Intelligent material distribution and scheduling optimization method and system for metal tile production line

Abstract

The invention relates to the technical field of metal tile production automation, and discloses an intelligent material distribution and scheduling optimization method and system for a metal tile production line. According to the method, distribution early warning is generated by obtaining material demand information and a production state, a material shortage risk index is calculated based on material consumption prediction, distribution request time sequence features are determined and sorted to form a task queue, then a distribution batch is generated, a shortest path is planned, and finally an instruction is sent to a distribution skip car to achieve distribution. The material distribution efficiency can be optimized, the material shortage risk of the production line is reduced, and the operation stability and the resource utilization rate of the production line are improved.

Description

Technical Field

[0001] This invention relates to the field of automation technology in metal tile production, and in particular to an intelligent material distribution and scheduling optimization method and system for metal tile production lines. Background Technology

[0002] In modern manufacturing, metal roofing tile production lines are a crucial component of building material production, and their production efficiency and stability directly impact a company's economic benefits. The production process involves multiple steps, including raw material preparation, forming, cutting, and coating, each requiring different types and quantities of materials. Traditionally, material distribution on metal roofing tile production lines relies heavily on manual experience and fixed-cycle delivery. However, with the expansion of production scale and the increase in product variety, the need for intelligent management of material distribution and production scheduling is becoming increasingly prominent.

[0003] Currently, the material distribution and scheduling system of metal tile production lines suffers from several problems: inaccurate material demand forecasting and the inability of traditional distribution methods to dynamically adjust distribution strategies based on actual production status lead to material backlogs at some workstations and material shortages at others, affecting the overall efficiency of the production line; the lack of scientific basis for determining material distribution priorities makes it impossible to effectively identify and respond to material demands at high-risk workstations, resulting in frequent production line downtime due to material shortages and increased production costs; and unreasonable material distribution path planning with arbitrary selection of material cart routes, failing to comprehensively consider the distribution needs of multiple workstations and material cart location information, leading to low distribution efficiency and serious resource waste.

[0004] With the development of intelligent manufacturing technology, using big data analysis and intelligent algorithms to optimize material distribution processes and achieve accurate prediction, reasonable scheduling and efficient distribution of material supply to the production line has become a key way to improve the production efficiency of metal tiles and reduce operating costs. Summary of the Invention

[0005] The present invention provides an intelligent material distribution and scheduling optimization method and system for metal tile production lines, which can at least solve some of the problems existing in the prior art.

[0006] A first aspect of this invention provides an intelligent material distribution and scheduling optimization method for a metal tile production line, comprising: Obtain material demand information for each workstation on the metal tile production line and real-time production status information of the metal tile production line; generate material delivery early warning information based on the material demand information and the real-time production status information. Based on the material delivery early warning information and the material consumption fluctuation data of each workstation, the predicted material consumption value is obtained, and the material shortage risk index of each workstation is determined based on the predicted material consumption value. The timing characteristics of material delivery requests at each workstation are determined based on the material shortage risk index. The material delivery requests at each workstation are prioritized based on the timing characteristics to generate a material delivery task queue. Based on the distance information between adjacent workstations in the material delivery task queue, material delivery batches are generated, and the shortest delivery path is planned for each material delivery batch based on the real-time location information of the material cart. Based on the material delivery batch and the shortest delivery route, a material delivery instruction is generated and sent to the corresponding delivery vehicle to instruct the delivery vehicle to carry out delivery.

[0007] Obtain material demand information for each workstation on the metal tile production line and real-time production status information of the metal tile production line; generate material delivery early warning information based on the material demand information and the real-time production status information, including: Based on the material demand information and the real-time production status information, the material consumption fluctuation characteristics are determined by analyzing the phase differences in the material transfer process between workstations. The material consumption fluctuation characteristics are then matched with the workstation production cycle time to obtain the dynamic correspondence between the production cycle time of each workstation and the material consumption. Based on the dynamic correspondence, analyze the material complementary configuration scheme between each workstation, and establish the correlation data between material consumption and production status based on the material complementary configuration scheme; Establish a cascading influence relationship of material consumption between workstations based on the associated data, determine the material consumption mutation point of the target workstation based on the cascading influence relationship, and take the material consumption mutation point as the critical state time point of material consumption. The material consumption critical state time point is compared with the material demand time to generate initial material delivery warning information. The initial material delivery warning information is then calibrated based on the real-time production status information to obtain calibrated material delivery warning information.

[0008] Based on the material delivery early warning information and the material consumption fluctuation data of each workstation, a predicted material consumption value is obtained. Based on the predicted material consumption value, a material shortage risk index for each workstation is determined, including: Based on the material delivery early warning information, the vibration frequency and material conveying speed of the material conveying line at each workstation are collected. Based on the correspondence between the vibration frequency and the material conveying speed, the stable range of material transmission is determined, and material consumption fluctuation data of each workstation is generated. Based on the material consumption fluctuation data and the material transmission stability range, the temporal variation characteristics of material consumption at each workstation are analyzed to obtain the material consumption evaluation benchmark. Based on the material consumption assessment benchmark, the material loss coefficient in the workstation material conversion process is determined, and based on the material loss coefficient, the corresponding curve of material consumption and product qualification rate is determined to generate the predicted value of material consumption. The predicted material consumption value is compared with the stable material transmission range to determine the fluctuation nodes in the material transmission process. Based on the deviation between the material backlog status of the fluctuation nodes and the predicted material consumption value, an initial material shortage risk index is generated. Based on the correlation between the initial material shortage risk index and the material backlog status, the material shortage risk index for each workstation is determined.

[0009] Based on the material consumption assessment benchmark, the material loss coefficient during the material conversion process at the workstation is determined. Based on the material loss coefficient, a curve showing the correlation between material consumption and product qualification rate is determined, and a predicted material consumption value is generated, including: The material consumption fluctuation cycle is obtained by calculating the real-time conversion ratio of material input parameters and output parameters at each workstation based on the material consumption assessment benchmark. Based on the material consumption fluctuation cycle, the stable range and fluctuation range of the material transportation process at each workstation are analyzed to obtain the material transportation state sequence; A workstation material loss coefficient is established based on the material transport state sequence and the material consumption fluctuation cycle, and the material loss trend characteristics are determined based on the workstation material loss coefficient and the workstation processing parameters. Based on the material loss trend characteristics, a mapping curve between material consumption and product qualification rate is generated, and the material loss correlation factor is determined. Based on the degree of matching between the material loss correlation factor and the material transportation state sequence, the key workstations for material loss are determined. Based on the process parameters of the key material loss stations and the mapping curve, material consumption prediction parameters are calculated, and material consumption prediction values ​​are generated according to the coupling relationship between the material consumption prediction parameters and the material loss trend characteristics.

[0010] Based on the material shortage risk index, the temporal characteristics of material delivery requests for each workstation are determined. Based on these temporal characteristics, the material delivery requests for each workstation are prioritized, and a material delivery task queue is generated, including: Based on the material shortage risk index, the frequency and duration of material delivery requests at each workstation are statistically analyzed to generate the temporal characteristics of material delivery requests at each workstation. The correlation coefficient between workstation capacity utilization and material consumption is determined based on the material shortage risk index, and material demand forecasting parameters are calculated based on the correlation coefficient and the time series characteristics. Based on the material demand forecasting parameters, the priority value of material delivery requests for each workstation is determined. Then, based on these priority values, the spatiotemporal overlap matrix of material delivery to each workstation and the material transport path topology between adjacent workstations are determined. The material delivery association strength is calculated based on the material transport path topology and the temporal characteristics to obtain the material delivery batch division parameters. Based on the material delivery batch division parameters and the spatiotemporal overlap matrix, material delivery requests are prioritized and sorted to generate a material delivery task queue.

[0011] Based on the distance information between adjacent workstations in the material delivery task queue, material delivery batches are generated. The shortest delivery path is planned for each material delivery batch based on the real-time location information of the material carts, including: Based on the material delivery start position information and material delivery target position information in the material delivery task queue, calculate the material delivery distance coefficient between adjacent workstations; Based on the material delivery distance coefficient, the material delivery connection strength between adjacent workstations is extracted, and a material delivery association sequence is generated. Based on the overlap between the material delivery association sequence and the material delivery time information in the material delivery task queue, multiple material delivery batches are obtained; Based on the multiple material delivery batches, extract the material vehicle running trajectory features from the real-time location information of the material vehicle, combine the material delivery association sequence to determine the material delivery spatial constraint boundary, and generate a multi-batch material delivery path planning scheme. Based on the multi-batch material delivery route planning scheme, analyze the degree of deviation between the material vehicle running trajectory and the material delivery space constraint boundary, and determine the material delivery route optimization parameters; The material delivery route optimization parameters are used to adjust the multi-batch material delivery route planning scheme to generate the shortest material delivery route.

[0012] Based on the multiple material delivery batches, the trajectory features of the material vehicles are extracted from the real-time location information of the material vehicles. Combined with the material delivery association sequence, the spatial constraint boundaries of the material delivery are determined, and a multi-batch material delivery path planning scheme is generated, including: Collect real-time location information of the material cart, and extract the change characteristics of the material cart's running acceleration and running speed based on the real-time location information to obtain the material cart's running trajectory characteristics; Based on the characteristics of the material car's running trajectory, the time of the running state transition during the material car's operation is determined, and a material delivery association sequence is established according to the running state transition time and the material demand of the workstation; Based on the material delivery association sequence, extract the time window constraints and spatial area constraints of workstation material delivery to obtain the material delivery spatial constraint boundary. Based on the material delivery space constraint boundary, the mapping relationship between the material car running trajectory segment and the workstation layout topology is determined, and the location information of key points of the material car running trajectory is obtained. Based on the key point location information of the material vehicle's running trajectory and the time of the running state transition, a segmented planning criterion for material delivery path is generated. Based on the segmented planning criterion for material delivery path, a batch path planning is performed for the multiple material delivery batches to generate a multi-batch material delivery path planning scheme.

[0013] A second aspect of the present invention provides an intelligent material distribution and scheduling optimization system for a metal tile production line, comprising: The first unit is used to acquire material demand information of each workstation on the metal tile production line and real-time production status information of the metal tile production line, and generate material delivery early warning information based on the material demand information and the real-time production status information. The second unit is used to predict the material consumption forecast value based on the material delivery early warning information and the material consumption fluctuation data of each workstation, and to determine the material shortage risk index of each workstation based on the material consumption forecast value. The third unit is used to determine the temporal characteristics of material delivery requests for each workstation based on the material shortage risk index, prioritize the material delivery requests for each workstation based on the temporal characteristics, and generate a material delivery task queue. The fourth unit is used to generate material delivery batches based on the distance information between adjacent workstations in the material delivery task queue, and to plan the shortest delivery path for each material delivery batch based on the real-time location information of the material cart. The fifth unit is used to generate a material delivery instruction based on the material delivery batch and the shortest delivery route, and send the material delivery instruction to the corresponding delivery vehicle to instruct the delivery vehicle to make delivery.

[0014] A third aspect of the present invention provides an electronic device, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0015] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0016] This invention obtains material demand information and real-time production status information of each workstation on the metal tile production line, and performs predictive analysis by combining material consumption fluctuation data. It can accurately identify the risk of material shortage at each workstation, thereby achieving scientific early warning and precise scheduling of material distribution, and effectively avoiding production line shutdowns due to material shortages.

[0017] By establishing a priority ranking mechanism based on the material shortage risk index and combining it with the distance information between workstations to generate reasonable material delivery batches, the arrangement of material delivery tasks is made more reasonable, improving material delivery efficiency, reducing unnecessary round-trip transportation, and lowering the time cost and energy consumption of material delivery.

[0018] Based on the real-time location information of the material carts, the shortest delivery route is planned and intelligent material delivery instructions are generated, realizing the automation and intelligence of the material delivery process, reducing manual intervention, improving the accuracy and timeliness of material delivery, thereby ensuring the continuous and stable operation of the metal tile production line and improving the overall production efficiency. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating the intelligent material distribution and scheduling optimization method for a metal tile production line according to an embodiment of the present invention.

[0020] Figure 2 This is a schematic diagram of the process for generating predicted material consumption values ​​according to an embodiment of the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0023] Figure 1 This is a flowchart illustrating the intelligent material distribution and scheduling optimization method for a metal tile production line according to an embodiment of the present invention. Figure 1 As shown, the method includes: Obtain material demand information for each workstation on the metal tile production line and real-time production status information of the metal tile production line; generate material delivery early warning information based on the material demand information and the real-time production status information. Based on the material delivery early warning information and the material consumption fluctuation data of each workstation, the predicted material consumption value is obtained, and the material shortage risk index of each workstation is determined based on the predicted material consumption value. The timing characteristics of material delivery requests at each workstation are determined based on the material shortage risk index. The material delivery requests at each workstation are prioritized based on the timing characteristics to generate a material delivery task queue. Based on the distance information between adjacent workstations in the material delivery task queue, material delivery batches are generated, and the shortest delivery path is planned for each material delivery batch based on the real-time location information of the material cart. Based on the material delivery batch and the shortest delivery route, a material delivery instruction is generated and sent to the corresponding delivery vehicle to instruct the delivery vehicle to carry out delivery.

[0024] In this embodiment, material demand information for each workstation on the metal tile production line is first acquired, including the current material inventory, material consumption rate, and minimum safety stock threshold for each workstation. This material demand information is collected in real-time by IoT sensors installed at each workstation, with a sensor acquisition frequency of once every 10 seconds. The collected data is transmitted to the central data processing server via industrial Ethernet. Simultaneously, real-time production status information for the metal tile production line is acquired, including the production line operating speed, the currently produced metal tile model, and the equipment operating status at each workstation. This real-time production status information is obtained through a data interface provided by the production line control system, with a acquisition frequency of once every 5 seconds.

[0025] Based on the acquired material demand information and real-time production status information, a material delivery early warning message is generated. The early warning mechanism calculates the estimated time for material depletion based on the difference between the current material inventory and the safety stock threshold at each workstation, combined with the material consumption rate. When the estimated time is less than a preset warning time (e.g., 30 minutes), a material delivery early warning message is generated. The early warning message includes key information such as workstation number, material type, current inventory, and estimated depletion time. For example, when the raw steel plate inventory at workstation A12 is 50kg, the current consumption rate is 5kg / minute, and the safety stock threshold is 20kg, it is calculated that the raw steel plate inventory at this workstation will drop to the safety stock threshold in 6 minutes, thus generating an early warning message.

[0026] For workstations where early warning information has been generated, the predicted material consumption is further calculated based on the material delivery early warning information and the material consumption fluctuation data for each workstation. Material consumption fluctuation data refers to the fluctuation in material consumption at each workstation over a certain period (e.g., the last 7 days), including statistical indicators such as average consumption rate, standard deviation, and peak consumption rate. Using time series analysis methods, considering factors such as the workday effect and the impact of production batch changes, the material consumption at each workstation is predicted for a specific future time period (e.g., the next 2 hours). The prediction results are expressed as predicted material consumption values, in units of material consumption per minute.

[0027] Based on the predicted material consumption, a material shortage risk index is determined for each workstation. The material shortage risk index measures the likelihood of a material shortage at a workstation, ranging from 0 to 100, with higher values ​​indicating a higher risk. The risk index calculation considers the following factors: the ratio of current inventory to the safety stock threshold, the stability of the predicted material consumption, whether the workstation is a production bottleneck, and the difficulty of material replenishment. When the current inventory at a workstation is 1.2 times the safety stock threshold, the predicted material consumption fluctuates significantly (coefficient of variation exceeding 0.3), and the workstation is a production bottleneck, its material shortage risk index reaches 85 or higher, indicating an extremely high risk.

[0028] Based on the material shortage risk index of each workstation, the temporal characteristics of material delivery requests for each workstation are determined. These temporal characteristics include delivery urgency, delivery time window, and delivery frequency. Delivery urgency is directly determined by the material shortage risk index; the delivery time window is calculated based on the current inventory level, safety stock threshold, and material consumption forecast; and the delivery frequency is determined based on the stability of the material consumption forecast and material storage capacity. For example, for a workstation with a risk index of 90, its delivery urgency can be determined as "extremely high," its delivery time window as "within 10 minutes," and its delivery frequency as "once every 30 minutes."

[0029] Based on the timing characteristics of material delivery requests at each workstation, these requests are prioritized. The prioritization algorithm comprehensively considers factors such as material shortage risk index, workstation importance, and delivery urgency. A weighted scoring mechanism is used for ranking, and the weights of each factor can be dynamically adjusted according to actual production conditions. The resulting material delivery task queue can be an ordered list of delivery tasks, with each task containing information such as workstation number, material type, delivery quantity, and expected delivery time.

[0030] After generating the material delivery task queue, tasks are merged into material delivery batches based on the distance information between adjacent workstations in the queue. This distance information comes from a pre-established production line workstation layout map, recording the actual physical distance between each workstation. A clustering algorithm is used to merge delivery tasks from workstations that are close together and have similar delivery urgency into a single delivery batch to improve delivery efficiency. For example, if workstations A12, A13, and A15 all need to deliver steel plate raw materials, and the physical distance between these three workstations is within 20 meters, their delivery tasks are merged into a single delivery batch.

[0031] For each material delivery batch, the shortest delivery route is planned based on the real-time location information of the material carts. This real-time location information is obtained through GPS or indoor positioning systems installed on the carts, with a positioning accuracy of ±1 meter and a location update frequency of once every 2 seconds. An improved shortest path algorithm is employed, considering factors such as aisle constraints, temporary obstacles, and the positions of other material carts within the production workshop, to plan the shortest path from the current material cart position to each target workstation. The path planning results include detailed navigation instructions, such as "Drive forward 15 meters, turn right, then drive forward 10 meters to reach workstation A12."

[0032] Finally, material delivery instructions are generated based on the material delivery batch and the shortest delivery route, and these instructions are sent to the corresponding delivery vehicles. The material delivery instructions use a standardized format and include the following information: delivery batch number, delivery material list (workstation-material type-quantity), delivery route navigation instructions, and estimated delivery completion time. The instructions are transmitted wirelessly to the terminal device on the delivery vehicle, which converts them into executable operation instructions and onboard display information. The delivery vehicle operator executes the material delivery task based on the displayed information, or, in the case of an autonomous delivery vehicle, the vehicle autonomously completes delivery route navigation and material unloading.

[0033] In a practical application scenario, assume a metal tile production line has 20 workstations distributed across a 500-square-meter workshop. At a certain moment, the material inventory levels at workstations A12, B05, and C07 are detected to be approaching the safety stock threshold, generating corresponding material delivery warnings. Based on historical data, the predicted material consumption for these three workstations in the next hour is 150kg, 80kg, and 120kg respectively, and their material shortage risk indices are calculated to be 85, 60, and 75 respectively. Based on these indices, workstation A12 is determined to have the highest delivery priority and should be delivered within 10 minutes; workstations B05 and C07 can be delivered within 20 minutes. It is detected that material cart 1 is currently closest to workstation A12 (25 meters) and its path passes by workstation C07; therefore, the delivery tasks for A12 and C07 are combined into one batch and assigned to material cart 1; the delivery task for workstation B05 is assigned to material cart 2. A delivery route of 65 meters in length was planned for material cart 1, with an estimated delivery time of 7 minutes; a delivery route of 80 meters in length was planned for material cart 2, with an estimated delivery time of 9 minutes. Ultimately, both material carts successfully completed their material delivery tasks according to the instructions, ensuring the continuous operation of the production line.

[0034] Through the above, the present invention realizes intelligent management of material distribution in metal tile production lines, effectively improves material distribution efficiency, reduces the risk of production line downtime due to material shortages, and is of great significance to improving the production efficiency of metal tiles.

[0035] In one optional implementation, material demand information for each workstation on the metal tile production line and real-time production status information of the metal tile production line are obtained. Material delivery early warning information is generated based on the material demand information and the real-time production status information, including: Based on the material demand information and the real-time production status information, the material consumption fluctuation characteristics are determined by analyzing the phase differences in the material transfer process between workstations. The material consumption fluctuation characteristics are then matched with the workstation production cycle time to obtain the dynamic correspondence between the production cycle time of each workstation and the material consumption. Based on the dynamic correspondence, analyze the material complementary configuration scheme between each workstation, and establish the correlation data between material consumption and production status based on the material complementary configuration scheme; Establish a cascading influence relationship of material consumption between workstations based on the associated data, determine the material consumption mutation point of the target workstation based on the cascading influence relationship, and take the material consumption mutation point as the critical state time point of material consumption. The material consumption critical state time point is compared with the material demand time to generate initial material delivery warning information. The initial material delivery warning information is then calibrated based on the real-time production status information to obtain calibrated material delivery warning information.

[0036] In this embodiment, it is first necessary to obtain the material requirements information for each workstation on the metal tile production line and the real-time production status information of the production line. The material requirements information includes parameters such as the type, quantity, and specifications of the materials required for each workstation, while the real-time production status information includes data such as the production rate of each workstation, equipment operating status, and current production batch.

[0037] Material consumption monitoring sensors deployed at each workstation collect data on material usage, while real-time production status data is obtained from the production control system. Taking workstation A as an example, this workstation is responsible for flattening metal coils and records the coil consumption every 10 minutes. After 24 hours of continuous monitoring, the data shows that under normal production conditions, this workstation consumes approximately 120 meters of coil per hour. However, during batch switching, there are consumption fluctuations of 15-20 minutes, with consumption dropping to 80 meters per hour or briefly increasing to 150 meters per hour.

[0038] Analyzing the collected data identified phase differences in material transfer between workstations. For example, after the metal coil at workstation A is flattened, it needs to be transferred to workstation B for cutting, and then to workstation C for forming. By comparing the material consumption timing data of the three workstations, it was found that the cutting operation at workstation B starts consuming materials approximately 8 minutes later than the flattening operation at workstation A, and the forming operation at workstation C starts consuming materials approximately 12 minutes later than the forming operation at workstation B. This phase difference constitutes the fluctuation characteristic of material consumption.

[0039] These material consumption fluctuations were time-matched with the production cycle time of each workstation. For example, workstation A's production cycle time is 2 meters of rolled material per minute, workstation B's is 1.8 meters of rolled material cut per minute, and workstation C's is 1.7 meters of cut material processed per minute. Through time-series analysis, a dynamic correspondence table was established, recording the material consumption patterns of each workstation at different production rates. When the production cycle time of workstation A increases to 2.5 meters per minute, the material demand of workstation B will increase accordingly after approximately 6 minutes, while the increase in material demand at workstation C will require an additional wait of approximately 10 minutes.

[0040] Based on the above dynamic correspondence, the material complementary configuration scheme between each workstation is analyzed. For example, when it is detected that workstation B temporarily slows down due to equipment adjustment, the amount of semi-finished products accumulated between workstations A and B is calculated, and the production time that this accumulation can support is predicted. If the accumulated material is sufficient for workstation B to digest after returning to normal speed, there is no need to adjust the upstream supply; if there is too much accumulated material, it is recommended to temporarily reduce the production speed of workstation A to avoid excessive accumulation of work-in-process.

[0041] These complementary material configuration schemes are linked to the actual production status to establish a database linking material consumption and production status. This database records the material flow under different production parameters. For example, at the standard production speed, the material transfer time per unit material from station A to station D is about 35 minutes; when the production line speed is increased by 10%, this transfer time is shortened to 32 minutes.

[0042] By utilizing relational databases, a cascading impact model of material consumption between workstations can be established. When material consumption changes at an upstream workstation, it's possible to predict how this change will propagate along the production chain and affect downstream workstations. Taking a metal tile production line as an example, if the roll material specification at workstation A changes from 0.8 mm thickness to 1.2 mm thickness, this change will cause the cutting equipment at workstation B to need parameter adjustments, the forming pressure at workstation C to need to increase, and the packaging material at workstation D to need to be replaced.

[0043] Based on the cascading effects, identify the abrupt change in material consumption at the target workstation. For example, when a production order switches from regular metal roofing sheets to corrosion-resistant metal roofing sheets, workstation C requires a special coating material, resulting in a significant abrupt change in its consumption pattern. Analysis of historical data reveals that this abrupt change typically occurs in the third production hour after the order switch, at which point the coating material consumption at workstation C increases from 5 kg per hour to 12 kg per hour. This point in time is marked as the critical point for material consumption.

[0044] The identified critical material consumption time point is compared with the material demand time. Assuming that the coating material inventory at workstation C is only 20 kg, based on the abrupt consumption rate, the material will be exhausted in 1.67 hours, while it takes 2 hours to transfer new material from the warehouse to workstation C. Based on this, an initial material delivery warning is generated, indicating that the coating material will be exhausted 0.33 hours before the expected time.

[0045] Finally, the initial warning was calibrated based on real-time production status information. If it was detected that the production speed at workstation C was temporarily reduced by 20% due to equipment maintenance, and the actual coating material consumption rate dropped to 9.6 kg per hour, the corrected material depletion time would be extended to 2.08 hours, slightly longer than the material delivery time. Based on this, the warning information was updated, and the emergency level was downgraded from "Immediate Delivery" to "Prepare for Delivery." It was also recommended that material management personnel closely monitor the production recovery at workstation C, and immediately initiate the material delivery process once the production speed rebounded.

[0046] The above technical solutions can accurately predict the material demand time of each workstation, issue early warning information in advance, effectively avoid production interruptions caused by material shortages, optimize material delivery routes and timing, and improve overall production efficiency.

[0047] In one optional implementation, a predicted material consumption value is obtained based on the material delivery early warning information and the material consumption fluctuation data of each workstation, and a material shortage risk index for each workstation is determined based on the predicted material consumption value, including: Based on the material delivery early warning information, the vibration frequency and material conveying speed of the material conveying line at each workstation are collected. Based on the correspondence between the vibration frequency and the material conveying speed, the stable range of material transmission is determined, and material consumption fluctuation data of each workstation is generated. Based on the material consumption fluctuation data and the material transmission stability range, the temporal variation characteristics of material consumption at each workstation are analyzed to obtain the material consumption evaluation benchmark. Based on the material consumption assessment benchmark, the material loss coefficient in the workstation material conversion process is determined, and based on the material loss coefficient, the corresponding curve of material consumption and product qualification rate is determined to generate the predicted value of material consumption. The predicted material consumption value is compared with the stable material transmission range to determine the fluctuation nodes in the material transmission process. Based on the deviation between the material backlog status of the fluctuation nodes and the predicted material consumption value, an initial material shortage risk index is generated. Based on the correlation between the initial material shortage risk index and the material backlog status, the material shortage risk index for each workstation is determined.

[0048] In one specific embodiment, the vibration frequency and material conveying speed of the material conveying line at each workstation are first collected based on material delivery early warning information. In practical applications, vibration sensors can be installed at key nodes of the material conveying line to collect vibration signals and convert them into vibration frequency values. For example, for workstation A on a standard assembly line, the vibration frequency of its material conveying line is maintained between 18Hz and 22Hz during normal operation, corresponding to a material conveying speed of 0.8m / s to 1.2m / s. Through the analysis of the correlation between vibration frequency and conveying speed, the stable material transmission range of this workstation is determined to be a vibration frequency of 20±2Hz and a conveying speed of 1.0±0.2m / s. Data points are collected at 15-minute intervals to generate a material consumption fluctuation dataset. For example, during a continuous 8-hour production process, the material consumption fluctuation data for workstation A shows that the average consumption rate for the first 4 hours is 100±5 units / hour, while the consumption rate for the last 4 hours becomes 95±8 units / hour due to equipment fine-tuning.

[0049] Based on the collected material consumption fluctuation data and the determined stable range of material transmission, the temporal variation characteristics of material consumption at each workstation are further analyzed. Specifically, the material consumption fluctuation data for seven consecutive days is divided into three groups according to work shifts: morning shift (8:00-16:00), afternoon shift (16:00-24:00), and night shift (0:00-8:00), and the material consumption patterns of different shifts are analyzed. Taking workstation A as an example, the data shows that the material consumption during the morning shift is relatively stable, with a fluctuation range of ±3%; the material consumption during the afternoon shift increases slightly, with a fluctuation range of ±5%; and the material consumption during the night shift decreases, with a fluctuation range of ±8%. Based on this, a material consumption temporal characteristic map is generated, identifying the peak period (usually occurring within 1 hour after shift change) and the trough period (usually occurring within 1 hour before the end of shift). Through these temporal characteristic analyses, a material consumption assessment benchmark is obtained. For example, the standard material consumption rate for workstation A is 98 units / hour, with a peak period coefficient of 1.15 and a trough period coefficient of 0.85.

[0050] After establishing the material consumption assessment benchmark, the material loss coefficient during the material conversion process at each workstation was further determined. In actual production environments, materials experience a certain degree of loss during transport and processing, which is related to various factors. By analyzing 30 batch material conversion records, the difference between the material input and the actual number of finished products was calculated, resulting in a material loss coefficient of 3.2% for workstation A. Based on this loss coefficient, a curve showing the correlation between material consumption and product qualification rate was established. Data shows that when the material loss coefficient is controlled within 3.0%, the product qualification rate can reach 99.5%; when the loss coefficient is between 3.0% and 4.0%, the product qualification rate is 97.8% to 99.5%; and when the loss coefficient exceeds 4.0%, the product qualification rate drops significantly to below 97.8%. Combining the material consumption assessment benchmark and the material loss coefficient, a predicted material consumption value was generated, estimating that workstation A will consume 785 ± 24 units of material in the next 8 hours.

[0051] The predicted material consumption values ​​were compared with the stable material transport range. The comparison method involved overlaying the predicted consumption curve with the upper and lower limits of the stable range to identify intersection points as potential fluctuation nodes in the material transport process. In the case of workstation A, three fluctuation nodes were identified: at hour 2 (predicted consumption rate rose to 105 units / hour, exceeding the upper limit of the stable range), hour 4 (predicted consumption rate decreased to 90 units / hour, approaching the lower limit of the stable range), and hour 7 (predicted consumption rate fluctuated drastically, ranging from 95±15 units / hour). Further analysis of the material backlog at these fluctuation nodes revealed that at hour 2, the material backlog was 15 units, far below the preset buffer value of 50 units; at hour 4, there was no material backlog; and at hour 7, the material backlog fluctuated between 0 and 35 units. The stockpiled material status was compared with the predicted material consumption, and the deviation values ​​were calculated: the deviation was 20 units in the second hour (predicted consumption 105 units, actual usable 90 units), the deviation was 0 units in the fourth hour, and the average deviation was 10 units in the seventh hour. Based on these deviation values, an initial material shortage risk index was generated: 0.8 (high risk) in the second hour, 0.3 (low risk) in the fourth hour, and 0.6 (medium risk) in the seventh hour.

[0052] Finally, based on the correlation between the initial material shortage risk index and the material backlog status, the final material shortage risk index for each workstation was determined. The correlation calculation considered the duration of material backlog, the trend of backlog changes, and the degree of matching with predicted consumption. For workstation A, the analysis revealed that the high-risk backlog in the second hour lasted for a short time (approximately 15 minutes), but the backlog changed significantly (rapidly decreasing from 30 to 15), showing a high correlation with predicted consumption; therefore, the final risk index was adjusted to 0.85. The low-risk backlog in the fourth hour was stable (no backlog at all), matching predicted consumption, and the final risk index remained at 0.3. The medium-risk backlog in the seventh hour fluctuated significantly and had a weak correlation with predicted consumption; therefore, the final risk index was adjusted to 0.5. Combining the risk indices at these three time points with their time distribution weights, the overall material shortage risk index for workstation A over the next 8 hours was calculated to be 0.58, which is considered a medium-risk level. Based on this risk assessment result, an early warning can be issued to production management personnel, suggesting timely adjustments to the material distribution plan to ensure the continuous and stable operation of the production line.

[0053] In one optional implementation, the material loss coefficient during the material conversion process at the workstation is determined based on the material consumption assessment benchmark. A corresponding curve between material consumption and product qualification rate is then determined based on the material loss coefficient, and a predicted material consumption value is generated, including: The material consumption fluctuation cycle is obtained by calculating the real-time conversion ratio of material input parameters and output parameters at each workstation based on the material consumption assessment benchmark. Based on the material consumption fluctuation cycle, the stable range and fluctuation range of the material transportation process at each workstation are analyzed to obtain the material transportation state sequence; A workstation material loss coefficient is established based on the material transport state sequence and the material consumption fluctuation cycle, and the material loss trend characteristics are determined based on the workstation material loss coefficient and the workstation processing parameters. Based on the material loss trend characteristics, a mapping curve between material consumption and product qualification rate is generated, and the material loss correlation factor is determined. Based on the degree of matching between the material loss correlation factor and the material transportation state sequence, the key workstations for material loss are determined. Based on the process parameters of the key material loss stations and the mapping curve, material consumption prediction parameters are calculated, and material consumption prediction values ​​are generated according to the coupling relationship between the material consumption prediction parameters and the material loss trend characteristics.

[0054] Figure 2 This is a schematic diagram of the process for generating predicted material consumption values ​​according to an embodiment of the present invention. Figure 2 As shown, in this embodiment, material consumption data and product quality data at each workstation are first collected to construct a material consumption assessment benchmark. This benchmark includes key indicators such as workstation material input parameters, workstation material output parameters, and product qualification rate. Sensors are used to monitor the material input and output at each workstation in real time. For example, at the injection molding workstation, the raw material particle input is recorded as 100 kg, the total weight of the molded parts is 92 kg, and the weight of scrap is 5 kg.

[0055] Based on the material consumption assessment benchmark, the real-time conversion ratio of material input and output parameters at each workstation was calculated. Taking the injection molding workstation as an example, with an input of 100kg of raw material and an effective output of 92kg, the real-time conversion ratio was 92%. Continuous monitoring of the conversion ratio over a production cycle (e.g., 8 hours) revealed that the conversion ratio was 90% in the initial stage of production (0-1 hour), 93% in the stable period (1-6 hours), and decreased to 91% in the later stage of production (6-8 hours). These data identified an 8-hour fluctuation cycle for material consumption, with significant fluctuations within each cycle.

[0056] Based on the material consumption fluctuation cycle, the stable and fluctuating ranges in the material transport process of each station are analyzed. In the injection molding station example, the 1-6 hour period is identified as the stable range (conversion rate fluctuation range ±1%), and the 0-1 hour and 6-8 hour periods are identified as the fluctuating ranges (conversion rate fluctuation range ±3%). These range information are encoded into a material transport state sequence, such as "fluctuation-stable-fluctuation", and mapped to specific time periods on the time axis.

[0057] Material loss coefficients for each workstation are established based on the material transport state sequence and material consumption fluctuation cycle. In the injection molding workstation example, the material loss coefficient is set to 0.07 (corresponding to 7% material loss) in the stable range, 0.10 in the initial stage of the fluctuation range, and 0.09 in the later stage of the fluctuation range. The correlation between these parameters and the material loss coefficient is analyzed in conjunction with the workstation's processing parameters, such as injection temperature, pressure, and holding time. For example, when the injection temperature increases from 195℃ to 210℃, the material loss coefficient increases from 0.07 to 0.09. Through these correlation analyses, the material loss trend characteristics are derived, indicating that increased temperature and unstable pressure lead to increased material loss.

[0058] Based on the material loss trend characteristics, a mapping curve between material consumption and product qualification rate is generated. In production data analysis, when the material loss coefficient is 0.07, the product qualification rate is 98.5%; when the material loss coefficient is 0.09, the product qualification rate drops to 97%; and when the material loss coefficient reaches 0.12, the product qualification rate further drops to 95%. A continuous mapping curve is fitted to these discrete data points, which describes the quantitative relationship between material loss and product quality.

[0059] Based on the mapping curves, the correlation factors for material loss were determined. The analysis revealed that the correlation between injection molding temperature fluctuation and the material loss coefficient was 0.85, pressure fluctuation was 0.78, and raw material moisture content was 0.65. Therefore, temperature fluctuation was identified as the primary correlation factor for material loss. Further matching analysis was performed on these correlation factors with the material transport state sequence to calculate the influence weight of each factor under different states. The results showed that the influence weight of the temperature factor reached 0.62 in the fluctuation range, while it was only 0.38 in the stable range.

[0060] Based on the matching analysis results, the critical workstations for material loss were identified. On a production line containing injection molding, assembly, and testing workstations, the matching degree of the material loss correlation factor for the injection molding workstation was 0.82, for the assembly workstation it was 0.45, and for the testing workstation it was 0.23. Therefore, the injection molding workstation was identified as a critical workstation for material loss.

[0061] Material consumption prediction parameters are calculated based on the process parameters and mapping curves of key material loss stations. For identified injection molding stations, historical production data is analyzed to extract the variation patterns of process parameters such as temperature, pressure, and time. For example, in continuous production, the injection molding machine temperature rises by approximately 5°C every 4 hours of operation, resulting in an increase of 0.01 in the material loss coefficient. By combining the variation rate of process parameters with the mapping curve, material consumption prediction parameters can be obtained, and a prediction model can be further constructed based on these parameters.

[0062] Based on the coupling relationship between material consumption forecast parameters and material loss trend characteristics, predicted material consumption values ​​are generated. Taking a 100-hour continuous production plan as an example, the predicted total material input demand is 12,500 kg, the estimated material loss is 1,050 kg, the average material loss rate is 8.4%, and the expected product qualification rate is 97.8%. Time-segmented forecast results are also provided, such as a material loss rate of 7.8% for the first 20 hours, 8.2% for the middle 60 hours, and 9.5% for the last 20 hours, with corresponding product qualification rates of 98.2%, 97.9%, and 97.2%, respectively.

[0063] The forecast results can guide production plan adjustments and material preparation. For example, for periods with predicted high loss rates, equipment maintenance or process parameter adjustments can be scheduled in advance. Practical application verification shows that the error between the predicted material consumption and the actual consumption using this method is controlled within 3%, providing a reliable basis for production management.

[0064] In one optional implementation, the timing characteristics of material delivery requests for each workstation are determined based on the material shortage risk index, and the material delivery requests for each workstation are prioritized based on the timing characteristics to generate a material delivery task queue, including: Based on the material shortage risk index, the frequency and duration of material delivery requests at each workstation are statistically analyzed to generate the temporal characteristics of material delivery requests at each workstation. The correlation coefficient between workstation capacity utilization and material consumption is determined based on the material shortage risk index, and material demand forecasting parameters are calculated based on the correlation coefficient and the time series characteristics. Based on the material demand forecasting parameters, the priority value of material delivery requests for each workstation is determined. Then, based on these priority values, the spatiotemporal overlap matrix of material delivery to each workstation and the material transport path topology between adjacent workstations are determined. The material delivery association strength is calculated based on the material transport path topology and the temporal characteristics to obtain the material delivery batch division parameters. Based on the material delivery batch division parameters and the spatiotemporal overlap matrix, material delivery requests are prioritized and sorted to generate a material delivery task queue.

[0065] In this implementation, the material shortage risk index data is first obtained, which reflects the likelihood of material shortages at each workstation. The material shortage risk index is usually calculated by combining factors such as material inventory, consumption rate, and replenishment cycle. For example, the material shortage risk index for engine mounting components at workstation A on a car assembly line is 0.78 (in the range of 0-1, with the risk increasing as it approaches 1).

[0066] When determining the temporal characteristics of material delivery requests for each workstation based on the material shortage risk index, the material request situation of each workstation is continuously monitored for 30 working days, recording the time, frequency, and duration of each request. Specifically, for workstations with a risk index higher than 0.6, data is collected hourly; for workstations with a risk index between 0.3 and 0.6, data is collected every 2 hours; and for workstations with a risk index lower than 0.3, data is collected every 4 hours. Through data collection, it was found that material requests for workstation B occurred 12 times within a week, with an average duration of 25 minutes each time, and were mostly concentrated in the two time periods of the morning shift (9-11 am) and afternoon shift (2-4 pm). These data constitute the temporal characteristics of material delivery requests for workstation B.

[0067] Next, the correlation coefficient between workstation capacity utilization and material consumption is determined. This is achieved by calculating the capacity utilization rate by analyzing the ratio of actual output to theoretical capacity at each workstation, and then performing a regression analysis between this ratio and actual material consumption to obtain the correlation coefficient. For example, workstation C has a capacity utilization rate of 85% and a material consumption of 30 units per hour. The calculated correlation coefficient is 0.92, indicating a high correlation between the workstation's capacity and material consumption.

[0068] By combining time-series characteristics and correlation coefficients, material demand forecasting parameters are calculated. These parameters include the probability of each workstation issuing a material request and the required quantity of materials within the next 4 hours. The calculation takes into account factors such as historical request frequency, duration, correlation coefficients, and current material inventory status. For example, the material demand forecasting parameters for workstation D show a 78% probability of issuing a material request within the next 2 hours, with an estimated need for 45 units of parts.

[0069] Based on material demand forecasting parameters, priority values ​​are assigned to material delivery requests at each workstation. Priority calculations consider factors such as the material shortage risk index, the predicted request probability, and the material's importance. For example, workstation E on the critical path has a material delivery request priority value of 92 (out of 100), while workstation F on the non-critical path has only 65.

[0070] Subsequently, a spatiotemporal overlap matrix for material delivery between workstations was determined. This matrix describes the temporal and spatial intersections of material delivery between different workstations. Larger matrix element values ​​indicate a higher degree of spatiotemporal overlap in material delivery between two workstations, and a greater likelihood of potential conflicts. In practical applications, the spatiotemporal overlap of material delivery between workstations G and H is 0.85, indicating that conflicts are highly likely to occur during their material delivery.

[0071] Simultaneously, the factory layout is analyzed to determine the topology of material transport paths between adjacent workstations. This structure is represented in the form of a graph, where nodes represent workstations, edges represent feasible material transport paths, and the weight of the edges reflects the path distance or traversal difficulty. For example, the material transport distance from workstation I to workstation J is 25 meters, the path smoothness is 0.9 (out of 1), and the overall evaluation weight of this path is 28.

[0072] Based on the topology and temporal characteristics of the material transport path, the material delivery association strength is calculated. This index reflects the feasibility of merging material delivery tasks from different workstations. The association strength is determined by factors such as path overlap, similarity of delivery times, and similarity of material types. The material delivery association strength for workstations K and L is 0.76, indicating that their material delivery tasks are suitable for merging. Through cluster analysis, parameters for dividing material delivery batches are obtained, such as dividing all delivery requests into 5 batches, each batch containing delivery tasks from 3-5 workstations.

[0073] Finally, material delivery requests are prioritized based on batch division parameters and a spatiotemporal overlap matrix. The prioritization algorithm comprehensively considers factors such as request priority values, spatial distribution of workstations within a batch, and material urgency. An example of the material delivery task queue output by the algorithm is as follows: Batch 1 (workstations E, G, I) priority 90; Batch 2 (workstations D, K, L) priority 85; Batch 3 (workstations B, C, J) priority 78. Logistics personnel or automated handling robots will execute material delivery tasks according to this queue order.

[0074] In a real-world application case, after applying this method to an electronics assembly line, material delivery efficiency improved by 32%, waiting time for materials at workstations decreased by 47%, and overall production line efficiency improved by 15%. This method is particularly suitable for multi-variety, small-batch production environments, effectively addressing the challenges of frequent changes in material demand and limited delivery resources, and achieving intelligent scheduling and optimization of material delivery.

[0075] In one optional implementation, material delivery batches are generated based on the distance information between adjacent workstations in the material delivery task queue, and the shortest delivery path is planned for each material delivery batch based on the real-time location information of the material carts, including: Based on the material delivery start position information and material delivery target position information in the material delivery task queue, calculate the material delivery distance coefficient between adjacent workstations; Based on the material delivery distance coefficient, the material delivery connection strength between adjacent workstations is extracted, and a material delivery association sequence is generated. Based on the overlap between the material delivery association sequence and the material delivery time information in the material delivery task queue, multiple material delivery batches are obtained; Based on the multiple material delivery batches, extract the material vehicle running trajectory features from the real-time location information of the material vehicle, combine the material delivery association sequence to determine the material delivery spatial constraint boundary, and generate a multi-batch material delivery path planning scheme. Based on the multi-batch material delivery route planning scheme, analyze the degree of deviation between the material vehicle running trajectory and the material delivery space constraint boundary, and determine the material delivery route optimization parameters; The material delivery route optimization parameters are used to adjust the multi-batch material delivery route planning scheme to generate the shortest material delivery route.

[0076] This implementation method calculates the distance coefficient between adjacent workstations, extracts the material delivery connection strength, divides material delivery batches, generates a path planning scheme, determines path optimization parameters, and finally generates the shortest delivery path.

[0077] Material delivery task queues typically contain multiple delivery records, each including material number, delivery start location, destination location, and delivery time. The material delivery distance coefficient between adjacent workstations is calculated based on this information. In practice, the coordinate information of each workstation in the delivery task queue is first extracted and represented in a Cartesian coordinate system. For example, workstation A is located at (10,15), and workstation B is located at (25,30). An initial distance value is obtained by calculating the straight-line distance between the two points, and then corrected based on the actual aisle conditions of the factory layout. If there are obstacles within the factory, detour distances are considered. For example, if there is equipment obstructing the path between workstation A and workstation B, the actual path requires a detour, increasing the distance to 30 meters, instead of the non-straight-line distance of approximately 21.2 meters. All distance data between adjacent workstations is stored as a distance matrix, serving as the basis for subsequent calculations.

[0078] Based on the calculated material delivery distance coefficient, the material delivery connection strength between adjacent workstations is further extracted. Connection strength reflects the frequency of material flow between workstations and is an important reference for planning delivery routes. Specifically, the number of material deliveries from workstation A to workstation B within a certain time period (e.g., a workday) is statistically analyzed. For example, if there are 12 deliveries from workstation A to workstation B and 5 deliveries from workstation B to workstation C in a day, the connection strength from A to B is higher than that from B to C. The influence of material weight and volume on connection strength is also considered. Materials with greater weight or volume will have a correspondingly higher delivery connection strength. Through comprehensive analysis, a material delivery association sequence is generated, which arranges all workstation pairs in descending order of connection strength.

[0079] Based on the time information in the material delivery sequence and the delivery task queue, material delivery batches are divided, taking into account both time constraints and delivery efficiency. Initially, a preliminary division is made according to time windows, such as grouping delivery tasks from 8:00 to 10:00 into one batch. Then, the overlap of material delivery sequences within each time window is checked. Tasks with high overlap are prioritized for the same batch to improve delivery efficiency. If the total amount of materials in a batch exceeds the material cart's load capacity, the batch will be further split. For example, 20 pieces of materials originally planned for delivery in the morning batch, exceeding the material cart's 300 kg load limit due to a total weight of 500 kg, will be split into two batches of 12 and 8 pieces respectively, based on delivery connection strength, ensuring that the weight of each batch does not exceed the load limit.

[0080] Based on the predefined material delivery batches, the trajectory characteristics of the material vehicles are extracted from their real-time location information. The real-time coordinates of the material vehicles are obtained through onboard GPS or indoor positioning systems. Historical trajectory data is analyzed to extract information such as common routes, average speed, and turning characteristics. For example, the analysis reveals that the average speed of the material vehicle on a certain route is 2 m / s, while the speed drops to 0.5 m / s at corners. Combined with the material delivery sequence, spatial constraints on material delivery are determined, such as workshop layout limitations and safety passage requirements. Based on this information, a path planning scheme is generated for each delivery batch. Taking a delivery batch with four workstations (workstations A, B, C, and D) as an example, the following path is generated: starting from the current location of the material vehicle (e.g., the warehouse, coordinates (5,5)), it passes through workstations A (10,15), C (30,15), D (35,25), and B (25,30) in sequence, finally returning to the warehouse. This path comprehensively considers distance, material vehicle characteristics, and delivery connection strength.

[0081] Further analysis of the deviation between the material cart's trajectory and the spatial constraints of material delivery is conducted to determine the optimization parameters for the material delivery path. In practice, the differences between historical trajectories and planned paths are compared to calculate deviation distances and frequencies. For example, it was found that when a material cart passes through a narrow passage, the actual trajectory deviates from the planned path by an average of 0.5 meters, and in 90% of cases, it chooses to detour. Time factors are also considered; for instance, near workstation C, frequent personnel activity often requires the material cart to slow down or wait, resulting in an actual transit time that is 2 minutes longer than expected. Based on these analyses, path optimization parameters are generated, including path adjustment coefficients and time correction values.

[0082] Finally, the material delivery route planning scheme for multiple batches is adjusted based on the material delivery route optimization parameters to generate the shortest material delivery route. The adjustment process considers various factors in actual operation, such as congestion in certain areas and the battery status of the material carts. Taking the above 4-station delivery batch as an example, the original route can be adjusted as follows: starting from the warehouse (5,5), first to station A (10,15), then to station B (25,30) (instead of station C), then to station D (35,25), and finally to station C (30,15) before returning to the warehouse. The total length of the adjusted route is 95 meters, saving approximately 7% of the distance compared to the original route of 102 meters. The optimized route is then sent to the material carts or operators for execution, and actual operating data is continuously collected during the execution process to provide a more accurate reference for subsequent route planning.

[0083] In one optional implementation, based on the multiple material delivery batches, the trajectory features of the material vehicles in the real-time location information are extracted, and the spatial constraint boundaries of material delivery are determined in conjunction with the material delivery association sequence to generate a multi-batch material delivery path planning scheme, including: Collect real-time location information of the material cart, and extract the change characteristics of the material cart's running acceleration and running speed based on the real-time location information to obtain the material cart's running trajectory characteristics; Based on the characteristics of the material car's running trajectory, the time of the running state transition during the material car's operation is determined, and a material delivery association sequence is established according to the running state transition time and the material demand of the workstation; Based on the material delivery association sequence, extract the time window constraints and spatial area constraints of workstation material delivery to obtain the material delivery spatial constraint boundary. Based on the material delivery space constraint boundary, the mapping relationship between the material car running trajectory segment and the workstation layout topology is determined, and the location information of key points of the material car running trajectory is obtained. Based on the key point location information of the material vehicle's running trajectory and the time of the running state transition, a segmented planning criterion for material delivery path is generated. Based on the segmented planning criterion for material delivery path, a batch path planning is performed for the multiple material delivery batches to generate a multi-batch material delivery path planning scheme.

[0084] In the specific implementation process, the real-time location information of the material cart is first collected. This information is usually obtained through positioning devices installed on the material cart, such as GPS or an indoor positioning system. The positioning data includes the position coordinates (x, y) of the material cart at different times, collected at a frequency of 10 times per second. Based on this position data, the position change of the material cart at consecutive time points is calculated, thereby obtaining the running speed of the material cart. For example, if the material cart's position is (10, 15) at time t1 and (11, 16) at time t2, the average speed during that time period can be obtained by calculating the distance between the two points and dividing by the time difference. Further, the acceleration information of the material cart is obtained by calculating the rate of change of speed. By analyzing the speed and acceleration change patterns over a period of time, the characteristics of the material cart's running trajectory are extracted, such as identifying typical motion states of the material cart, such as starting, constant speed travel, deceleration, and stopping.

[0085] After acquiring trajectory features, the state transition times during the material cart's operation are determined. Specifically, when the material cart's speed rises from zero to a preset threshold (e.g., 0.5 m / s), it is determined to be in the starting state; when the speed fluctuates within a certain range (e.g., fluctuation not exceeding 0.2 m / s) and lasts for a certain period (e.g., 5 seconds), it is determined to be in the uniform speed state; when the speed starts to decrease continuously from a stable value and the acceleration is negative, it is determined to be in the deceleration state; when the speed drops to near zero and remains stationary (e.g., speed below 0.1 m / s for 3 seconds), it is determined to be in the stopping state. The times of these state transitions are recorded, such as the material cart changing from a uniform speed state to a deceleration state at time t=120 seconds and coming to a complete stop at t=128 seconds. Simultaneously, material demand information for each workstation is acquired, including workstation identifier, required material type, quantity, and expected delivery time window. By analyzing the correspondence between the material cart's stopping position and the workstation position, and the matching degree between the stopping time and the workstation's material demand time, a material delivery association sequence is established, recording the workstation order in which the material cart stops and the corresponding delivery time.

[0086] Based on the established material delivery sequence, the time window constraints and spatial area constraints of material delivery at each workstation are further extracted. The time window constraints include the earliest and latest material arrival times for each workstation. For example, workstation A requires materials to arrive no earlier than 8:00 AM and no later than 8:30 AM. The spatial area constraints describe the area around the workstation that can be used for material cart parking and operation. For example, the material unloading point coordinates for workstation B are (50, 75), and the effective operating area is within a 2-meter radius. By comprehensively analyzing these constraints of multiple workstations, the overall spatial constraint boundary of material delivery is determined, including the drivable road network, parking areas, and obstacle areas that need to be avoided.

[0087] After defining the spatial constraints of material delivery, the material cart's trajectory is mapped to the workstation layout topology. Specifically, key points in the material cart's actual trajectory are identified, including turning points, bifurcation points, and stopping points. For example, a turning point is marked when the material cart's trajectory changes direction by more than 45 degrees; a workstation stopping point is marked when the material cart stays at a specific location for more than 10 seconds and the distance between that location and the workstation is less than 3 meters. These key points are then matched with nodes in the factory layout diagram to establish a correspondence between the actual trajectory and the workstation topology. For example, key point (32, 48) corresponds to intersection 3 in the factory layout, and key point (45, 60) corresponds to the material receiving point at workstation C. This mapping relationship allows us to understand the material cart's movement patterns in the actual environment.

[0088] Finally, based on the key point location information of the material cart's running trajectory and the time of running state transition, segmented planning criteria for material delivery routes are generated. These criteria consider the traffic characteristics of different road segments, such as average driving speed, congestion level, and turning difficulty. For example, for the frequently congested road segment AB, the average travel time for this segment is recorded as 45 seconds, with a standard deviation of 8 seconds; for the narrow passage CD requiring slowing down, the maximum speed limit is set to 0.8 meters per second. Based on these criteria, batch path planning is performed for multiple material delivery batches. For the first batch of delivery tasks (material delivery to workstations D, E, and F), the optimal path is calculated as start point - ACDEF - end point, with an estimated completion time of 25 minutes; for the second batch of delivery tasks (material delivery to workstations G, H, and I), the optimal path is start point - BGHI - end point, with an estimated completion time of 30 minutes. Combining these batch planning results, a complete multi-batch material delivery route planning scheme is generated, including detailed information such as the workstation access order, estimated arrival time, path length, and risk point warnings for each batch.

[0089] Through the above-mentioned technologies, this invention can generate an efficient and feasible material delivery route planning scheme based on the actual operation data of the material cart, combined with the material demand and spatial constraints of the workstation. This effectively improves the material delivery efficiency of the factory, reduces the risk of delivery delays, and has important application value for the intelligent logistics system of modern manufacturing enterprises.

[0090] The intelligent material distribution and scheduling optimization system for the metal tile production line of this invention includes: The first unit is used to acquire material demand information of each workstation on the metal tile production line and real-time production status information of the metal tile production line, and generate material delivery early warning information based on the material demand information and the real-time production status information. The second unit is used to predict the material consumption forecast value based on the material delivery early warning information and the material consumption fluctuation data of each workstation, and to determine the material shortage risk index of each workstation based on the material consumption forecast value. The third unit is used to determine the temporal characteristics of material delivery requests for each workstation based on the material shortage risk index, prioritize the material delivery requests for each workstation based on the temporal characteristics, and generate a material delivery task queue. The fourth unit is used to generate material delivery batches based on the distance information between adjacent workstations in the material delivery task queue, and to plan the shortest delivery path for each material delivery batch based on the real-time location information of the material cart. The fifth unit is used to generate a material delivery instruction based on the material delivery batch and the shortest delivery route, and send the material delivery instruction to the corresponding delivery vehicle to instruct the delivery vehicle to make delivery.

[0091] A third aspect of the present invention provides an electronic device, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0092] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0093] This invention can be a method, apparatus, system, and / or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for performing various aspects of the invention.

[0094] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent material distribution and scheduling optimization method for metal tile production lines, characterized in that, include: Obtain material demand information for each workstation on the metal tile production line and real-time production status information of the metal tile production line; generate material delivery early warning information based on the material demand information and the real-time production status information. Based on the material delivery early warning information and the material consumption fluctuation data of each workstation, the predicted material consumption value is obtained, and the material shortage risk index of each workstation is determined based on the predicted material consumption value. The timing characteristics of material delivery requests at each workstation are determined based on the material shortage risk index. The material delivery requests at each workstation are prioritized based on the timing characteristics to generate a material delivery task queue. Based on the distance information between adjacent workstations in the material delivery task queue, material delivery batches are generated, and the shortest delivery path is planned for each material delivery batch based on the real-time location information of the material cart. Based on the material delivery batch and the shortest delivery route, a material delivery instruction is generated and sent to the corresponding delivery vehicle to instruct the delivery vehicle to carry out delivery.

2. The method according to claim 1, characterized in that, Obtain material demand information for each workstation on the metal tile production line and real-time production status information of the metal tile production line; generate material delivery early warning information based on the material demand information and the real-time production status information, including: Based on the material demand information and the real-time production status information, the material consumption fluctuation characteristics are determined by analyzing the phase differences in the material transfer process between workstations. The material consumption fluctuation characteristics are then matched with the workstation production cycle time to obtain the dynamic correspondence between the production cycle time of each workstation and the material consumption. Based on the dynamic correspondence, analyze the material complementary configuration scheme between each workstation, and establish the correlation data between material consumption and production status based on the material complementary configuration scheme; Establish a cascading influence relationship of material consumption between workstations based on the associated data, determine the material consumption mutation point of the target workstation based on the cascading influence relationship, and take the material consumption mutation point as the critical state time point of material consumption. The material consumption critical state time point is compared with the material demand time to generate initial material delivery warning information. The initial material delivery warning information is then calibrated based on the real-time production status information to obtain calibrated material delivery warning information.

3. The method according to claim 1, characterized in that, Based on the material delivery early warning information and the material consumption fluctuation data of each workstation, a predicted material consumption value is obtained. Based on the predicted material consumption value, a material shortage risk index for each workstation is determined, including: Based on the material delivery early warning information, the vibration frequency and material conveying speed of the material conveying line at each workstation are collected. Based on the correspondence between the vibration frequency and the material conveying speed, the stable range of material transmission is determined, and material consumption fluctuation data of each workstation is generated. Based on the material consumption fluctuation data and the material transmission stability range, the temporal variation characteristics of material consumption at each workstation are analyzed to obtain the material consumption evaluation benchmark. Based on the material consumption assessment benchmark, the material loss coefficient in the workstation material conversion process is determined, and based on the material loss coefficient, the corresponding curve of material consumption and product qualification rate is determined to generate the predicted value of material consumption. The predicted material consumption value is compared with the stable material transmission range to determine the fluctuation nodes in the material transmission process. Based on the deviation between the material backlog status of the fluctuation nodes and the predicted material consumption value, an initial material shortage risk index is generated. Based on the correlation between the initial material shortage risk index and the material backlog status, the material shortage risk index for each workstation is determined.

4. The method according to claim 3, characterized in that, Based on the material consumption assessment benchmark, the material loss coefficient during the material conversion process at the workstation is determined. Based on the material loss coefficient, a curve showing the correlation between material consumption and product qualification rate is determined, and a predicted material consumption value is generated, including: The material consumption fluctuation cycle is obtained by calculating the real-time conversion ratio of material input parameters and output parameters at each workstation based on the material consumption assessment benchmark. Based on the material consumption fluctuation cycle, the stable range and fluctuation range of the material transportation process at each workstation are analyzed to obtain the material transportation state sequence; A workstation material loss coefficient is established based on the material transport state sequence and the material consumption fluctuation cycle, and the material loss trend characteristics are determined based on the workstation material loss coefficient and the workstation processing parameters. Based on the material loss trend characteristics, a mapping curve between material consumption and product qualification rate is generated, and the material loss correlation factor is determined. Based on the degree of matching between the material loss correlation factor and the material transportation state sequence, the key workstations for material loss are determined. Based on the process parameters of the key material loss stations and the mapping curve, material consumption prediction parameters are calculated, and material consumption prediction values ​​are generated according to the coupling relationship between the material consumption prediction parameters and the material loss trend characteristics.

5. The method according to claim 1, characterized in that, Based on the material shortage risk index, the temporal characteristics of material delivery requests for each workstation are determined. Based on these temporal characteristics, the material delivery requests for each workstation are prioritized, and a material delivery task queue is generated, including: Based on the material shortage risk index, the frequency and duration of material delivery requests at each workstation are statistically analyzed to generate the temporal characteristics of material delivery requests at each workstation. The correlation coefficient between workstation capacity utilization and material consumption is determined based on the material shortage risk index, and material demand forecasting parameters are calculated based on the correlation coefficient and the time series characteristics. Based on the material demand forecasting parameters, the priority value of material delivery requests for each workstation is determined. Then, based on these priority values, the spatiotemporal overlap matrix of material delivery to each workstation and the material transport path topology between adjacent workstations are determined. The material delivery association strength is calculated based on the material transport path topology and the temporal characteristics to obtain the material delivery batch division parameters. Based on the material delivery batch division parameters and the spatiotemporal overlap matrix, material delivery requests are prioritized and sorted to generate a material delivery task queue.

6. The method according to claim 1, characterized in that, Based on the distance information between adjacent workstations in the material delivery task queue, material delivery batches are generated. The shortest delivery path is planned for each material delivery batch based on the real-time location information of the material carts, including: Based on the material delivery start position information and material delivery target position information in the material delivery task queue, calculate the material delivery distance coefficient between adjacent workstations; Based on the material delivery distance coefficient, the material delivery connection strength between adjacent workstations is extracted, and a material delivery association sequence is generated. Based on the overlap between the material delivery association sequence and the material delivery time information in the material delivery task queue, multiple material delivery batches are obtained; Based on the multiple material delivery batches, extract the material vehicle running trajectory features from the real-time location information of the material vehicle, combine the material delivery association sequence to determine the material delivery spatial constraint boundary, and generate a multi-batch material delivery path planning scheme. Based on the multi-batch material delivery route planning scheme, analyze the degree of deviation between the material vehicle running trajectory and the material delivery space constraint boundary, and determine the material delivery route optimization parameters; The material delivery route optimization parameters are used to adjust the multi-batch material delivery route planning scheme to generate the shortest material delivery route.

7. The method according to claim 6, characterized in that, Based on the multiple material delivery batches, the trajectory features of the material vehicles are extracted from the real-time location information of the material vehicles. Combined with the material delivery association sequence, the spatial constraint boundaries of the material delivery are determined, and a multi-batch material delivery path planning scheme is generated, including: Collect real-time location information of the material cart, and extract the change characteristics of the material cart's running acceleration and running speed based on the real-time location information to obtain the material cart's running trajectory characteristics; Based on the characteristics of the material car's running trajectory, the time of the running state transition during the material car's operation is determined, and a material delivery association sequence is established according to the running state transition time and the material demand of the workstation; Based on the material delivery association sequence, extract the time window constraints and spatial area constraints of workstation material delivery to obtain the material delivery spatial constraint boundary. Based on the material delivery space constraint boundary, the mapping relationship between the material car running trajectory segment and the workstation layout topology is determined, and the location information of key points of the material car running trajectory is obtained. Based on the key point location information of the material vehicle's running trajectory and the time of the running state transition, a segmented planning criterion for material delivery path is generated. Based on the segmented planning criterion for material delivery path, a batch path planning is performed for the multiple material delivery batches to generate a multi-batch material delivery path planning scheme.

8. An intelligent material distribution and scheduling optimization system for a metal tile production line, used to implement the method as described in any one of claims 1-7, characterized in that, include: The first unit is used to acquire material demand information of each workstation on the metal tile production line and real-time production status information of the metal tile production line, and generate material delivery early warning information based on the material demand information and the real-time production status information. The second unit is used to predict the material consumption forecast value based on the material delivery early warning information and the material consumption fluctuation data of each workstation, and to determine the material shortage risk index of each workstation based on the material consumption forecast value. The third unit is used to determine the temporal characteristics of material delivery requests for each workstation based on the material shortage risk index, prioritize the material delivery requests for each workstation based on the temporal characteristics, and generate a material delivery task queue. The fourth unit is used to generate material delivery batches based on the distance information between adjacent workstations in the material delivery task queue, and to plan the shortest delivery path for each material delivery batch based on the real-time location information of the material cart. The fifth unit is used to generate a material delivery instruction based on the material delivery batch and the shortest delivery route, and send the material delivery instruction to the corresponding delivery vehicle to instruct the delivery vehicle to make delivery.

9. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 7.

Images