Warehouse Management Forecasting Methods and Devices

Abstract

This application provides a warehouse management forecasting method and apparatus, belonging to the field of warehouse management technology. The method includes: acquiring multi-source heterogeneous data from multiple containers; preprocessing the multi-source heterogeneous data to obtain standard input data; the multi-source heterogeneous data includes container model and size data, material consumption data, production scheduling data, equipment operation data, logistics status data, and warehouse inventory data; determining the production stage type of each container based on the standard input data; assigning production weight coefficients to each container according to its production stage type to obtain a production weight coefficient matrix corresponding to multiple containers; and performing time-series forecasting based on the standard input data and the production weight coefficient matrix, whether the production stage type is manual, automatic, or intelligent, to obtain warehouse management forecasting data. This application can achieve intelligent warehouse management with precise matching of warehousing and production cycles.

Description

Technical Field

[0001] This application belongs to the field of warehouse management technology, and more specifically, relates to a warehouse management forecasting method and apparatus. Background Technology

[0002] Warehouse management is a core control link in the production and circulation of automotive parts containers. Containers, as specialized assembly tools used for carrying, transferring, and storing parts during automotive manufacturing, are highly correlated with their model specifications, storage quantity, inbound / outbound cycle time, production progress, material consumption, and logistics scheduling. With the continuous shortening of new car product iteration cycles, container production is characterized by multiple models, small batches, and rapid changeovers, placing higher demands on warehouse turnover efficiency, precise material supply, and real-time logistics scheduling.

[0003] Existing warehouse management methods can only record and query static information such as inventory quantity and storage space occupancy. They cannot dynamically predict material demand, inbound and outbound timing, and storage space occupancy. This can easily lead to problems such as material supply delays or inventory backlog, mismatch between logistics scheduling and production completion timing, and unreasonable storage space allocation. As a result, warehouse turnover efficiency is low, on-site waiting time is long, and production continuity is difficult to guarantee. They cannot meet the needs of efficient and intelligent container production and warehouse management. Summary of the Invention

[0004] To address the aforementioned technical problems, this application provides a warehouse management prediction method and apparatus to solve the problem that existing container warehouse management cannot dynamically predict in conjunction with the production stage, resulting in a disconnect between warehousing and production, thereby achieving intelligent warehouse management with precise matching between warehousing and production rhythms.

[0005] The embodiments of this application disclose the following technical solutions: Firstly, a warehouse management forecasting method is provided, including: Acquire multi-source heterogeneous data from multiple containers, preprocess the multi-source heterogeneous data to obtain standard input data; the multi-source heterogeneous data includes container model and size data, material unit consumption data, production scheduling data, equipment operation data, logistics status data and warehouse inventory data; Based on standard input data, the production stage type of each container is determined, and production weight coefficients are assigned to each container according to its production stage type, resulting in a production weight coefficient matrix for multiple containers; the production stage type is manual production stage, automatic production stage, or intelligent production stage. Time-series forecasting is performed based on standard input data and production weight coefficient matrix to obtain warehouse management forecast data.

[0006] Secondly, a warehouse management forecasting device is provided, comprising: The multi-source data acquisition module is used to acquire multi-source heterogeneous data from multiple containers, preprocess the multi-source heterogeneous data to obtain standard input data; the multi-source heterogeneous data includes container model and size data, material unit consumption data, production scheduling data, equipment operation data, logistics status data and warehouse inventory data; The production stage weight determination module is used to determine the production stage type of each container based on standard input data, and to assign production weight coefficients to each container according to the production stage type, thereby obtaining a production weight coefficient matrix for multiple containers; the production stage type is manual production stage, automatic production stage, or intelligent production stage. The warehouse management forecasting module is used to perform time-series forecasting based on standard input data and the production weight coefficient matrix to obtain warehouse management forecasting data.

[0007] Thirdly, embodiments of this application also provide an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the warehouse management prediction method provided by any possible implementation of the first aspect.

[0008] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the warehouse management prediction method provided by any possible implementation of the first aspect.

[0009] The beneficial effects of the technical solution provided in this application are as follows: The warehouse management forecasting method and apparatus provided in this application, compared with related technologies, are as follows: This application embodiment collects multi-source heterogeneous data from the entire container production process and preprocesses it into standard input data. This achieves the normalization and integration of production, material, equipment, logistics, and warehousing information, breaking down information silos and providing a data foundation for accurate prediction. By identifying the manual, automatic, and intelligent production stages of containers and constructing corresponding production weight coefficient matrices, this application embodiment can objectively reflect the differentiated impacts of different production modes on material consumption, production cycle time, and warehousing turnover, ensuring that the prediction results align with the actual production pace.

[0010] This application's embodiments utilize standard data and a weight matrix to perform time-series forecasting, dynamically determining material demand, inbound / outbound timing, and storage location allocation schemes. This transforms warehouse management from passive recording to proactive prediction. Compared to traditional methods, this application's embodiments effectively avoid material delays or stockpiling, misaligned logistics scheduling, and unreasonable storage location utilization. It shortens on-site waiting time, ensures production continuity, improves warehouse turnover efficiency and overall control accuracy, and is well-suited to the automotive industry's needs for multi-model, small-batch, and quick-change container production and intelligent warehouse management. It possesses strong practicality and promotional value. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 A flowchart illustrating the warehouse management forecasting method provided in this application embodiment; Figure 2 A structural block diagram of the warehouse management forecasting device provided in the embodiments of this application; Figure 3 A schematic block diagram of a computer device provided in an embodiment of this application; Figure 4 Another schematic block diagram of the computer device provided in the embodiments of this application. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0014] It should be noted that the terms "first," "second," and similar terms used in this application do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Unless the context clearly indicates otherwise, the singular forms "a," "one," or "the," etc., do not indicate a quantity limitation, but rather indicate the presence of at least one. The quantities of "multiple" or "multiple copies" mentioned in the embodiments of this application all refer to a quantity of "at least two," for example, "multiple" means "at least two," and "multiple copies" means "at least two copies." The terms "comprising" and "having," and any variations thereof, as used in this application, are intended to cover non-exclusive inclusion. The term "and / or" as used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0015] like Figure 1 As shown in the embodiments of this application, the warehouse management forecasting method can be executed by computer equipment in each step. Computer equipment refers to electronic equipment with data computing, processing, and storage capabilities. The method may include: S101: Acquire multi-source heterogeneous data from multiple containers, preprocess the multi-source heterogeneous data to obtain standard input data; the multi-source heterogeneous data includes container model and size data, material unit consumption data, production scheduling data, equipment operation data, logistics status data and warehouse inventory data.

[0016] In this embodiment, multi-source heterogeneous data is preprocessed to obtain standard input data, including: Missing values ​​are filled in multi-source heterogeneous data to obtain data with missing values ​​completed. The missing value-completed data is then subjected to outlier removal to obtain outlier-filtered data. The units of the outlier filtering data are standardized to obtain data with standardized units. The unified data of each unit is classified and labeled according to the container model, material type and production stage to obtain classified label data; The categorical label data is time-aligned to obtain standard input data.

[0017] In this embodiment, multi-source heterogeneous data refers to business data from different systems and formats throughout the entire container production process, including six types of core data such as container model and size data. Container model and size data refers to basic data characterizing container specifications, such as container length, width, height, and matching rectangular tube specifications. Material consumption data refers to the material quota consumption data for producing a single container, such as the amount of square tube required for a single container. Production scheduling data refers to the planned time sequence data for container production, such as daily planned output and process time data. Equipment operation data refers to the real-time operating status data of production workstations, such as equipment start / stop and process progress data. Logistics status data refers to the real-time status data of transfer equipment, such as AGV location and task progress data.

[0018] Warehouse inventory data refers to real-time warehouse storage data, such as inventory quantity and storage location occupancy data. Preprocessing refers to a series of operations to organize and optimize raw data, including five operations such as missing value imputation. Standard input data refers to preprocessed, formatted, and usable data used for subsequent forecasting calculations. Missing value imputation refers to the operation of filling in missing fields, such as using valid values ​​from the previous period. Outlier removal refers to the operation of filtering invalid or deviating data, such as removing data with fluctuations exceeding thresholds. Unit standardization refers to the operation of standardizing measurement standards, such as standardizing the length unit to millimeters. Classification and labeling refers to the operation of categorizing and labeling by dimension, such as labeling by container type. Time-series alignment refers to the operation of organizing data to the same time axis, such as aligning data according to fixed time slices.

[0019] For example, this embodiment first obtains multi-source heterogeneous data corresponding to multiple containers in real time through standardized interfaces of various business systems throughout the production process. Among them, container model and size data come from the basic parameter file output by the container design, material unit consumption data comes from the container production bill of materials file, production scheduling data comes from the planning file of the advanced production scheduling system, equipment operation data comes from the real-time data collected by the programmable logic controller of the intensive production workstation, logistics status data comes from the real-time reporting data of the vehicle terminal of the in-plant transfer equipment, and warehouse inventory data comes from the real-time ledger data of the warehouse management system.

[0020] In this embodiment, the acquired multi-source heterogeneous data is first filled with missing values. For field gaps caused by device offline or network interruption, the corresponding valid value of the same model container in the previous valid period of the data item is used to fill the gaps. At the same time, a temporary status mark is added to the filled data. For scheduling data gaps caused by plan adjustments, the latest effective production plan data is used to cover and fill the gaps. After the processing is completed, the missing value filled data is obtained.

[0021] This embodiment performs outlier removal processing on the missing value completion data, with a preset data fluctuation threshold of 30%. When the value of a single data item deviates from the mean of the previous three periods in the same dimension by more than 30%, it is determined to be abnormal fluctuation data, and the abnormal data is replaced by the mean of the previous three periods. For signal drift data where the location data of the on-site transfer equipment changes by more than 5 meters, it is directly filtered out, and the outlier filtered data is obtained after processing.

[0022] This embodiment performs unit standardization processing on the outlier filtering data, converting the measurement units of all length data to millimeters, weight data to kilograms, time data to minutes, and quantity data to pieces, resulting in data with unified units.

[0023] This embodiment categorizes and labels the unit's unified data according to three dimensions: container model, material type, and production stage. Container model is labeled according to the standard model code output from the design; material type is labeled according to material categories such as rectangular tubes and cold-rolled steel sheets; and production stage is labeled according to manual production stage, automatic production stage, and intelligent production stage. The resulting categorized and labeled data is obtained after processing.

[0024] This embodiment performs time-series alignment processing on the classification and labeling data, setting a standard time slice of 10 minutes to normalize all dimensions of data onto the same time axis. Time slices without data are filled with default zero values ​​for the corresponding dimensions to ensure that the time-series nodes of all data are completely consistent, resulting in standard input data after processing.

[0025] This embodiment effectively eliminates noise and measurement discrepancies in the original data by performing a series of preprocessing operations on multi-source heterogeneous data, achieving normalized integration of data from the entire production process. This provides a stable and reliable standard data foundation for subsequent warehouse management forecasting, effectively ensuring the accuracy and practicality of the forecast results.

[0026] S102: Determine the production stage type of each container based on the standard input data, and assign production weight coefficients to each container according to the production stage type to obtain a production weight coefficient matrix corresponding to multiple containers; the production stage type is manual production stage, automatic production stage or intelligent production stage.

[0027] In this embodiment, the production stage type of each container is determined based on standard input data, including: Identify the process type of each container based on the equipment operation data and production scheduling data in the standard input data; When the process type is manual part removal, manual welding, manual repair welding or manual painting, the container is determined to be in the manual production stage; When the process type is laser cutting, automatic feeding, automatic welding, induction heating, correction or robot head sealing, the container is determined to be in the automatic production stage; When the process type is project design, review and scheduling, material procurement or electronic contract management, the container is determined to be in the intelligent production stage.

[0028] In this embodiment, production weight coefficients are assigned to each container according to its production stage type, resulting in a production weight coefficient matrix corresponding to multiple containers, including: Assign a first production weight coefficient to containers in the manual production stage, assign a second production weight coefficient to containers in the automatic production stage, and assign a third production weight coefficient to containers in the intelligent production stage; the second production weight coefficient is greater than the first production weight coefficient, and the first production weight coefficient is greater than the third production weight coefficient. The production weight coefficients corresponding to each container are arranged in a matrix according to the container number and production stage type to obtain a matrix of production weight coefficients corresponding to multiple containers.

[0029] In this embodiment, the production stage type refers to the classification of the production control stage throughout the entire lifecycle of the container, such as manual production stage, automatic production stage, and intelligent production stage. The production weight coefficient refers to a quantitative value characterizing the degree of impact of different production stages on warehouse management, such as a first production weight coefficient, a second production weight coefficient, and a third production weight coefficient. The production weight coefficient matrix refers to a set of weight coefficients arranged by container number and production stage type, used for unified prediction calculation of batch containers. The manual production stage refers to the production stage that relies on manual labor to complete core processes, such as stages corresponding to manual part handling and manual welding. The automatic production stage refers to the production stage that relies on automated equipment to complete core processes, such as stages corresponding to laser cutting and automatic feeding.

[0030] The intelligent manufacturing stage refers to the pre-production stage that relies on digital systems for project management, including stages such as project design, review, and scheduling. Equipment operation data refers to pre-processed real-time operating data of production equipment, such as the start / stop status of workstation equipment and process execution progress data. Production scheduling data refers to pre-processed time-series data of the container production plan, such as process planning nodes and production task allocation data. Process type refers to the subdivided operational stages in the entire container production process, such as specific processes like manual welding and robotic end capping. Container number refers to a unique code that identifies each container, used to distinguish different containers when arranged in a matrix.

[0031] For example, this embodiment first extracts equipment operation data and production scheduling data corresponding one-to-one with each container number from the pre-processed standard input data. The equipment operation data comes from real-time collection and pre-processed process execution data from the production station's programmable logic controller, equipment start / stop status data, and check-in and work confirmation data from manual operation terminals. The production scheduling data comes from pre-processed process plan node data, production task allocation data, and process stage identifier data issued by the advanced production scheduling system. This ensures that the extracted data has a unique binding relationship with each individual container, avoiding data matching errors.

[0032] This embodiment identifies the current process type of each container based on extracted equipment operation data and production scheduling data. Specifically, it matches the process code that the container is currently scheduled to execute in the production scheduling data, and combines the real-time operation status of the corresponding workstation in the equipment operation data to complete the accurate identification of the process type. The preset process identification verification rule is that when the matching degree between the planned process and the real-time operation status exceeds 90%, the process is confirmed as the current process type of the container. When the identified process type is manual part handling, manual welding, manual repair welding, or manual painting, and there is a corresponding workstation with a sign-in record of human operation and no status data showing that automated equipment has been running continuously for more than 30 minutes, the container is determined to be in the manual production stage. When the identified process type is laser cutting, automatic feeding, automatic welding, induction heating, calibration, or robotic end capping, and there is a corresponding workstation with data showing that automated equipment has been running continuously and there is no record of human continuous operation for more than 10 minutes, the container is determined to be in the automatic production stage. When the identified process type is project design, review scheduling, material procurement, or electronic contract management, and there is a corresponding process in the digital system approval process and no production workstation operation data, the container is determined to be in the intelligent production stage.

[0033] This embodiment assigns a corresponding production weight coefficient to each container based on the production stage type determined for each container. The preset weight allocation rule is as follows: containers in the manual production stage are assigned a first production weight coefficient, containers in the automatic production stage are assigned a second production weight coefficient, and containers in the intelligent production stage are assigned a third production weight coefficient. The second production weight coefficient is greater than the first production weight coefficient, and the first production weight coefficient is greater than the third production weight coefficient. Specifically, this embodiment sets the second production weight coefficient to 0.7, the first production weight coefficient to 0.25, and the third production weight coefficient to 0.05. These values ​​are determined based on the degree of impact of different production stages on warehouse material demand, logistics scheduling, and storage space occupancy. The automatic production stage is the core link in the physical production of containers and has the highest impact on warehouse management; therefore, it is assigned the highest weight. The intelligent production stage is a pre-project control stage and has the lowest direct impact on warehouse management; therefore, it is assigned the lowest weight.

[0034] In this embodiment, after all containers have had their production weight coefficients assigned, the production weight coefficients for each container are arranged in a matrix according to the container number and production stage type. Specifically, the unique container number is used as the matrix row identifier, and the manual production stage, automatic production stage, and intelligent production stage are used as the matrix column identifiers. The column position corresponding to the production stage of each container is filled with the assigned weight coefficient, and the remaining column positions are filled with zero values. This ultimately forms a multi-row, three-column, two-dimensional array that matches the number of containers, thus obtaining the production weight coefficient matrix for multiple containers. For example, when there are 5 containers, of which 1 is in the manual production stage, 3 are in the automatic production stage, and 1 is in the intelligent production stage, the corresponding rows of the matrix are filled with 0.25, 0.7, 0.7, 0.7, and 0.05 respectively, and the remaining positions are filled with zero values, completing the standardized construction of the matrix.

[0035] This embodiment accurately identifies the production stage of the containers based on standard input data, assigns differentiated production weight coefficients to different production stages, and constructs a corresponding production weight coefficient matrix. This objectively quantifies the differentiated impact of different production links on warehouse management, aligns with the progressive production mode of containers from manual to automatic to intelligent, provides accurate weight basis for subsequent warehouse management timing prediction, effectively improves the matching degree between prediction results and actual production rhythm, and adapts to the production characteristics of containers with multiple models and quick changeover.

[0036] S103: Based on standard input data and the production weight coefficient matrix, perform time series prediction to obtain warehouse management prediction data.

[0037] In this embodiment, the warehouse management forecast data includes container material demand forecast data; Time-series forecasting based on standard input data and a production weighting coefficient matrix yields warehouse management forecast data, including: Extract the planned production quantity of each container within a future preset time period from the production scheduling data in the standard input data; Based on the material unit consumption data and planned production quantity in the standard input data, the theoretical material requirement for each container is calculated. The theoretical material requirement for each container is weighted and calculated by combining it with the production weight coefficient corresponding to that container in the production weight coefficient matrix to obtain the weighted material requirement. The container material demand forecast data is obtained by calculating the difference between the weighted material demand and the warehouse inventory data in the standard input data.

[0038] In this embodiment, after performing time-series prediction based on standard input data and the production weight coefficient matrix to obtain warehouse management prediction data, the method further includes: generating and issuing warehouse scheduling execution instructions based on the warehouse management prediction data, obtaining corresponding on-site execution feedback data; and updating the production weight coefficient matrix according to the deviation between the on-site execution feedback data and the warehouse management prediction data.

[0039] In this embodiment, time-series forecasting refers to an analytical method that predicts future warehouse management indicators based on data from the entire production process, such as predicting the demand for containers in the next 24 hours. Warehouse management forecast data refers to the warehouse management prediction results obtained through time-series forecasting, with container material demand forecast data being a core component. Container material demand forecast data refers to the predicted shortage data of materials needed for container production, used to guide material scheduling and procurement. The future preset time period refers to a pre-set forecast coverage period, such as the next 24 hours of production operation days. The planned production quantity refers to the planned total production volume of a single container within the future preset time period, such as a planned production of 100 containers per day. Material unit consumption data refers to the quota consumption of materials required for the production of a single container, such as a single container requiring four 40×40 square tubes.

[0040] Theoretical material requirement refers to the total material requirement calculated based on planned output and unit consumption. Weighted material requirement refers to the material requirement adjusted by production weighting coefficients. Warehouse inventory data refers to the real-time available inventory quantity of corresponding materials in the warehouse. Warehouse scheduling execution instructions refer to warehousing operation instructions generated based on forecast results, such as material outbound instructions. On-site execution feedback data refers to the actual execution data after the completion of warehousing operations, such as the actual material outbound quantity. Deviation refers to the difference between forecast data and actual execution data, used for iterative optimization of the weight matrix. The production weighting coefficient matrix refers to a set of weighting coefficients constructed according to container number and production stage, used to correct forecast data.

[0041] For example, this embodiment first completes the matching and binding of standard input data and production weight coefficient matrix. The standard input data is the preprocessed normalized data of the entire container production process. The production weight coefficient matrix is ​​a set of weights constructed according to the container number and production stage type, ensuring that each production weight coefficient in the matrix has a unique correspondence with a single container, and avoiding data matching misalignment.

[0042] This embodiment extracts the planned production quantity of each container within a preset future time period from the production scheduling data in the standard input data. The preset future time period is 24 hours, i.e., a complete production work day. The production scheduling data comes from the preprocessed daily production plan data issued by the advanced production scheduling system, and includes a unique number corresponding to each container, the planned production start time, the planned completion time, and the total planned daily production quantity. The preset extraction rule in this embodiment is to filter all container production tasks in the production scheduling data whose planned production start time is within the next 24-hour period, and extract the corresponding daily planned production quantity one by one according to the container number. For example, for container number SJ001, the planned production quantity extracted for the next 24 hours is 100 pieces; for container number SJ002, the planned production quantity extracted for the next 24 hours is 50 pieces, thus completing the extraction and aggregation of the planned production quantities of all containers.

[0043] This embodiment calculates the theoretical material requirement for each container based on the material consumption data and planned production quantity in the standard input data. The material consumption data comes from the container production bill of materials file and is preprocessed quota data, including the specifications and unit consumption of various materials such as rectangular tubes and cold-rolled plates required for the production of a single container. The preset calculation rule in this embodiment is to match the corresponding material consumption data according to the container number, and multiply the unit consumption value of each type of material for a single container by the planned production quantity of that container to obtain the theoretical material requirement for each type of material. For example, container number SJ001 requires 4 x 40x40 square tubes and 2 x 2 mm cold-rolled plates per unit. With a planned production quantity of 100 units, the theoretical material requirement for the 40x40 square tubes is calculated to be 400 units, and the theoretical material requirement for the 2 mm cold-rolled plates is 200 units. Container number SJ002 requires 6 x 40x40 square tubes and 3 x 2 mm cold-rolled plates per unit. With a planned production quantity of 50 units, the theoretical material requirement for the 40x40 square tubes is calculated to be 300 units, and the theoretical material requirement for the 2 mm cold-rolled plates is 150 units. This completes the calculation of the theoretical material requirements for all types of materials for all containers.

[0044] This embodiment calculates the weighted material requirement by weighting the theoretical material requirement of each container with the corresponding production weight coefficient in the production weight coefficient matrix. In the production weight coefficient matrix, the production weight coefficient for each container represents the weight assigned to the production stage of that container: 0.7 for automatic production, 0.25 for manual production, and 0.05 for intelligent production. The preset weighting calculation rule is as follows: the corresponding production weight coefficient is matched from the production weight coefficient matrix according to the container number; the theoretical material requirement for each type of material corresponding to that container is multiplied by the matched production weight coefficient to obtain the weighted material requirement for each type of material corresponding to that container. For example, container number SJ001 is in the automatic production stage, with a weighting coefficient of 0.7. Its theoretical demand for 40×40 cubic meter tubing is 400 pieces, and after weighted calculation, the weighted material demand is 280 pieces. The theoretical demand for 2mm cold-rolled steel is 200 pieces, and after weighted calculation, it is 140 pieces. Container number SJ002 is in the manual production stage, with a weighting coefficient of 0.25. Its theoretical demand for 40×40 cubic meter tubing is 300 pieces, and after weighted calculation, it is 75 pieces. The theoretical demand for 2mm cold-rolled steel is 150 pieces, and after weighted calculation, it is 37.5 pieces. This completes the calculation of the weighted demand for all materials of all containers.

[0045] This embodiment calculates the difference between the weighted material demand and the warehouse inventory data in the standard input data to obtain the container material demand forecast data. The warehouse inventory data comes from the real-time collected and pre-processed available inventory data of the warehouse management system, including the real-time in-stock quantity, locked quantity, and available quantity of various materials. The preset difference calculation rule in this embodiment is as follows: match the corresponding weighted material demand and warehouse available inventory data according to the material category, subtract the available inventory quantity of the corresponding material from the weighted material demand. When the calculation result is greater than 0, this value is the shortfall demand forecast value for that material; when the calculation result is less than or equal to 0, the demand forecast value for that material is 0. After all the demand forecast values ​​for materials are aggregated by container number, the container material demand forecast data is obtained, completing the generation of this time-series forecast and warehouse management forecast data.

[0046] After obtaining the warehouse management forecast data, this embodiment generates and issues warehouse scheduling execution instructions based on the forecast data. According to the material shortage value in the container material demand forecast data, it generates corresponding material replenishment purchase instructions, in-warehouse sorting and outbound instructions, and material delivery instructions. These instructions are then sent to the warehouse management terminal, the on-site logistics terminal, and the procurement management terminal, specifying the operation completion deadlines and requirements. Subsequently, this embodiment acquires the corresponding on-site execution feedback data in real time, including the actual material purchase and warehousing quantity, the actual outbound sorting quantity, the actual material quantity delivered to the production workstation, and the actual material consumption in production. All feedback data is categorized by container number and material type to ensure complete matching with the dimensions of the forecast data. This embodiment updates the production weight coefficient matrix based on the deviation between on-site execution feedback data and warehouse management forecast data. The preset deviation calculation rule is to calculate the deviation rate between the forecast data and the actual execution data according to the material category and container number. The preset deviation threshold is 15%. When the overall deviation rate corresponding to a single container exceeds 15%, the weight coefficient update mechanism is triggered. The weight coefficient of the production stage to which the container belongs is adjusted up or down in a step of 0.02. At the same time, upper and lower limits of the weight coefficient are set to ensure that the weight coefficient is always within a reasonable range, thus completing the update of the production weight coefficient matrix for the next iterative optimization of time series forecast.

[0047] This embodiment combines standard input data with a production weighting coefficient matrix to perform time-series forecasting, accurately calculating the predicted material demand data for containers. This transforms warehouse management from passive recording to proactive prediction, effectively avoiding problems such as material shortages or inventory backlogs. Simultaneously, by iteratively updating the weighting matrix through feedback and deviation analysis, the forecast accuracy is continuously improved, adapting to the production characteristics of containers with multiple models, small batches, and quick changeovers, thereby enhancing warehouse management efficiency and production continuity.

[0048] In one embodiment of this application, the warehouse management forecast data includes container material demand forecast data and storage location allocation forecast data; based on standard input data and a production weight coefficient matrix, time-series forecasting is performed to obtain the warehouse management forecast data, including: The baseline production efficiency of each container is determined based on the equipment operation data and production scheduling data in the standard input data. The baseline production efficiency is then weighted and calculated with the production weight coefficient corresponding to that container in the production weight coefficient matrix to obtain the weighted corrected production efficiency. Based on weighted corrected production efficiency, material unit consumption data in standard input data and warehouse inventory data, time series extrapolation is performed to obtain container material demand forecast data; Based on weighted adjusted production efficiency, container model and size data in standard input data, and warehouse inventory data, time-series forecasts of storage space occupancy are performed to obtain storage space allocation forecast data.

[0049] In this embodiment, the baseline production efficiency refers to the standard production capacity of the containers per unit time at the corresponding production stage, such as the standard production capacity of one container per minute in the automated production stage. The weighted corrected production efficiency refers to the predicted production efficiency value after correction by production weight coefficients, used for subsequent time-series calculations. Container material demand forecast data refers to the material shortage data for container production based on the corrected efficiency prediction. Storage location allocation forecast data refers to the storage location occupancy and allocation scheme data based on the corrected efficiency prediction. Time-series calculation refers to the progressive calculation process of material demand for future periods based on the corrected efficiency. Storage location occupancy time-series prediction refers to the prediction process of storage location occupancy for future periods. Standard input data refers to the preprocessed normalized data of the entire container production process. The production weight coefficient matrix refers to the set of weight coefficients constructed according to container number and production stage. Equipment operation data refers to the real-time operating status data of production equipment. Production scheduling data refers to the time-series data of container production plans. Material unit consumption data refers to the quota consumption data of a single container material. Warehouse inventory data refers to the real-time status data of warehouse materials and storage locations. The container model and size data refer to the container's specifications and external parameters.

[0050] For example, this embodiment first completes the one-to-one matching and binding of standard input data and production weight coefficient matrix. The standard input data is the preprocessed normalized data of the entire container production process. The production weight coefficient matrix is ​​a set of weights constructed according to the container number and production stage type, ensuring that each production weight coefficient in the matrix corresponds precisely to the unique number of a single container, and avoiding data matching misalignment.

[0051] This embodiment determines the baseline production efficiency of each container based on equipment operation data and production scheduling data from the standard input data. The equipment operation data comes from historical unit-time output data, equipment utilization rate data, and process execution time data, collected in real-time and preprocessed by the programmable logic controllers at the production workstations. The production scheduling data comes from process planned cycle time data and daily planned capacity data issued by the advanced production scheduling system and preprocessed. The baseline production efficiency is determined by taking the average unit-time output of the corresponding container in the same production stage over the past seven production cycles, and calibrating it with the planned cycle time data from the production scheduling data to obtain the baseline production efficiency of that container. For example, the container numbered SJ001 is in the automatic production stage, with an average output of 1 piece per minute over the past 7 cycles and a planned production cycle of 0.95 pieces per minute. After calibration, its baseline production efficiency is determined to be 0.95 pieces per minute. The container numbered SJ002 is in the manual production stage, with an average output of 1 piece per 5 minutes over the past 7 cycles and a planned cycle of 1 piece per 6 minutes. After calibration, its baseline production efficiency is determined to be 1 piece per 6 minutes.

[0052] In this embodiment, the baseline production efficiency of each container is weighted and calculated by combining it with the production weight coefficient of the corresponding production stage in the production weight coefficient matrix to obtain a weighted corrected production efficiency. The automatic production stage corresponds to a second production weight coefficient of 0.7, the manual production stage corresponds to a first production weight coefficient of 0.25, and the intelligent production stage corresponds to a third production weight coefficient of 0.05. The preset weighting calculation rule in this embodiment is to multiply the baseline production efficiency of a single container by its corresponding weight coefficient to obtain the weighted corrected production efficiency of that container. For example, the baseline production efficiency of container SJ001 is 0.95 pieces per minute, with a corresponding weight of 0.7, resulting in a weighted corrected production efficiency of 0.665 pieces per minute; the baseline production efficiency of container SJ002 is 1 piece per 6 minutes, with a corresponding weight of 0.25, resulting in a weighted corrected production efficiency of 1 piece per 24 minutes.

[0053] This embodiment uses weighted adjusted production efficiency, material consumption data per unit of standard input data, and warehouse inventory data to perform time-series extrapolation to obtain container material demand forecast data. The future extrapolation period is preset to 24 hours. The material consumption data is the quota consumption of the material corresponding to a single container, and the warehouse inventory data is the real-time available inventory data of the corresponding material. This embodiment first calculates the expected total output for the next 24 hours based on weighted adjusted production efficiency, then calculates the total material demand based on the material consumption data, and finally calculates the difference between the total material demand and the available inventory to obtain the material gap, i.e., the container material demand forecast data.

[0054] This embodiment uses weighted adjusted production efficiency, container model and size data from standard input data, and warehouse inventory data to predict storage space occupancy over time, resulting in storage space allocation prediction data. The container model and size data refer to the container's length, width, and height dimensions, while the warehouse inventory data includes the size, location, and quantity of available warehouse storage spaces. This embodiment first calculates the completed and inbound container volume for each time period within the next 24 hours based on weighted adjusted production efficiency. Then, it calculates the corresponding storage space occupancy based on the size of individual containers. Finally, it matches the size and location of available warehouse storage spaces and allocates storage spaces according to the rule of grouping similar models together and facilitating entry and exit, thus obtaining storage space allocation prediction data. For example, the SJ001 container has dimensions of 400mm×600mm×800mm, and a single standard storage location has dimensions of 1200mm×1000mm×1000mm. A single storage location can store 4 containers of this model. Based on the estimated production of 950 units in the next 24 hours, it is estimated that 238 standard storage locations are needed. After matching the available storage locations in the corresponding area of ​​the warehouse, accurate storage location allocation prediction data is generated.

[0055] This embodiment uses a weighted correction based on benchmark production efficiency and weighting coefficients to simultaneously predict the timing of container material demand and storage location allocation. This ensures that the warehousing forecast results accurately match the actual production rhythm at different production stages, effectively avoiding material supply and demand imbalances and storage location waste, and improving the accuracy and adaptability of container warehousing management.

[0056] Based on the same principle as the warehouse management forecasting method provided in the embodiments of this application, the embodiments of this application also provide a warehouse management forecasting device, such as... Figure 2 As shown, the warehouse management forecasting device 20 may specifically include: a multi-source data acquisition module 21, a production stage weight determination module 22, and a warehouse management forecasting module 23. The multi-source data acquisition module 21 is used to acquire multi-source heterogeneous data of multiple containers, preprocess the multi-source heterogeneous data to obtain standard input data, and the multi-source heterogeneous data includes container model and size data, material unit consumption data, production scheduling data, equipment operation data, logistics status data, and warehouse inventory data. The production stage weight determination module 22 is used to determine the production stage type of each container based on standard input data, and to assign production weight coefficients to each container according to the production stage type, thereby obtaining a production weight coefficient matrix corresponding to multiple containers; the production stage type is manual production stage, automatic production stage or intelligent production stage. The warehouse management forecasting module 23 is used to perform time-series forecasting based on standard input data and production weight coefficient matrix to obtain warehouse management forecasting data.

[0057] In one embodiment of this application, the multi-source data acquisition module 21 is specifically used for: Missing values ​​are filled in multi-source heterogeneous data to obtain data with missing values ​​completed. The missing value-completed data is then subjected to outlier removal to obtain outlier-filtered data. The units of the outlier filtering data are standardized to obtain data with standardized units. The unified data of each unit is classified and labeled according to the container model, material type and production stage to obtain classified label data; The categorical label data is time-aligned to obtain standard input data.

[0058] In one embodiment of this application, the production stage weight determination module 22 is specifically used for: Identify the process type of each container based on the equipment operation data and production scheduling data in the standard input data; When the process type is manual part removal, manual welding, manual repair welding or manual painting, the container is determined to be in the manual production stage; When the process type is laser cutting, automatic feeding, automatic welding, induction heating, correction or robot head sealing, the container is determined to be in the automatic production stage; When the process type is project design, review and scheduling, material procurement or electronic contract management, the container is determined to be in the intelligent production stage.

[0059] In one embodiment of this application, the production stage weight determination module 22 is further configured to: assign a first production weight coefficient to containers in the manual production stage, assign a second production weight coefficient to containers in the automatic production stage, and assign a third production weight coefficient to containers in the intelligent production stage; the second production weight coefficient is greater than the first production weight coefficient, and the first production weight coefficient is greater than the third production weight coefficient. The production weight coefficients corresponding to each container are arranged in a matrix according to the container number and production stage type to obtain a matrix of production weight coefficients corresponding to multiple containers.

[0060] In one embodiment of this application, the warehouse management forecast data includes container material demand forecast data; the warehouse management forecast module 23 is specifically used for: Extract the planned production quantity of each container within a future preset time period from the production scheduling data in the standard input data; Based on the material unit consumption data and planned production quantity in the standard input data, the theoretical material requirement for each container is calculated. The theoretical material requirement for each container is weighted and calculated by combining it with the production weight coefficient corresponding to that container in the production weight coefficient matrix to obtain the weighted material requirement. The container material demand forecast data is obtained by calculating the difference between the weighted material demand and the warehouse inventory data in the standard input data.

[0061] In one embodiment of this application, the warehouse management forecast data includes container material demand forecast data and storage location allocation forecast data; the warehouse management forecast module 23 is specifically used for: The baseline production efficiency of each container is determined based on the equipment operation data and production scheduling data in the standard input data. The baseline production efficiency is then weighted and calculated with the production weight coefficient corresponding to that container in the production weight coefficient matrix to obtain the weighted corrected production efficiency. Based on weighted corrected production efficiency, material unit consumption data in standard input data and warehouse inventory data, time series extrapolation is performed to obtain container material demand forecast data; Based on weighted adjusted production efficiency, container model and size data in standard input data, and warehouse inventory data, time-series forecasts of storage space occupancy are performed to obtain storage space allocation forecast data.

[0062] In one embodiment of this application, the warehouse management prediction device 20 further includes a feedback module, configured to: generate and issue warehouse scheduling execution instructions based on warehouse management prediction data, obtain corresponding on-site execution feedback data, and update the production weight coefficient matrix according to the deviation between the on-site execution feedback data and the warehouse management prediction data.

[0063] Each module in the aforementioned warehouse management forecasting device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0064] In some embodiments, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 3 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores data such as container dimensions. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a warehouse management forecasting method.

[0065] In some embodiments, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 4As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a warehouse management prediction method. The display unit of the computer device is used to form a visually visible image. It can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0066] Those skilled in the art will understand that Figure 3 , Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0067] In some embodiments, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described warehouse management forecasting method.

[0068] In some embodiments, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described warehouse management prediction method.

[0069] In some embodiments, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the above-described warehouse management forecasting method.

[0070] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0071] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, data processing logic devices, etc., and are not limited to these.

[0072] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0073] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A warehouse management forecasting method, characterized in that, include: Acquire multi-source heterogeneous data from multiple containers, preprocess the multi-source heterogeneous data to obtain standard input data; the multi-source heterogeneous data includes container model and size data, material unit consumption data, production scheduling data, equipment operation data, logistics status data, and warehouse inventory data; Based on the standard input data, the production stage type of each container is determined, and a production weight coefficient is assigned to each container according to the production stage type, thus obtaining the production weight coefficient matrix corresponding to the multiple containers; the production stage type is manual production stage, automatic production stage, or intelligent production stage. Based on the standard input data and the production weight coefficient matrix, time-series prediction is performed to obtain warehouse management prediction data.

2. The warehouse management forecasting method as described in claim 1, characterized in that, The preprocessing of the multi-source heterogeneous data to obtain standard input data includes: The missing values ​​of the multi-source heterogeneous data are filled to obtain the missing value-completed data; The missing value-completed data is then subjected to outlier removal to obtain outlier-filtered data. The outlier filtering data is then standardized in terms of units to obtain standardized data. The unified data of the unit is classified and labeled according to the container model, material type and production stage to obtain classified label data; The classification label data is time-aligned to obtain the standard input data.

3. The warehouse management forecasting method as described in claim 1, characterized in that, The process of determining the production stage type for each container based on the standard input data includes: The process type of each container is identified based on the equipment operation data and production scheduling data in the standard input data. When the process type is manual part removal, manual welding, manual repair welding or manual painting, the container is determined to be in the manual production stage; When the process type is laser cutting, automatic feeding, automatic welding, induction heating, correction, or robot head sealing, the container is determined to be in the automatic production stage. When the process type is project design, review scheduling, material procurement, or electronic contract management, the container is determined to be in the intelligent production stage.

4. The warehouse management forecasting method as described in claim 1, characterized in that, The process of assigning production weight coefficients to each container according to its production stage type, resulting in a production weight coefficient matrix corresponding to the multiple containers, includes: A first production weight coefficient is assigned to containers in the manual production stage, a second production weight coefficient is assigned to containers in the automatic production stage, and a third production weight coefficient is assigned to containers in the intelligent production stage; the second production weight coefficient is greater than the first production weight coefficient, and the first production weight coefficient is greater than the third production weight coefficient. The production weight coefficients corresponding to each container are arranged in a matrix according to the container number and production stage type to obtain the production weight coefficient matrix corresponding to the multiple containers.

5. The warehouse management forecasting method as described in claim 1, characterized in that, The warehouse management forecast data includes forecast data on the demand for containers; The process of performing time-series prediction based on the standard input data and the production weight coefficient matrix to obtain warehouse management prediction data includes: Extract the planned production quantity of each container within a future preset time period from the production scheduling data in the standard input data; Based on the material consumption data in the standard input data and the planned production quantity, the theoretical material requirement for each container is calculated. The theoretical material requirement for each container is weighted and calculated by combining it with the production weight coefficient corresponding to that container in the production weight coefficient matrix to obtain the weighted material requirement. The container material demand forecast data is obtained by calculating the difference between the weighted material demand and the warehouse inventory data in the standard input data.

6. The warehouse management forecasting method as described in claim 1, characterized in that, The warehouse management forecast data includes container material demand forecast data and storage location allocation forecast data; The process of performing time-series prediction based on the standard input data and the production weight coefficient matrix to obtain warehouse management prediction data includes: Based on the equipment operation data and production scheduling data in the standard input data, the baseline production efficiency of each container is determined. The baseline production efficiency is then weighted and calculated with the production weight coefficient corresponding to the container in the production weight coefficient matrix to obtain the weighted corrected production efficiency. Based on the weighted corrected production efficiency, the material unit consumption data in the standard input data and the warehouse inventory data, time series extrapolation is performed to obtain the container material demand forecast data; Based on the weighted corrected production efficiency, the container model and size data in the standard input data, and the warehouse inventory data, the time series prediction of storage space occupancy is performed to obtain storage space allocation prediction data.

7. The warehouse management forecasting method as described in claim 1, characterized in that, After performing time-series prediction based on the standard input data and the production weight coefficient matrix to obtain warehouse management prediction data, the method further includes: Based on the warehouse management prediction data, generate and issue warehouse scheduling execution instructions, and obtain corresponding on-site execution feedback data; The production weight coefficient matrix is ​​updated based on the deviation between the on-site execution feedback data and the warehouse management forecast data.

8. A warehouse management forecasting device, characterized in that, include: The multi-source data acquisition module is used to acquire multi-source heterogeneous data from multiple containers, preprocess the multi-source heterogeneous data to obtain standard input data; the multi-source heterogeneous data includes container model and size data, material unit consumption data, production scheduling data, equipment operation data, logistics status data and warehouse inventory data; The production stage weight determination module is used to determine the production stage type of each container based on the standard input data, and to assign production weight coefficients to each container according to the production stage type, thereby obtaining a production weight coefficient matrix corresponding to the multiple containers; the production stage type is manual production stage, automatic production stage, or intelligent production stage. The warehouse management forecasting module is used to perform time-series forecasting based on the standard input data and the production weight coefficient matrix to obtain warehouse management forecasting data.

9. A computer device, characterized in that, The computer device includes a processor and memory: The memory is used to store computer programs; The processor is configured to execute the warehouse management forecasting method according to any one of claims 1-7 according to the computer program.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program, which, when executed by a computer device, implements the warehouse management prediction method according to any one of claims 1-7.

Images