous material supply is strictly maintained, reducing the number and duration of downtime due to material shortages. In addition, through the earliest call constraint and the reminder against pre-stocking, excessive inventory events can be suppressed, reducing work-in-process inventory, channel occupation, and secondary handling, improving logistics efficiency. Moreover, the traceable closed loop formed by recommended reason codes, prediction error analysis, and anomaly records supports continuous iterative optimization of thresholds and models.

[0135] Figure 9This is a schematic diagram of the structure of a material scheduling device according to an embodiment of the present invention. Figure 9 As shown, the device includes: a data acquisition module 410, a latest call time determination module 420, a delivery load attribute determination module 430, and a target material demand method determination module 440.

[0136] The system includes a data acquisition module 410, which acquires delivery equipment data, delivery task information to the target machine, and material operation data of the target machine at the current time. A latest call time determination module 420 determines the latest call time for the target machine to initiate a material request based on the current time, the delivery equipment data, the delivery task information, and the material operation data. A delivery load attribute determination module 430 determines the delivery load attribute based on task flow data to the target machine within a preset time window when the current time is earlier than the latest call time. A target material request method determination module 440 determines the target material request method based on the delivery load attribute, the latest call time, the current time, and the material operation data, and schedules the material request of the target machine based on the target material request method.

[0137] The technical solution of this embodiment obtains the delivery equipment data of the scheduling system, the delivery task information sent to the target machine, and the material operation data of the target machine at the current time; based on the current time, delivery equipment data, delivery task information, and material operation data, it determines the latest call time for the target machine to initiate a material request; when the current time is earlier than the latest call time, it determines the delivery load attribute according to the task flow data sent to the target machine within a preset time window; based on the delivery load attribute, the latest call time, the current time, and the material operation data, it determines the target material request method, and initiates a material request based on the target material request method. This solves the problem in the prior art that material requests based on fixed cycles or manual experience are prone to material delivery delays or premature material arrival, thereby affecting the safety and smoothness of the production site. It realizes that by combining the current time, the delivery equipment data of the scheduling system, the delivery task information sent to the target machine, and the material operation data of the target machine, it can determine the latest call time for the target machine to initiate a material request, thereby avoiding the backlog of line-side inventory caused by calling too early and avoiding the production risks caused by calling too late. Furthermore, by analyzing the task flow data sent to the target machine when the current time is earlier than the latest call time, the delivery load attribute is determined. This allows for the ability to anticipate the delivery logistics capacity before issuing new material requests. Based on the delivery load attribute, the latest call time, the current time, and material operation data, the target material request method is determined. This ensures that the target material request method is precisely matched with current production needs and delivery supply capacity. Material requests are then initiated based on the target material request method, improving the on-time rate and reliability of material delivery, reducing the risk of machine downtime due to material shortages and material blockages at the production line, thereby ensuring the safety and smooth operation of the production site.

[0138] Optionally, based on the above-described device, the latest call time determination module 420 includes: The estimated material delivery time determination unit is used to determine the estimated material delivery time based on the delivery equipment data and the delivery task information; The latest call time determination unit is used to determine the latest call time based on the current time, the estimated material delivery time, and the material operation data.

[0139] Based on the above-mentioned device, optionally, the material operation data includes currently available material data and material consumption rate; the latest call time determination unit includes: The remaining support time determination subunit is used to determine the remaining support time based on the current available material data and the material consumption rate; Theoretical material shortage time determination subunit is used to determine the theoretical material shortage time based on the current time and the remaining support time; The latest call time determination subunit is used to determine the latest call time based on the theoretical material shortage time, the expected material delivery time, and the preset safety margin time.

[0140] Based on the above-mentioned device, optionally, the task flow data includes task inventory data, task incremental data, and flow efficiency data; the delivery load attribute determination module 430 includes: The cumulative load attribute determination unit is used to determine the cumulative load attribute based on the task inventory data; wherein, the task inventory data includes the number of tasks in the pending execution state and the in-transit state; A traffic growth load attribute determination unit is used to determine the traffic growth load attribute based on the task increment data; wherein, the task increment data includes the number of delivery tasks newly created within the preset duration window; A congestion load attribute determination unit is used to determine congestion load attributes based on the flow efficiency data; The delivery load attribute determination unit is used to determine the delivery load attribute based on the cumulative load attribute, the traffic growth load attribute, and the congestion load attribute.

[0141] Based on the above-mentioned device, optionally, the flow efficiency data includes congestion characteristic data of at least one material consumption station in the target machine: the congestion load attribute determination unit is used to determine the congestion load attribute based on the congestion characteristic data of at least one material consumption station; The congestion characteristic data includes the average waiting time, queuing time, or stall time of the delivery tasks of the scheduling system at the material consumption stations within the preset time window.

[0142] Based on the above-mentioned device, the optional target material determination module 440 includes: The earliest call time determination unit is used to determine the earliest call time based on the current time and the material operation data; The target call time determination unit is used to generate a target call time based on the latest call time and the earliest call time; The target material request method determination unit is used to generate a target material request method that includes the target call time based on the delivery load attributes.

[0143] Based on the above-mentioned device, optionally, the material operation data includes currently available material data, the upper limit of material quantity at the line edge, the material consumption rate, and the amount of material replenished per cycle; the earliest call time determination unit includes: The expected overflow determination unit is used to determine the expected overflow based on the currently available material data, the upper limit of the material at the line edge, and the amount of material replenished in a single operation; A buffer waiting time determination unit is used to determine the buffer waiting time based on the material consumption rate and the expected overflow amount; The earliest call time determination subunit is used to determine the earliest call time based on the current time, the buffer waiting time, and the preset zero lower limit.

[0144] Based on the above-mentioned device, optionally, a target material demand determination unit is used to generate a recommended waiting instruction including the target call time when the delivery load attribute is greater than a preset high load threshold; and to generate a material demand advance instruction including the target call time when the delivery load attribute is not greater than the preset high load threshold.

[0145] Optionally, based on the above-described apparatus, the apparatus is further configured to generate a target request method including an immediate call instruction when the current time is not earlier than the latest call time.

[0146] The material scheduling device provided in the embodiments of the present invention can execute the material scheduling method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method execution.

[0147] Figure 10 This is a schematic diagram of the structure of an electronic device implementing the material scheduling method of embodiments of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0148] like Figure 10 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory 12 or a random access memory 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the read-only memory 12 or a computer program loaded from storage unit 18 into the random access memory 13. The random access memory 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, read-only memory 12, and random access memory 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0149] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0150] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as material scheduling methods.

[0151] In some embodiments, the material scheduling method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via read-only memory 12 and / or communication unit 19. When the computer program is loaded into random access memory 13 and executed by processor 11, one or more steps of the material scheduling method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to execute the material scheduling method by any other suitable means (e.g., by means of firmware).

[0152] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0153] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0154] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0155] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0156] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0157] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0158] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication unit 19, or installed from storage unit 18, or installed from read-only memory 12. When the computer program is executed by processor 11, it performs the functions defined in the methods of the embodiments of the present invention.

[0159] This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the material scheduling method provided in any embodiment of this invention.

[0160] In implementing the computer program product, computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0161] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0162] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A material scheduling method, characterized in that, include: Obtain the current delivery equipment data, delivery task information to the target machine, and material operation data of the target machine from the dispatch system; Based on the current time, the delivery equipment data, the delivery task information, and the material operation data, determine the latest call time for the target machine to initiate a material request; When the current time is earlier than the latest call time, the delivery load attribute is determined based on the task flow data sent to the target machine within a preset duration window; Based on the delivery load attributes, the latest call time, the current time, and the material operation data, a target material request method is determined, and the material request is initiated based on the target material request method.

2. The method according to claim 1, characterized in that, The process of determining the latest call time for the target machine to initiate a material request based on the current time, the delivery equipment data, the delivery task information, and the material operation data includes: Based on the delivery equipment data and the delivery task information, the estimated material delivery time is determined; The latest call time is determined based on the current time, the estimated material delivery time, and the material operation data.

3. The method according to claim 2, characterized in that, The material operation data includes currently available material data and material consumption rate. Determining the latest call time based on the current time, the estimated material delivery time, and the material operation data includes: Based on the currently available material data and the material consumption rate, determine the remaining support time; Based on the current moment and the remaining available time, determine the theoretical moment of material shortage; The latest call time is determined based on the theoretical material shortage time, the estimated material delivery time, and the preset safety margin time.

4. The method according to claim 1, characterized in that, The task flow data includes task inventory data, task increment data, and flow efficiency data. Determining the delivery load attributes based on the task flow data sent to the target machine within a preset time window includes: Based on the task inventory data, the cumulative load attribute is determined; wherein, the task inventory data includes the number of tasks in the pending execution state and the in-transit state; Based on the incremental task data, the traffic growth load attribute is determined; wherein, the incremental task data includes the number of delivery tasks newly created within the preset duration window; Based on the aforementioned flow efficiency data, the congestion load attributes are determined; The delivery load attribute is determined based on the cumulative load attribute, the traffic growth load attribute, and the congestion load attribute.

5. The method according to claim 4, characterized in that, The flow efficiency data includes congestion characteristic data of at least one material consumption station in the target machine; the determination of congestion load attributes based on the flow efficiency data includes: The congestion load attribute is determined based on congestion characteristic data of at least one of the material consumption sites; The congestion characteristic data includes the average waiting time, queuing time, or stall time of the delivery tasks of the scheduling system at the material consumption stations within the preset time window.

6. The method according to claim 1, characterized in that, The method of determining the target material requisition method based on the delivery load attributes, the latest call time, the current time, and the material operation data includes: Based on the current time and the material operation data, determine the earliest call time; The target call time is generated based on the latest call time and the earliest call time; Based on the delivery load attributes, a target material requisition method including the target call time is generated.

7. The method according to claim 6, characterized in that, The material operation data includes currently available material data, line-side material upper limit, material consumption rate, and single replenishment quantity; determining the earliest call time based on the current time and the material operation data includes: Based on the currently available material data, the upper limit of material quantity at the line edge, and the amount of material replenished at one time, the expected overflow amount is determined; The buffer waiting time is determined based on the material consumption rate and the expected overflow amount; The earliest call time is determined based on the current time, the buffer waiting time, and the preset zero lower limit.

8. The method according to claim 6, characterized in that, The step of generating a target material request method that includes the target call time based on the delivery load attributes includes: When the delivery load attribute is greater than the preset high load threshold, a recommended waiting instruction including the target call time is generated; When the delivery load attribute is not greater than the preset high load threshold, a material demand advance instruction including the target call time is generated.

9. The method according to claim 1, characterized in that, The method further includes: When the current time is not earlier than the latest call time, a target request method including an immediate call instruction is generated.

10. A material dispatching device, characterized in that, include: The data acquisition module is used to acquire the delivery equipment data of the scheduling system at the current moment, the delivery task information sent to the target machine, and the material operation data of the target machine; The latest call time determination module is used to determine the latest call time for the target machine to initiate a material request based on the current time, the delivery equipment data, the delivery task information, and the material operation data; The delivery load attribute determination module is used to determine the delivery load attribute based on the task flow data sent to the target machine within a preset duration window when the current time is earlier than the latest call time. The target material request method determination module is used to determine the target material request method based on the delivery load attribute, the latest call time, the current time and the material operation data, so as to schedule the material request request of the target machine based on the target material request method.

Images