Inventory quota interval prediction method and device, medium and product

Abstract

The invention discloses an inventory quota interval prediction method and device, a medium and a product. The method comprises the following steps: in response to a prediction request of an inventory quota interval, determining target material data of target materials in a target prediction organization according to a prediction time period and the target prediction organization; determining an evaluation feature matrix corresponding to the target material data, and predicting an optimal inventory theoretical value corresponding to the target material based on a preset random forest model according to the target material data and the evaluation feature matrix; and determining a material importance weight corresponding to the target material, and determining an inventory quota interval of the target material according to the material importance weight and the optimal inventory theoretical value, thereby performing supervision and early warning on the material inventory condition of the target prediction organization based on the inventory quota interval. According to the technical scheme, the material data can be comprehensively evaluated, and the accurate inventory quota interval can be determined in combination with the random forest model, so that effective and comprehensive inventory management can be realized.

Description

Technical Field

[0001] This invention relates to the field of big data, and in particular to a method, apparatus, medium and product for predicting inventory quota ranges. Background Technology

[0002] Inventory quotas are a core indicator for enterprises to balance "material support capability" and "inventory cost control". Traditional inventory quota setting often relies on the experience and judgment of managers or adopts a simple "historical average method", which has problems such as single data dimension, insufficient forecast accuracy and lack of dynamic adjustment mechanism.

[0003] As digital transformation progresses, enterprises have accumulated a large amount of historical data on procurement, consumption, and inventory. How to use this data to build intelligent prediction models and output scientific inventory quota ranges (rather than a single value) has become the key to solving the pain points of traditional inventory management. Summary of the Invention

[0004] This invention provides a method, apparatus, medium, and product for predicting inventory quota ranges, which comprehensively evaluates material data and combines it with a random forest model to determine accurate inventory quota ranges, thereby helping to achieve effective and comprehensive inventory management.

[0005] According to one aspect of the present invention, an inventory quota range forecasting method is provided, comprising: In response to a forecast request for an inventory quota range, the target material data for the target materials in the target forecast organization is determined based on the forecast period and the target forecast organization. Determine the evaluation feature matrix corresponding to the target material data, and based on the preset random forest model, predict the optimal theoretical inventory value corresponding to the target material according to the target material data and the evaluation feature matrix; Determine the material importance weight corresponding to the target material, and based on the material importance weight and the optimal inventory theoretical value, determine the inventory quota range of the target material, so as to monitor and warn of the material inventory status of the target forecast organization based on the inventory quota range.

[0006] According to another aspect of the present invention, an inventory quota range prediction device is provided, comprising: The data determination module is used to respond to the forecast request for the inventory quota range and determine the target material data of the target material in the target forecast organization based on the forecast period and the target forecast organization. The prediction module is used to determine the evaluation feature matrix corresponding to the target material data, and based on the preset random forest model, predict the optimal theoretical inventory value of the target material according to the target material data and the evaluation feature matrix. The interval determination module is used to determine the material importance weight corresponding to the target material, and to determine the inventory quota range of the target material based on the material importance weight and the optimal inventory theoretical value, so as to monitor and warn the material inventory status of the target forecast organization based on the inventory quota range.

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

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

[0009] According to another aspect of the present invention, a computer program product is also provided, the computer program product including a computer program that, when executed by a processor, implements the inventory quota range prediction method of any embodiment of the present invention.

[0010] The technical solution of this invention, in response to a forecast request for an inventory quota range, determines the target material data of the target materials in the target forecast organization based on the forecast period and the target forecast organization; determines the evaluation feature matrix corresponding to the target material data, and predicts the optimal theoretical inventory value corresponding to the target materials based on a preset random forest model, according to the target material data and the evaluation feature matrix; determines the material importance weight corresponding to the target materials, and determines the inventory quota range of the target materials based on the material importance weight and the optimal theoretical inventory value, so as to monitor and warn of the material inventory status of the target forecast organization based on the inventory quota range. By comprehensively evaluating the material data and combining it with the random forest model to determine an accurate inventory quota range, it can help achieve effective and comprehensive inventory management.

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

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

[0013] Figure 1 This is a flowchart of an inventory quota range prediction method provided by an embodiment of the present invention; Figure 2 This is a flowchart of an inventory quota range prediction method provided by an embodiment of the present invention; Figure 3 This is a structural block diagram of an inventory quota range prediction device provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of the electronic device provided in an embodiment of the present invention. Detailed Implementation

[0014] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

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

[0016] Traditional inventory quota forecasting schemes suffer from the following problems: First, they rely on a single data dimension, setting quotas based solely on historical inventory data or static consumption data, failing to consider dynamic factors such as procurement cycle fluctuations, production plan adjustments, and market supply risks, leading to a disconnect between quotas and actual business needs. Second, their forecasting accuracy is insufficient; traditional inventory quota forecasting schemes often output single numerical quotas, unable to cope with business fluctuations (such as sudden project demands or supplier delivery delays), easily resulting in "inventory backlog" or "material shortages." Third, they lack a dynamic adjustment mechanism; quotas remain unchanged for a long period after being set, failing to adapt in a timely manner when the company's business expands, new material categories are added, or external markets change, leading to low inventory management efficiency. Specifically, some existing technologies use single time series models to predict inventory demand, but struggle to handle multi-dimensional features; others focus on cost optimization, neglecting the rigid requirements of production assurance, failing to simultaneously meet the three major objectives of "accurate forecasting," "dynamic adaptation," and "cost balance." To address the aforementioned issues, this invention provides a corporate inventory quota range prediction scheme based on multi-dimensional historical data. By integrating multi-source data and constructing a prediction model, it outputs an inventory quota range that includes "guaranteed lower limit - optimal value - cost upper limit". At the same time, it combines a business function feedback mechanism to solve the problems of experience dependence, insufficient accuracy, and poor dynamism in traditional inventory quota setting. The specific implementation method will be described in detail in subsequent embodiments.

[0017] Example 1 Figure 1 This is a flowchart of an inventory quota range prediction method provided by an embodiment of the present invention. This embodiment is applicable to situations where a comprehensive evaluation of material data is performed and an accurate inventory quota range is determined by combining a random forest model, thereby contributing to effective and comprehensive inventory management. This method can be executed by an inventory quota range prediction device, which can be implemented in hardware and / or software. This inventory quota range prediction device can be configured in an electronic device, such as an electronic device equipped with an inventory quota range prediction system. Figure 1 As shown, the inventory quota range forecasting method includes: S101. In response to a forecast request for an inventory quota range, determine the target material data for the target materials in the target forecast organization based on the forecast period and the target forecast organization.

[0018] The inventory quota range refers to a pre-defined assessment range for materials within an organization. The inventory quota range includes a lower limit, an upper limit, and an optimal value. A forecast request is a request to forecast the inventory quota range for each target material within the target forecasting organization. The forecast period can be the next year or the next month. The target forecasting organization is the organization that manages the target forecasted inventory quota range. The target forecasting organization corresponds to the procurement, consumption, and inventory management of at least one target material. Target materials can be, for example, wire rope, bearings, and bolts. Target material data can include target procurement data, target consumption data, and target inventory data, and may also include cost accounting data.

[0019] Optionally, the target procurement data, target consumption data, and target inventory data corresponding to the target forecasting organization can be determined from the target database based on the organization identifier, business type, and organizational level of the target forecasting organization, and used as target material data.

[0020] For example, the inventory quota range forecasting system can obtain parameters through front-end interface form components (such as organization pop-up boxes and date selectors) to determine the forecast period and target forecasting organization. The forecast period is a future year of the target forecast (e.g., if the user selects 2026, the inventory quota range for 2026 will be forecast). The system verifies the validity of the year format (e.g., it must be a 4-digit number) using a date verification algorithm. The target forecasting organization is obtained by instructing the user to select an organization table. The system backend reads the unique identifier (ORG_ID, i.e., organization identifier), organization level, business type, and other related information of the corresponding organization from the organization information table to ensure that the parameters are accurately mapped to the material data of the target forecasting organization.

[0021] For example, business types can include port logistics, port operation, harbor construction, container shipping services, technological innovation, investment and operation, healthcare, or equipment manufacturing. Organizational levels can be group-level, subsidiary-level, or branch-level.

[0022] Optionally, before determining the target material data of the target materials in the target forecasting organization based on the forecast period and the target forecasting organization, the process further includes: obtaining the management material data of each management organization from the material management system of the management organization through a preset data collection tool based on a preset collection cycle, and determining the organization identifier, business type and organizational level of each management organization; performing anomaly removal and missing data supplementation on the management material data of each management organization, and generating a material information table, a procurement data table, a consumption data table and an inventory data table based on the processed management material data, organization identifier, business type and organizational level, and storing them in the target database through a table-segmented storage method.

[0023] The materials data includes at least one of the following: purchased materials data, consumed materials data, and inventory materials data. The default data collection tool is Kettle (also known as Pentaho Data Integration), a tool specifically designed for data extraction, transformation, and loading. The default collection period is daily. The management organization refers to all organizations managed by the materials management system.

[0024] For example, a material information table can store static information such as material code, material name, specifications, category (critical spare parts / bulk consumables), and importance weight, with the material code as the primary key. The procurement data table includes fields such as order number, material code, purchase quantity, unit price, tax-free amount, tax rate, tax amount, total amount including tax, supplier number, order date, arrival date, arrival days, warehousing time, warehousing quantity, and data cleaning status (normal / corrected). The primary key is order number + material code, and the foreign key links to the material basic information table and the supplier information table. The consumption data table covers fields such as material requisition organization, material requisition department, material code, consumption quantity, unit price, consumption amount, consumption date, corresponding production work order / maintenance task number, equipment, asset, project, and cleaning status. The primary key is material requisition number + material code, and the foreign key links to the material basic information table. The inventory data table includes fields such as organization, warehouse code, warehouse description, storage location, material code, material description, current inventory quantity, average unit price, inventory amount, inventory turnover rate, stagnant inventory identifier, and cleaning status. The primary key is material code + warehouse code, and the foreign key links to the material basic information table and the warehouse information table.

[0025] For example, the table partitioning method refers to the data storage using a combination of row storage and column storage. High-frequency query fields (such as material codes and dates) are stored in rows, while batch analysis fields (such as purchase quantity and consumption quantity) are stored in columns, thereby improving data read and write efficiency.

[0026] For example, at a fixed time each day (such as 3 a.m.), the Kettle tool can be used to extract the previous day's procurement, consumption, and inventory data from the corresponding material management system of each management organization, thus determining the managed material data.

[0027] For example, when performing data extraction using the Kettle tool, it connects to the business database of the organization's internal materials management system through a preset ETL (Extract-Transform-Load) job script, and reads the specified data tables accordingly. Specifically, procurement data can be extracted from the purchase order table, goods receipt and acceptance table, and receiving and warehousing table, including fields such as order number, material code, purchase quantity, unit price, tax-free amount, tax rate, tax amount, total amount including tax, supplier number, order date, arrival date, arrival days, warehousing time, and warehousing quantity; consumption data can be extracted from the material requisition form and material issuance form, covering information such as material requisition organization, material requisition department, material code, consumption quantity, unit price, consumption amount, consumption date, corresponding production work order / maintenance task number, equipment, assets, and project; inventory data can be extracted from the inventory table, storage location table, storage location batch inventory table, and organization table, including the organization, warehouse code, warehouse description, storage location, material code, material description, current inventory quantity, average unit price, inventory amount, batch, and warehousing time.

[0028] For example, if the managed material is steel wire rope, the corresponding procurement material data will additionally include fields such as steel wire rope specifications (diameter, strength grade), quantity, and required date. The consumption material data will be associated with information such as specific work order number, equipment number, equipment description, project number, and project description. The inventory material data will record the steel wire rope's warehousing time, shelf life warning information, etc.

[0029] For example, Kettle can be used to directly connect to the organization's corresponding materials management system. The procurement management system and financial system, on the other hand, connect to the materials management system via message queues, synchronizing key business data and metrics to the materials management system. Then, during data extraction, procurement, consumption, inventory, and cost accounting data are extracted incrementally on a daily basis to obtain managed materials data. Managed materials data can include procured materials data, consumed materials data, and inventory materials data.

[0030] Optionally, a daily collection script can be executed to extract managed material data from the material management system daily and perform cleaning processing, that is, to remove abnormal data in the procurement data and fill in missing values ​​in the inventory data (such as missing inventory on November 2nd, which is filled in by the average inventory of November 1st and November 3rd).

[0031] Optionally, anomaly removal processing is performed on the managed material data of each management organization, including: for the material data of each management organization, determining the target procurement quantity of each target material based on the procurement data, and determining whether there are quantity anomalies in the procurement data based on the target procurement quantity, the historical maximum regular procurement quantity, and the historical average procurement quantity; determining the procurement unit price of each target material based on the procurement data, and determining whether there are price anomalies in the procurement data based on the procurement unit price, the reasonable range of market unit prices, and the lower limit of supplier quotations; determining the daily consumption of each target material based on the consumption data, and determining whether there are sudden abnormal consumption based on the correlation between the monthly average total consumption and the daily consumption of the target materials; and performing anomaly removal processing if quantity anomalies, price anomalies, or sudden abnormal consumption are determined.

[0032] Optionally, if the target purchase quantity exceeds the historical maximum regular purchase quantity or the historical average purchase quantity, a quantity anomaly is confirmed; otherwise, no quantity anomaly is considered to exist. If the purchase unit price exceeds the reasonable range of market unit prices or is 30% lower than the supplier's lower limit, a price anomaly is confirmed; otherwise, no price anomaly is considered to exist. If the daily consumption of the target material exceeds 50% of the monthly average total consumption, a sudden abnormal consumption is confirmed; otherwise, no sudden abnormal consumption is considered to exist.

[0033] For example, regarding procurement data, if a single order's purchase quantity far exceeds the historical reasonable range (e.g., a single order for 1200 units of model 6205 bearings, while the historical maximum regular purchase quantity is 500 units; a single batch of steel wire rope purchases 200 tons, while the average purchase quantity over the past 3 years is 80 tons), then an anomaly in quantity is identified. Regarding consumption data, if a record shows a daily consumption quantity exceeding 50% of the average monthly consumption of that material, then a sudden abnormal consumption event is identified. Regarding inventory data, if there are invalid records showing a negative inventory quantity (inconsistent between book and physical inventory) or an inventory value of 0 but a non-zero quantity, then an anomaly in quantity is identified.

[0034] For example, regarding procurement data, if the unit price of procurement deviates from the reasonable market range (e.g., the unit price of a certain type of bolt is 5 yuan / piece, and the average unit price in the past 3 years is 1.2 yuan / piece, exceeding the ±50% fluctuation threshold, then it is determined that there is a price anomaly; when the unit price of core chips is lower than 30% of the lower limit of the qualified supplier's quotation, it can also be considered that there is a quality risk, i.e., a price anomaly).

[0035] For example, if the order status associated with the managed material data is returned or canceled, it can be determined that there is an invalid status, and the corresponding managed material data is identified as abnormal data and removed.

[0036] For example, by calling the daily collection script, anomaly filtering can be performed during business cleanup, and abnormal records can be directly filtered out; then it can be determined whether the data from the previous day exists. If it does not exist, it is determined that there is missing data, and missing data can be supplemented.

[0037] Optionally, missing data can be supplemented based on the missing type. Specifically, if the missing type is missing material data for a preset duration, the supplementary value is determined based on the average value of material data in the adjacent time period to supplement the missing material data; if the missing type is missing at least one data item, the supplementary value is determined for each data item based on the value of the corresponding data item in the historical material data to supplement the missing material data.

[0038] Optionally, missing data processing can be performed on the management material data of each management organization, including: missing data processing on inventory material data in the management material data based on the average of adjacent dates; and missing data processing on purchased material data and consumed material data in the management material data based on the average of the same type of material in the same period.

[0039] For example, missing inventory data can be filled using the "average of adjacent dates" strategy. If a certain material is missing on November 2, it can be filled using the average of 800 tons on November 1 and 820 tons on November 3, which is 810 tons. Missing procurement and consumption data can be filled using the "average of the same type of material during the same period" strategy. (For example, if the consumption data for a certain type of bolt is missing on October 5, the average of the last 5 days of the past 3 months can be used to fill the missing data.)

[0040] Optionally, after only missing data processing, the data format can be standardized (e.g., a unified date format of "YYYY-MM-DD" and a unified unit of measurement of "tons / pieces"), and the cleaned data can be stored in the target database.

[0041] S102. Determine the evaluation feature matrix corresponding to the target material data, and based on the preset random forest model, predict the optimal theoretical inventory value corresponding to the target material according to the target material data and the evaluation feature matrix.

[0042] The evaluation feature matrix refers to a matrix that characterizes the evaluation features of the target material data. The evaluation feature matrix may include at least one evaluation indicator value. Evaluation indicator values ​​include at least one of the following: consumption volatility, monthly peak-to-valley difference, consumption trend slope, procurement cycle stability, on-time delivery rate, procurement cost volatility, inventory turnover rate, percentage of obsolete inventory, and inventory holding cost rate.

[0043] Optionally, determine the evaluation feature matrix corresponding to the target material data, including: for each target material, perform data evaluation based on the corresponding target procurement data, target consumption data, and target inventory data, and determine the evaluation index value; generate the evaluation feature matrix based on the determined evaluation index value; For example, consumption volatility can be the annual consumption volatility, such as the standard deviation / mean of consumption data over the past 3 years. Monthly peak-to-trough difference can be the difference between the maximum and minimum monthly consumption. The consumption trend slope can be obtained by fitting the trend of consumption data over the past 3 years using linear regression. Procurement cycle stability can be the coefficient of variation of procurement delivery cycles over the past 5 years, obtained by the standard deviation / mean. On-time delivery rate can be the number of on-time orders delivered by the supplier / total number of orders. Procurement cost volatility can be the standard deviation / mean of the unit price of purchases over the past 3 years. Inventory turnover rate can be the annual inventory turnover rate, determined by the total annual consumption / average annual inventory. Obsolete inventory ratio can be the quantity of obsolete inventory / total inventory quantity. Obsolete inventory is defined as inventory that has not been consumed for 24 consecutive months. Inventory holding cost rate can be the annual inventory holding cost / total value of annual inventory.

[0044] For example, based on a pre-defined random forest model, the optimal theoretical inventory value Y_opt for the target material can be determined according to the following Steps 1-5: Step 1: Data Preprocessing. From the data tables in the target database, determine the evaluation indicator values ​​for the past 5 years by "organization + month," construct an evaluation feature matrix X, and uniformly perform z-score standardization (rows = monthly / yearly samples, columns = 9 feature variables: consumption volatility, monthly peak-to-trough difference, consumption trend slope, procurement cycle stability, on-time delivery rate, procurement cost volatility, inventory turnover rate, obsolete inventory ratio, and inventory holding cost rate). Simultaneously, extract the actual optimal inventory value for the corresponding month as the label. Pearson coefficient checks are performed on the features to avoid the impact of collinearity on the model.

[0045] Step 2: Divide the data into training and test sets. Use a 7:3 ratio for the data, with 70% used as the training set and 30% as the test set.

[0046] Step 3: Random Forest Model Configuration. Based on the previous parameter search and validation, a random forest model with 200 trees was ultimately chosen as the optimal inventory predictor. The model parameters were set as follows: random seed was set to 42 to ensure experimental repeatability; maximum tree depth was set to 10, minimum number of splits was 5, and minimum number of leaf node samples was set to 2 to control model complexity; node splits used "squared_error" as the criterion, and feature importance was measured based on the reduction in the mean squared error of nodes. These parameters were determined through offline parameter tuning and remained fixed during model training.

[0047] Step 4: Model Training and Optimization. During model training, a resampling with replacement mechanism is used to randomly generate a training subset for each tree to enhance the model's diversity across samples. When splitting a node, three features are randomly selected (the feature subset size is approximately 3) to enhance the model's randomness. To determine the optimal parameter combination, offline cross-validation based on time series is used for key parameters such as tree depth, minimum number of samples per leaf node, and feature subset size during the model development phase, with the validation error used as the selection criterion. After the parameters are determined, the model is trained once on the complete training set without online or real-time parameter tuning, thus ensuring a stable training process and controllable results.

[0048] Step 5: The model outputs the predicted optimal theoretical inventory value Y_opt.

[0049] S103. Determine the material importance weight corresponding to the target material, and determine the inventory quota range of the target material based on the material importance weight and the optimal inventory theoretical value, so as to monitor and warn the material inventory status of the target forecast organization based on the inventory quota range.

[0050] Among them, the material importance weight refers to the weight determined by judging whether the target material belongs to the critical spare parts and bulk consumables.

[0051] Optionally, based on the importance weight of materials and the theoretical value of optimal inventory, a two-way asymmetric constraint mechanism of "supply guarantee" and "cost mitigation" can be adopted to calculate the final quota range, that is, to determine the inventory quota range of the target materials.

[0052] Optionally, the importance weight of the target material can be calculated as "critical spare parts x bulk consumables," such as a weight of 0.9 for core scarce materials and a weight of 0.1 for auxiliary abundant materials. For example, core scarce materials at container port terminals are mostly facilities, specialized equipment, and key technical components that directly determine the port's throughput capacity and operational efficiency. These materials often have high construction or R&D costs and are irreplaceable. Auxiliary abundant materials, on the other hand, are general-purpose tools and consumables that ensure the daily operation of the port, which are easy to procure, can be stockpiled in bulk, and have substitutable functions. For example, core scarce materials may include quay cranes (quay cranes), rail-mounted yard cranes, and automated guided vehicles, while auxiliary abundant materials may include cleaning tools, high-pressure cleaning equipment, lubricants, seals, fire-fighting equipment, general lighting equipment, rubber anti-collision pads, and general mooring ropes.

[0053] Optionally, the material importance weight corresponding to the target material is determined, including: determining whether the target material belongs to the critical spare parts as a first judgment result and whether the target material belongs to the bulk consumables as a second judgment result; and determining the material importance weight corresponding to the target material based on the first judgment result, the second judgment result and the preset adjustment coefficient.

[0054] Optionally, the importance weight of materials can be determined by matrix weighting based on the classification attributes of the materials (such as critical spare parts / non-critical spare parts, bulk consumables / non-bulk consumables). Specifically, first, assign values ​​to the two types of attributes respectively (such as critical spare parts = 0.8, non-critical spare parts = 0.2; bulk consumables = 0.7, non-bulk consumables = 0.3). Then, the importance weight of the material = the attribute value of critical spare parts × the attribute value of bulk consumables + the preset adjustment coefficient (the preset adjustment coefficient is preset according to the degree of impact of historical shortages of materials, ranging from 0 to 0.2).

[0055] For example, if a special spindle bearing is a "critical spare part + non-bulk consumable", then its material importance weight = 0.8 × 0.3 + 0.1 = 0.34; if a special resin is a "non-critical spare part + bulk consumable", then its material importance weight = 0.2 × 0.7 + 0.1 = 0.24; if a common bolt is a "non-critical spare part + non-bulk consumable", then its material importance weight = 0.2 × 0.3 + 0 = 0.06.

[0056] Optionally, the inventory quota range for the target material is determined based on the material importance weight and the optimal theoretical inventory value, including: determining the lower limit of the target material based on the material importance weight, the preset supply guarantee gap coefficient, and the optimal theoretical inventory value corresponding to the target material; determining the upper limit of the target material based on the preset cost smoothing slack coefficient, the procurement cost volatility, and the optimal theoretical inventory value corresponding to the target material; determining the optimal value based on the upper and lower limits; and determining the inventory quota range for the target material based on the upper, lower, and optimal values.

[0057] The lower limit is based on the supply guarantee logic, which aims to ensure the continuity of production. The more important the material (the higher the weight of the material's importance) and the more unstable the supply, the greater the guarantee provided by the system, and the closer the lower limit is to the optimal value.

[0058] For example, based on the material importance weight, the preset supply guarantee gap coefficient, and the optimal theoretical inventory value corresponding to the target material, the lower limit value of the target material can be determined using the formula Y_min=Y_opt×(1-α×(1-W)), where Y_min is the lower limit value, Y_opt is the optimal theoretical inventory value, α is the preset supply guarantee gap coefficient (e.g., 0.25), and W is the material importance weight. The dynamic adjustment logic is that α is negatively correlated with the procurement cycle variation coefficient (CV, i.e., risk). For every 10% increase in CV above the baseline, α decreases by 0.05 (i.e., the greater the risk, the more safety stock is needed).

[0059] The upper limit is based on cost mitigation logic and aims to control capital occupation. The higher the holding cost, the more limited the upper limit, in order to avoid unnecessary accumulation.

[0060] For example, based on the preset cost smoothing slack coefficient, procurement cost volatility, and the optimal theoretical inventory value corresponding to the target material, the upper limit of the target material can be determined using the formula Y_max = Y_opt × (1 + β × P_vol), where Y_max is the upper limit, Y_opt is the optimal theoretical inventory value, β is the cost smoothing slack coefficient with a preset benchmark value of 0.20, and P_vol is the procurement cost volatility. The dynamic adjustment logic is that β is negatively correlated with the inventory holding cost rate; for every 5% increase in the holding cost rate, β decreases by 0.05.

[0061] Optionally, the midpoint between the upper and lower limits can be determined as the optimal value. For example, consider the dedicated spindle bearing of organization A: Lower limit calculation: CV=20% (high risk), which causes the supply security gap coefficient α to narrow from the benchmark 0.25 to 0.20. Y_min=50×[1-0.20×(1-0.84)]=50×0.968≈48 sets.

[0062] Upper limit calculation: P_vol=15%; the high inventory holding cost rate leads to a decrease in the cost smoothing slack coefficient β to 0.15. Y_max=50×(1+0.15×0.15)≈51 sets. The final output range is 48-51 sets, with an optimal value of 50 sets.

[0063] Optionally, after determining the inventory quota range for the target materials, the actual purchase amount, consumption amount, and inventory amount for the year can be collected at the end of each year and compared with the predicted inventory quota range. If the actual inventory of a certain type of material is lower than the lower limit of the range (material shortage risk) or higher than the upper limit of the range (inventory backlog risk) for two consecutive months, the model will be iterated and the feature weights will be updated again (such as adding the impact of the purchase delay feature) to correct the inventory quota range prediction model for the next year.

[0064] For example, if the actual inventory is below the lower limit of the range for two consecutive months (risk of material shortage), the feature weights of procurement cycle stability and supplier on-time delivery rate can be increased. The specific adjustment is as follows: the weight of procurement cycle stability is increased by 50% (e.g., the original weight 0.2→0.3), and the weight of supplier on-time delivery rate is increased by 30% (e.g., the original weight 0.15→0.195); at the same time, the weight of inventory turnover rate is reduced by 20% (e.g., the original weight 0.1→0.08) to reduce the suppressive effect of "high turnover" on inventory quotas.

[0065] Optionally, a "Purchase Delay Frequency" feature (the percentage of delayed orders in the past year) can be added to the evaluation feature matrix, with a weight of 0.15. When retraining the random forest model, the weight of delayed sample purchases can be increased (the weight of the delayed order sample is multiplied by 1.2), the lower limit calculation logic can be optimized, and the baseline value of the α coefficient can be increased by 0.05 (e.g., from 0.1 to 0.15), widening the gap between the lower limit and the optimal value and improving inventory assurance capabilities. After increasing the weights of procurement cycle stability and on-time delivery rate, the model will focus more on the impact of supplier fulfillment capabilities on inventory. When the supplier delivery risk is high, the lower limit of assurance will be automatically increased to reduce the risk of shortages. Adding the purchase delay feature can enhance the model's perception of delivery uncertainty and improve prediction robustness.

[0066] Optionally, the group's business management department can determine and officially issue annual inventory quota targets for each subordinate unit based on a defined inventory quota range. It can also collect real-time dynamic inventory amounts from each organization and automatically compare them with the pre-set quota targets. If the inventory amount exceeds the quota range, an alert will be automatically triggered and pushed to the group's management unit, allowing managers to promptly identify and follow up on anomalies. At the end of the year, the group will conduct a special performance evaluation of each subordinate unit's inventory control based on the pre-set inventory quotas. If the actual inventory amount exceeds the quota standard, the corresponding unit will be evaluated according to relevant evaluation rules, thereby achieving monitoring and early warning of the material inventory status of target organizations based on inventory quota ranges.

[0067] The technical solution of this invention, in response to a forecast request for an inventory quota range, determines the target material data of the target materials in the target forecast organization based on the forecast period and the target forecast organization; determines the evaluation feature matrix corresponding to the target material data, and predicts the optimal theoretical inventory value corresponding to the target materials based on a preset random forest model, according to the target material data and the evaluation feature matrix; determines the material importance weight corresponding to the target materials, and determines the inventory quota range of the target materials based on the material importance weight and the optimal theoretical inventory value, so as to monitor and warn of the material inventory status of the target forecast organization based on the inventory quota range. By comprehensively evaluating the material data and combining it with the random forest model to determine an accurate inventory quota range, it can help achieve effective and comprehensive inventory management.

[0068] Example 2 Figure 2 This is a flowchart of an inventory quota range prediction method provided by an embodiment of the present invention. Based on the above embodiments, this embodiment provides a preferred example that comprehensively evaluates material data and combines it with a random forest model to determine an accurate inventory quota range, thereby contributing to effective and comprehensive inventory management. Figure 2 As shown, the method includes: S201. Based on the preset collection cycle, obtain the management material data of each management organization from the material management system of the management organization through the preset data collection tool.

[0069] S202. Determine the organizational identifier, business type, and organizational level of each management organization, and perform anomaly removal and missing data supplementation on the management material data of each management organization.

[0070] S203. Based on the processed management material data, organization identifier, business type and organizational level, generate a material information table, a procurement data table, a consumption data table and an inventory data table, and store them in the target database using a table-splitting storage method.

[0071] S204. In response to a forecast request for an inventory quota range, determine the target material data of the target material in the target forecast organization from the target database, based on the forecast period and the target forecast organization.

[0072] S205. Determine the evaluation feature matrix corresponding to the target material data, and based on the preset random forest model, predict the optimal theoretical inventory value corresponding to the target material according to the target material data and the evaluation feature matrix.

[0073] S206. The first judgment result is to determine whether the target material belongs to the critical spare parts and the second judgment result is to determine whether the target material belongs to the bulk consumables.

[0074] S207. Based on the first judgment result, the second judgment result, and the preset adjustment coefficient, determine the material importance weight corresponding to the target material.

[0075] S208. Determine the lower limit of the target material based on the material importance weight, the preset supply guarantee gap coefficient, and the optimal inventory theoretical value corresponding to the target material.

[0076] S209. Determine the upper limit of the target material based on the preset cost smoothing slack coefficient, the procurement cost volatility, and the optimal inventory theoretical value corresponding to the target material.

[0077] S210. Determine the optimal value based on the upper and lower limits, and determine the inventory quota range of the target material based on the upper, lower, and optimal values, so as to monitor and warn of the material inventory status of the target forecast organization based on the inventory quota range.

[0078] Example 3 Figure 3This is a structural block diagram of an inventory quota range prediction device provided in an embodiment of the present invention. This embodiment is applicable to situations where comprehensive evaluation of material data is performed and an accurate inventory quota range is determined by combining a random forest model, thereby contributing to effective and comprehensive inventory management. The inventory quota range prediction device provided by the present invention can execute the inventory quota range prediction method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of executing the method. This inventory quota range prediction device can be implemented in hardware and / or software and configured in an electronic device with inventory quota range prediction function, such as... Figure 3 As shown, the inventory quota range forecasting device may specifically include: The data determination module 301 is used to respond to the forecast request of the inventory quota range and determine the target material data of the target material in the target forecast organization according to the forecast period and the target forecast organization. The prediction module 302 is used to determine the evaluation feature matrix corresponding to the target material data, and based on the preset random forest model, predict the optimal theoretical inventory value corresponding to the target material according to the target material data and the evaluation feature matrix. The interval determination module 303 is used to determine the material importance weight corresponding to the target material, and to determine the inventory quota range of the target material based on the material importance weight and the optimal inventory theoretical value, so as to monitor and warn the material inventory status of the target forecast organization based on the inventory quota range.

[0079] The technical solution of this invention, in response to a forecast request for an inventory quota range, determines the target material data of the target materials in the target forecast organization based on the forecast period and the target forecast organization; determines the evaluation feature matrix corresponding to the target material data, and predicts the optimal theoretical inventory value corresponding to the target materials based on a preset random forest model, according to the target material data and the evaluation feature matrix; determines the material importance weight corresponding to the target materials, and determines the inventory quota range of the target materials based on the material importance weight and the optimal theoretical inventory value, so as to monitor and warn of the material inventory status of the target forecast organization based on the inventory quota range. By comprehensively evaluating the material data and combining it with the random forest model to determine an accurate inventory quota range, it can help achieve effective and comprehensive inventory management.

[0080] Furthermore, the interval determination module 303 is specifically used for: The first judgment result is to determine whether the target material belongs to the critical spare parts, and the second judgment result is to determine whether the target material belongs to the bulk consumables. Based on the first judgment result, the second judgment result, and the preset adjustment coefficient, determine the material importance weight corresponding to the target material.

[0081] Furthermore, the interval determination module 303 is specifically used for: The lower limit of the target material is determined based on the material importance weight, the preset supply guarantee gap coefficient, and the optimal inventory theoretical value corresponding to the target material. The upper limit of the target material is determined based on the preset cost smoothing slack coefficient, the procurement cost volatility, and the optimal inventory theoretical value corresponding to the target material. The optimal value is determined based on the upper and lower limits, and the inventory quota range for the target material is determined based on the upper, lower, and optimal values.

[0082] Furthermore, the prediction module 302 is specifically used for: For each target material, data evaluation is conducted based on the corresponding target procurement data, target consumption data, and target inventory data to determine the evaluation index value; Based on the determined evaluation index values, an evaluation feature matrix is ​​generated; the evaluation index values ​​include at least one of the following: consumption volatility, monthly peak-to-valley difference, consumption trend slope, procurement cycle stability, on-time delivery rate, procurement cost volatility, inventory turnover rate, and obsolete inventory ratio.

[0083] Furthermore, the above-mentioned device is also used for: Based on a preset collection cycle, the management material data of each management organization is obtained from the management organization's material management system through a preset data collection tool, and the organization identifier, business type and organizational level of each management organization are determined. The system performs anomaly removal and missing data supplementation on the management material data of each management organization. Based on the processed management material data, organization identifier, business type, and organizational level, it generates material information table, procurement data table, consumption data table, and inventory data table, which are then stored in the target database using a table-splitting storage method.

[0084] Furthermore, the above-mentioned device is also used for: For each management organization's material data, the target procurement quantity for each target material is determined based on the procurement data, and the procurement data is used to determine whether there are any quantity anomalies based on the target procurement quantity, the historical maximum regular procurement quantity, and the historical average procurement quantity. The unit price of each target material is determined based on the procurement data, and the procurement data is used to determine whether there are any price anomalies based on the unit price, the reasonable range of market unit prices, and the lower limit of supplier quotations. Based on the consumption data, determine the daily consumption of each target material, and based on the correlation between the monthly average total consumption and the daily consumption of the target materials, determine whether there is any sudden abnormal consumption. If abnormalities are found in quantity, price, or sudden abnormal consumption, anomaly removal procedures will be performed.

[0085] Furthermore, the above-mentioned device is also used for: The missing inventory data in the managed materials data is supplemented based on the average of adjacent dates. Based on the strategy of supplementing the data with the average value of similar materials in the same period, the missing data of procurement materials and consumption materials in the managed materials data are supplemented.

[0086] Example 4 Figure 4 This is a schematic diagram of the structure of the electronic device provided in an embodiment of the present invention. Figure 4 A schematic diagram of an electronic device 10, which can be used to implement embodiments of the present invention, is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0087] like Figure 4 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory 12 or a random access memory 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the read-only memory 12 or loaded from storage unit 18 into the random access memory 13. The random access memory 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, read-only memory 12, and random access memory 13 are interconnected via a bus 14. An input / output interface 15 is also connected to the bus 14.

[0088] Multiple components in electronic device 10 are connected to input / output 15, including: input unit 16, such as a keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as a disk, optical disk, etc.; and communication unit 19, such as a network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0089] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, central processing units, graphics processing units, various special-purpose artificial intelligence computing chips, various processors running machine learning model algorithms, digital signal processors, and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as inventory quota range forecasting methods.

[0090] In some embodiments, the inventory quota range forecasting method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via read-only memory 12 and / or communication unit 19. When the computer program is loaded into random access memory 13 and executed by processor 11, one or more steps of the inventory quota range forecasting method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the inventory quota range forecasting method by any other suitable means (e.g., by means of firmware).

[0091] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays, application-specific integrated circuits (ASICs), application-specific standard products (ASICs), system-on-a-chip (SoCs), complex programmable logic devices (PLCs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0092] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0093] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory, optical fibers, portable compact disk read-only memory, optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0094] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a cathode ray tube or liquid crystal display) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (e.g., voice input, speech input, or tactile input).

[0095] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0096] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product within the cloud computing service system to address the shortcomings of traditional physical hosts and virtual reality services, such as high management difficulty and weak business scalability.

[0097] In one embodiment, the present invention further includes a computer program product, which includes a computer program that, when executed by a processor, implements the inventory quota range prediction method of any embodiment of the present invention.

[0098] In the implementation of a computer program product, computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​as well as conventional procedural programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including local area networks (LANs) or wide area networks (WANs), or it can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0099] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0100] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for predicting inventory quota ranges, characterized in that, include: In response to a forecast request for an inventory quota range, the target material data for the target materials in the target forecast organization is determined based on the forecast period and the target forecast organization. Determine the evaluation feature matrix corresponding to the target material data, and based on the preset random forest model, predict the optimal theoretical inventory value corresponding to the target material according to the target material data and the evaluation feature matrix; Determine the material importance weight corresponding to the target material, and based on the material importance weight and the optimal inventory theoretical value, determine the inventory quota range of the target material, so as to monitor and warn of the material inventory status of the target forecast organization based on the inventory quota range.

2. The method according to claim 1, characterized in that, Determine the material importance weight corresponding to the target material, including: The first judgment result is to determine whether the target material belongs to the critical spare parts, and the second judgment result is to determine whether the target material belongs to the bulk consumables. Based on the first judgment result, the second judgment result, and the preset adjustment coefficient, determine the material importance weight corresponding to the target material.

3. The method according to claim 1, characterized in that, Based on the importance weight of materials and the theoretical value of optimal inventory, determine the inventory quota range for the target materials, including: The lower limit of the target material is determined based on the material importance weight, the preset supply guarantee gap coefficient, and the optimal inventory theoretical value corresponding to the target material. The upper limit of the target material is determined based on the preset cost smoothing slack coefficient, the procurement cost volatility, and the optimal inventory theoretical value corresponding to the target material. The optimal value is determined based on the upper and lower limits, and the inventory quota range for the target material is determined based on the upper, lower, and optimal values.

4. The method according to claim 1, characterized in that, Determine the evaluation feature matrix corresponding to the target material data, including: For each target material, data evaluation is conducted based on the corresponding target procurement data, target consumption data, and target inventory data to determine the evaluation index value; Based on the determined evaluation index values, an evaluation feature matrix is ​​generated; the evaluation index values ​​include at least one of the following: consumption volatility, monthly peak-to-valley difference, consumption trend slope, procurement cycle stability, on-time delivery rate, procurement cost volatility, inventory turnover rate, and obsolete inventory ratio.

5. The method according to claim 1, characterized in that, Before determining the target material data for the target materials within the target forecasting organization, based on the forecast period and the target forecasting organization, the following steps are also included: Based on a preset collection cycle, the management material data of each management organization is obtained from the management organization's material management system through a preset data collection tool, and the organization identifier, business type and organizational level of each management organization are determined. The system performs anomaly removal and missing data supplementation on the management material data of each management organization. Based on the processed management material data, organization identifier, business type, and organizational level, it generates material information table, procurement data table, consumption data table, and inventory data table, which are then stored in the target database using a table-splitting storage method.

6. The method according to claim 5, characterized in that, Anomaly removal processing was performed on the management material data of each management organization, including: For each management organization's material data, the target procurement quantity for each target material is determined based on the procurement data, and the procurement data is used to determine whether there are any quantity anomalies based on the target procurement quantity, the historical maximum regular procurement quantity, and the historical average procurement quantity. The unit price of each target material is determined based on the procurement data, and the procurement data is used to determine whether there are any price anomalies based on the unit price, the reasonable range of market unit prices, and the lower limit of supplier quotations. Based on the consumption data, determine the daily consumption of each target material, and based on the correlation between the monthly average total consumption and the daily consumption of the target materials, determine whether there is any sudden abnormal consumption. If abnormalities are found in quantity, price, or sudden abnormal consumption, anomaly removal procedures will be performed.

7. The method according to claim 5, characterized in that, Missing data on management materials for each management organization were supplemented, including: The missing inventory data in the managed materials data is supplemented based on the average of adjacent dates. Based on the strategy of supplementing the data with the average value of similar materials in the same period, the missing data of procurement materials and consumption materials in the managed materials data are supplemented.

8. An inventory quota range prediction device, characterized in that, include: The data determination module is used to respond to the forecast request for the inventory quota range and determine the target material data of the target material in the target forecast organization based on the forecast period and the target forecast organization. The prediction module is used to determine the evaluation feature matrix corresponding to the target material data, and based on the preset random forest model, predict the optimal theoretical inventory value of the target material according to the target material data and the evaluation feature matrix. The interval determination module is used to determine the material importance weight corresponding to the target material, and to determine the inventory quota range of the target material based on the material importance weight and the optimal inventory theoretical value, so as to monitor and warn the material inventory status of the target forecast organization based on the inventory quota range.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the inventory quota range forecasting method according to any one of claims 1-7.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the inventory quota range forecasting method according to any one of claims 1-7.

Images