Commodity intelligent purchase order generation method, related device and storage medium

Abstract

The invention discloses a commodity intelligent purchase list generation method, a related device and a storage medium. The method comprises the following steps: collecting current commodity data of each commodity from each business system; analyzing current commodity feature data of each commodity based on the current commodity data; inputting the current commodity feature data of each commodity into a prediction model corresponding to the type to which the current commodity feature data belongs, and performing prediction through the prediction model to obtain sales prediction data of each commodity; calculating the safe inventory of each commodity by using the sales prediction data of each commodity and the inventory data of each commodity; according to the sales volume prediction data of each kind of commodities and the safe inventory of each kind of commodities, determining the purchase quantity of each kind of commodities; generating a commodity purchasing suggestion list based on the purchasing quantity of each commodity and the current commodity supply information of each supplier; and generating a commodity purchase order according to the commodity purchase suggestion order.

Description

Technical Field

[0001] This application relates to the technical field of commodity data analysis, and particularly relates to a method for generating an intelligent purchase order for commodities, related devices, and storage media. Background Art

[0002] In the process of e-commerce operation, a large number of commodity purchases and sales are involved, and the management of commodity purchases and sales is a crucial link. Therefore, it is necessary to analyze the relevant data of commodities to determine the corresponding purchase data and generate corresponding commodity purchase orders to ensure the normal sales of commodities.

[0003] Currently, in order to generate commodity purchase orders, mainly the recent sales data, order data, and inventory of commodities are collected, and relevant personnel, based on experience, use data statistical analysis methods to statistically analyze these data to obtain the recent sales situation and order data, so as to determine the future demand for each commodity, and then determine the purchase quantity of the commodity, and generate a purchase order based on the analyzed purchase quantity of each commodity to purchase commodities according to the purchase order.

[0004] However, this analysis method not only has low efficiency, but also is based on the recent sales situation and experience, and obtains the analysis results and generates a purchase order through simple statistical analysis, so it cannot effectively guarantee the accuracy of the results, and thus easily leads to overstocking or out-of-stock of commodities, affecting normal operations. Summary of the Invention

[0005] Based on the above deficiencies of the prior art, this application provides a method for generating an intelligent purchase order for commodities, related devices, and storage media to solve the problems of low efficiency and inability to guarantee accuracy in the prior art.

[0006] To achieve the above object, this application provides the following technical solutions:

[0007] The first aspect of this application provides a method for generating an intelligent purchase order for commodities, including:

[0008] Collecting the current commodity data of each commodity from various business systems;

[0009] Analyzing the current commodity feature data of each commodity based on the current commodity data;

[0010] Inputting the current commodity feature data of each commodity into the prediction model corresponding to its type, and predicting the sales prediction data of each commodity through the prediction model;

[0011] Calculating the safety inventory of each commodity by using the sales prediction data of each commodity and the inventory data of each commodity;

[0012] Determine the purchase quantity of each commodity according to the sales volume prediction data of each commodity and the safety inventory of each commodity;

[0013] Generate a commodity purchase recommendation form based on the purchase quantity of each commodity and the current commodity supply information of each supplier;

[0014] Generate a commodity purchase order according to the commodity purchase recommendation form.

[0015] Optionally, in the above method for generating a commodity intelligent purchase order, after collecting the current commodity data of each commodity from each business system, it further includes:

[0016] Preprocess the current commodity data of each commodity according to a preset data processing method.

[0017] Optionally, in the above method for generating a commodity intelligent purchase order, the calculating the safety inventory of each commodity by using the sales volume prediction data of each commodity and the inventory data of each commodity includes:

[0018] Based on the sales volume prediction data of each commodity, determine the standard deviation of the demand of the commodity within the predicted time period;

[0019] Calculate the safety inventory quantity of each commodity respectively through the supply chain delivery cycle of each commodity, the standard normal distribution value corresponding to the service level, and the standard deviation of the demand.

[0020] Optionally, in the above method for generating a commodity intelligent purchase order, the determining the purchase quantity of each commodity according to the sales volume prediction data of each commodity and the safety inventory of each commodity includes:

[0021] Calculate the reorder point of each commodity respectively by using the safety inventory of each commodity, the sales volume prediction data of each commodity, and the supply chain delivery cycle of each commodity;

[0022] Calculate the optimal economic order quantity of the reorder point of each commodity according to the sales volume prediction data of each commodity and the order cost of each commodity;

[0023] Determine the purchase quantity of each commodity based on the optimal economic order quantity of the reorder point of each commodity.

[0024] Optionally, in the above method for generating a commodity intelligent purchase order, it further includes:

[0025] Analyze the value level of each commodity by using the sales volume prediction data of each commodity or the inventory data of each commodity;

[0026] Based on the sales volume prediction data of each of the said commodities and the current inventory of each of the said commodities, determine the total sales amount, the beginning inventory, and the ending inventory of each of the said commodities within the prediction time period;

[0027] Respectively use the total sales amount, the beginning inventory, and the ending inventory of each of the said commodities within the prediction time period to calculate the turnover rate of each commodity;

[0028] Among them, the optimal economic order quantity based on the reorder point of each of the said commodities is determined as the purchase quantity of each of the said commodities, including:

[0029] Determine the purchase quantity of each of the said commodities based on the optimal economic order quantity, value level, and turnover rate of the reorder point of each of the said commodities.

[0030] Optionally, in the above method for generating a commodity intelligent purchase order, the generating of a commodity purchase recommendation order based on the purchase quantity of each of the said commodities and the current commodity supply information of each supplier includes:

[0031] For each of the said commodities, respectively calculate the comprehensive scores of each of the suppliers according to the current delivery cycle, current supply capacity value, current application price, and supply phenotype information of each of the suppliers of the commodity;

[0032] According to the comprehensive scores of each of the suppliers and the purchase quantity of the commodity, select the current supplier of the commodity and allocate the purchase quantity of the commodity for each of the current suppliers of the commodity;

[0033] Generate a purchase recommendation order for each of the said commodities according to the current supply information and purchase quantity of each of the current suppliers of each of the said commodities.

[0034] Optionally, in the above method for generating a commodity intelligent purchase order, after generating the commodity purchase order according to the commodity purchase recommendation order, it further includes:

[0035] Respond to the user's modification operation to modify the information in the commodity purchase order;

[0036] When the user confirms the commodity purchase order, perform approval processing on the commodity purchase order through an automated approval process;

[0037] When the commodity purchase order passes the approval, send the commodity purchase order to each supplier.

[0038] The second aspect of the present application provides a device for generating a commodity intelligent purchase order, including:

[0039] A data acquisition unit for acquiring the current product data of each product from various business systems;

[0040] A feature analysis unit for analyzing the current product feature data of each product based on the current product data;

[0041] A sales volume prediction unit for inputting the current product feature data of each product into the prediction model corresponding to its type, and predicting the sales volume prediction data of each product through the prediction model;

[0042] A safety stock analysis unit for calculating the safety stock quantity of each product by using the sales volume prediction data of each product and the inventory data of each product;

[0043] A purchase quantity determination unit for determining the purchase quantity of each product according to the sales volume prediction data of each product and the safety stock quantity of each product;

[0044] A recommendation form generation unit for generating a product purchase recommendation form based on the purchase quantity of each product and the current product supply information of each supplier;

[0045] A purchase order generation unit for generating a product purchase order according to the product purchase recommendation form.

[0046] Optionally, in the above product intelligent purchase order generation device, it further includes:

[0047] A preprocessing unit for preprocessing the current product data of each product according to a preset data processing method.

[0048] Optionally, in the above product intelligent purchase order generation device, the safety stock analysis unit includes:

[0049] A standard deviation determination unit for determining the standard deviation of the demand of the product within the prediction time period based on the sales volume prediction data of each product;

[0050] A safety stock quantity determination unit for calculating the safety stock quantity of each product respectively through the supply chain delivery cycle of each product, the standard normal distribution value corresponding to the service level, and the standard deviation of the demand.

[0051] Optionally, in the above product intelligent purchase order generation device, the purchase quantity determination unit includes:

[0052] A reorder point determination unit for calculating the reorder point of each product by using the safety stock quantity of each product, the sales volume prediction data of each product, and the supply chain delivery cycle of each product respectively;

[0053] An order quantity calculation unit for calculating the optimal economic order quantity of the reorder point for each commodity according to the sales forecast data of each commodity and the order cost of each commodity.

[0054] A purchase quantity calculation unit for determining the purchase quantity for each commodity based on the optimal economic order quantity of the reorder point for each commodity.

[0055] Optionally, in the above-mentioned device for generating an intelligent purchase order for commodities, it further includes:

[0056] A level analysis unit for analyzing the value level of each commodity by using the sales forecast data of each commodity or the inventory data of each commodity.

[0057] An index determination unit for determining the total sales amount, the beginning inventory quantity, and the ending inventory quantity of each commodity within the forecast time period based on the sales forecast data of each commodity and the current inventory quantity of each commodity.

[0058] A turnover rate calculation unit for calculating the turnover rate of each commodity by using the total sales amount, the beginning inventory quantity, and the ending inventory quantity of each commodity within the forecast time period respectively.

[0059] Among them, the purchase quantity calculation unit includes:

[0060] The purchase quantity calculation subunit for determining the purchase quantity for each commodity based on the optimal economic order quantity of the reorder point for each commodity, the value level, and the turnover rate.

[0061] Optionally, in the above-mentioned device for generating an intelligent purchase order for commodities, the recommendation order generation unit includes:

[0062] A score calculation unit for calculating the comprehensive score of each supplier for each commodity according to the current delivery cycle, the current supply capacity value, the current application price, and the supply phenotype information of each supplier of the commodity.

[0063] A selection unit for selecting the current supplier of the commodity and allocating the commodity purchase quantity of each current supplier of the commodity according to the comprehensive score of each supplier and the purchase quantity of the commodity.

[0064] A purchase recommendation order generation unit for generating a purchase recommendation order for each commodity according to the current supply information of each current supplier of each commodity and the commodity purchase quantity.

[0065] Optionally, in the above-mentioned device for generating an intelligent purchase order for commodities, it further includes:

[0066] A modification unit, configured to modify the information in the commodity purchase order in response to a user's modification operation;

[0067] An approval unit, configured to, after the user confirms the commodity purchase order, perform an approval process on the commodity purchase order through an automated approval process;

[0068] A sending unit, configured to, when the commodity purchase order passes the approval, send the commodity purchase order to each supplier.

[0069] A third aspect of the present application provides an electronic device, including:

[0070] A memory and a processor;

[0071] Wherein, the memory is used to store a program;

[0072] The processor is configured to execute the program, and when the program is executed, it is specifically configured to implement the method for generating a smart commodity purchase order as described in any one of the above.

[0073] A fourth aspect of the present application provides a computer storage medium, configured to store a computer program, and when the computer program is executed by a processor, it is used to implement the method for generating a smart commodity purchase order as described in any one of the above.

[0074] The present application provides a method for generating a smart commodity purchase order. Current commodity data of each commodity is collected from each business system. Then, based on the current commodity data analysis, the current commodity feature data of each commodity is obtained, so as to facilitate subsequent analysis through a model. Then, the current commodity feature data of each commodity is input into the prediction model corresponding to its type, and the sales volume prediction data of each commodity is obtained through the prediction model, so as to accurately predict the future sales volume of the commodity through the prediction model, and use it to analyze the purchase quantity of the commodity, rather than using the historical sales situation as the future sales volume for analyzing the purchase quantity. Then, using the sales volume prediction data of each commodity and the inventory data of each commodity, the safety inventory of each commodity is calculated, so as to determine the inventory quantity that can ensure sales and will not cause inventory backlog. Therefore, the purchase quantity of each commodity can be determined according to the sales volume prediction data of each commodity and the safety inventory of each commodity. Finally, based on the purchase quantity of each commodity and the current commodity supply information of each supplier, a commodity purchase recommendation form is generated, and a commodity purchase order is generated according to the commodity purchase recommendation form. Thus, accurate analysis of commodity information is realized, accurate future sales volume is obtained, and further accurate purchase quantity is analyzed, which can not only ensure sales but also avoid backlog. And finally, the order is automatically generated according to the analyzed information, which can not only ensure the accuracy of the generated purchase order but also improve the efficiency. Description of the Drawings

[0075] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only the embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on the provided drawings.

[0076] Figure 1 It is a flowchart of a method for generating an intelligent purchase order for goods provided by an embodiment of the present application;

[0077] Figure 2 It is a flowchart of a method for calculating the safety stock quantity of goods provided by an embodiment of the present application;

[0078] Figure 3 It is a flowchart of a method for determining the purchase quantity of goods provided by an embodiment of the present application;

[0079] Figure 4 It is a flowchart of a method for analyzing inventory-related indicators of goods provided by an embodiment of the present application;

[0080] Figure 5 It is a flowchart of a method for generating a purchase advice order for goods provided by an embodiment of the present application;

[0081] Figure 6 It is a schematic diagram of the architecture of a device for generating an intelligent purchase order for goods provided by an embodiment of the present application;

[0082] Figure 7 It is a schematic diagram of the architecture of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0083] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0084] In this application, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprising", "including" or any other variant thereof are intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element.

[0085] An embodiment of this application provides a method for generating an intelligent purchase order for commodities, as Figure 1 shown, including the following steps:

[0086] S101. Collect the current commodity data of each commodity from various business systems.

[0087] Among them, the current commodity data refers to the specified data related to the commodity currently, for example, commodity basic information, commodity inventory, order quantity, historical sales volume, etc. The business systems are each business system involved in commodity operation, for example, ERP system, order management system (OMS), warehouse management system (WMS), etc.

[0088] Optionally, specifically, the current data of various commodities can be obtained in real time from each business system through the interface of the business system and stored in the system database. Optionally, a distributed architecture can be adopted for data collection to ensure the real-time nature and high availability of data collection. For data collection in a high-concurrency environment, message queues (such as RabbitMQ, KafKa) etc. can be used to cache real-time data. Specifically, a timed polling method can be adopted to pull the latest data regularly.

[0089] Optionally, the data formats of different data sources may be different, so the system needs to perform preliminary format conversion when collecting data, unify the data into the standard JSON format, and when performing data verification and cleaning, adopt the ETL process to extract, transform and load the collected data. If there are duplicate or conflicting data situations, data deduplication and consistency processing etc. are required to ensure data consistency and quality. And during the data transmission process, in order to prevent data from being stolen or tampered with during network transmission, the data can be encrypted and then the encrypted data is transmitted. For example, the SSL protocol is used to encrypt the data.

[0090] Optionally, in order to improve the quality of the collected data and facilitate subsequent data analysis, in another embodiment of the present application, after performing step S101 and before performing step S102, the following steps may be further performed:

[0091] Preprocess the current commodity data of each commodity according to a preset data processing method.

[0092] Optionally, specifically, the collected data can be cleaned, format-converted, and standardized. By removing noise, outliers, filling missing values, processing duplicate data, data augmentation, etc., the consistency and reliability of the data can be improved, thereby converting the original data into high-quality input data.

[0093] Optionally, for the detection and processing of outliers, statistical indicators such as mean, standard deviation, or quartiles can be used to detect outliers. For extremely abnormal and meaningless data points detected, they can be directly deleted. For relatively common outliers, the median or mean can be used to replace the outliers. Slightly abnormal values can be smoothed by the moving average method.

[0094] For the processing of duplicate data, after multi-source data fusion, deduplication can be performed through the primary key or other unique identifiers to ensure the uniqueness of each record. For records that may be duplicates but have slightly different contents, they can be merged through data aggregation. For data consistency checking, the consistency of the same data field in different data sources can be checked to ensure the consistency of the formats of fields such as dates, amounts, and quantities. For missing values, specifically, indirect missing values are identified through conditional judgment, and the missing values in the identified data are directly displayed as NA or NAN. Then, for the identified missing values, the mean, median, mode, etc. are specifically used to fill the missing values. For missing values in time series data, linear interpolation methods are specifically used to fill the missing values. For data format conversion, the original data can be converted into a format and structure suitable for model processing. mainly data type conversion. For example, all numerical fields are converted into a consistent type (such as integer or floating point). For time series conversion, specifically, the date and time are converted into timestamps or standardized time formats, and according to the prediction requirements, the time series data can be divided into different time windows such as days, weeks, and months.

[0095] Furthermore, the data can also be normalized and standardized. Among them, the normalization process can be to scale the numerical data to the range of [0,1], and specifically, it can be processed through the normalization formula. For the standardization process, specifically, the numerical values can be adjusted to a standard overall distribution with a mean of 0 and a standard deviation of 1.

[0096] S102. Analyze the current commodity feature data of each commodity based on the current commodity data.

[0097] It should be noted that since some of the original data cannot be directly used for model prediction, and in order to further optimize the input data and improve the performance of the model, further statistical analysis or extraction is required. Therefore, it is necessary to analyze the current product feature data of each product based on the current product data. Specifically, according to the required features, corresponding methods are used for feature analysis.

[0098] Optionally, for statistical features, the average value, variance, maximum value, minimum value, kurtosis, skewness and other statistical features can be specifically extracted from the sales data in the current product data. For time features, they can be extracted from the date and time fields. For example, time features such as day of the week, beginning and end of the month, season, etc. can be extracted. For interaction features, interaction terms between two or more relevant features can be extracted, such as the interaction effect between sales volume and promotional activities.

[0099] Optionally, after analyzing the current product feature data, feature selection can be further performed. Specifically, the correlation between the feature and the target variable, that is, the correlation with the sales volume, can be calculated to retain the features with higher correlation. And, the principal components can be analyzed through the principal component analysis method to extract the main information in the data, thereby reducing redundant features. For the selected features, feature interaction and feature combination can also be performed to capture non-linear relationships by creating polynomial combinations. Moreover, data augmentation can be used to generate new data to expand the dataset, especially when the amount of data is insufficient, it can provide sufficient data volume. Among them, for time series data, more data can be generated through methods such as jittering, scaling, and time offset. Among them, for time series, multiple samples can be generated through the sliding window technique to increase the amount of training data, and data interpolation can generate new data points through interpolation methods in continuous time series.

[0100] S103. Input the current product feature data of each product into the prediction model corresponding to its type, and obtain the sales volume prediction data of each product through the prediction model.

[0101] It should be noted that in order to accurately determine the future demand of the product, instead of using the current sales situation as the future sales situation. Therefore, in the embodiments of the present application, a prediction model is pre-trained, and then the future sales situation of the product is predicted through the prediction model.

[0102] Due to different scenarios and different products having different characteristics that affect their sales, the currently collected product characteristic data will vary. And since the characteristics of different currently collected product characteristic data are different, in order to adapt to the characteristics of the characteristic data and accurately predict sales volume, corresponding prediction models are constructed for different types of characteristic data, such as time series models, regression analysis, and machine learning models, so as to process the corresponding characteristic data and accurately predict the sales volume data. Therefore, when making predictions, it is necessary to input the currently collected product characteristic data of each product into the prediction model corresponding to its type, so as to predict the sales volume data of each product in a future period through the corresponding prediction model.

[0103] Optionally, the selected algorithm model can mainly use the ARIMA model (p, d, q) to process non-stationary data, that is, make the data stationary through the regression order (P), the number of differences (d), and the moving average order (q), and then use ARIMA for modeling. If the data fluctuates greatly with seasonality, the SARIMA model can be used to process it.

[0104] A linear regression model can also be used, that is, there is a linear relationship between the sales volume and multiple influencing factors. Among them, the expression of the linear regression model is:

[0105]

[0106] Among them, is the regression coefficient to be estimated. is the error term.

[0107] A machine learning algorithm model can also be used. For example, random forest, LSTM (long short-term memory network), etc. are used for prediction. Among them, random forest predicts the sales volume through the combination of multiple decision trees and has good anti-noise ability and non-linear fitting ability. LSTM processes the long-term dependence relationship of data time through deep learning of time series data and predicts data with long-term dependence relationships.

[0108] Specifically, a large amount of historical data can be used to form a training set to train the prediction models corresponding to each type. Among them, for the model belonging to the time series, the training set is generally continuous data in chronological order. For the machine learning model, the training set is the data processed through feature engineering. And during the process of model training, indicators such as mean squared error (MSE) and mean absolute error (MAE) can be used to evaluate the accuracy of the model, and the model parameters can be adjusted or feature selection can be carried out according to the verification results to continuously optimize the model performance. For the machine learning model, parameter tuning can also be carried out through methods such as cross-validation.

[0109] S104. Calculate the safety stock quantity for each commodity by using the sales volume prediction data and inventory data of each commodity.

[0110] Among them, the safety stock quantity of a commodity is the inventory quantity that will not affect the normal sales of the commodity and will not cause inventory backlog of the commodity. Therefore, purchasing according to the safety stock quantity of the commodity can ensure the normal sales of the commodity and will not cause inventory backlog of the commodity.

[0111] It should be noted that the sales volume prediction data of a commodity is the quantity that the commodity needs to be sold in the future. Therefore, based on the sales volume prediction data of the commodity, the future demand of the commodity can be estimated. Then, by combining the demand of the commodity with the current inventory, the safety stock quantity of the commodity can be determined.

[0112] Optionally, in another embodiment of the present application, a specific implementation manner of step S104 is as Figure 2 shown and includes the following steps:

[0113] S201. Based on the sales volume prediction data of each commodity, determine the standard deviation of the demand of the commodity within the prediction time period.

[0114] Among them, the prediction time period is the time period predicted by the sales volume prediction data.

[0115] Therefore, the standard deviation of the demand within the prediction time period can be specifically calculated by the following formula:

[0116]

[0117] Among them, N is the total number of days in the target future time period. D i is the demand on the i-th day, that is, the sales volume on the i-th day, so it is obtained from the sales volume prediction data of the commodity. D is the average demand.

[0118] S202. Calculate the safety stock quantity of each commodity respectively through the supply chain delivery cycle of each commodity, the standard normal distribution value corresponding to the service level, and the standard deviation of the demand.

[0119] Specifically, take the result of multiplying the standard normal distribution value corresponding to the service level, the standard deviation of the demand within the prediction time period, and the square root of the supply chain delivery cycle as the safety stock quantity of the commodity.

[0120] Among them, the standard normal distribution value Z corresponding to the service level can be obtained through the standard normal distribution table or the calculated standard score according to the set supplier service level. For example, when the service level is 90%, Z≈1.28; when the service level is 95%, Z≈1.645; when the service level is 99%, Z≈2.33. The supply chain delivery cycle is the time required from placing an order to delivering the goods to the warehouse, usually measured in days or weeks.

[0121] S105. Determine the purchase quantity of each commodity according to the sales forecast data of each commodity and the safety stock quantity of each commodity.

[0122] The safety stock quantity of a commodity is the quantity of the commodity's inventory that is safe within a future time period, and the sales forecast data is the sales volume within a future time period. Therefore, according to the sales forecast data of each commodity and the safety stock quantity of each commodity, it is possible to determine how many commodities need to be purchased to meet the sales volume and the safety stock quantity.

[0123] Optionally, in another embodiment of the present application, a specific implementation manner of step S105 is as Figure 3 shown, including the following steps:

[0124] S301. Calculate the reorder point of each commodity by using the safety stock quantity of each commodity, the sales forecast data of each commodity, and the supply chain delivery cycle of each commodity respectively.

[0125] Specifically, the daily average demand, that is, the daily average sales volume, in the safety stock quantity of the commodity, the supply chain delivery cycle of the commodity, and the sales forecast data of the commodity can be multiplied to obtain the reorder point of the commodity.

[0126] S302. Calculate the optimal economic order quantity of the reorder point of each commodity according to the sales forecast data of each commodity and the order cost of each commodity.

[0127] Specifically, the EOQ model can be used to calculate the quantity of the commodity that needs to be ordered to achieve the best economy at the reorder point. The specific calculation method is as follows:

[0128]

[0129] Among them, D is the total demand, that is, the total demand of the commodity within the forecast time period, so it can be specifically obtained according to the sales forecast data of the commodity. S is the fixed cost per order. H is the unit inventory holding cost, that is, the cost of storing the commodity in the warehouse for a specified time.

[0130] S303. Determine the purchase quantity of each commodity based on the optimal economic order quantity of the reorder point of each commodity.

[0131] The optimal economic order quantity is the order quantity of goods that can achieve the best economic benefits. Therefore, the procurement quantity of goods can be determined with reference to the optimal economic order quantity. Optionally, specifically, the procurement quantity of goods can be directly determined as the procurement quantity of goods. Considering that other factors also need to be taken into account in the procurement of goods, the procurement quantity of goods can be determined with reference to the procurement quantity of goods and other factor information.

[0132] Optionally, in another embodiment of the present application, in order to analyze a more accurate procurement quantity, without affecting normal sales and without causing inventory backlogs. Therefore, it can further include analyzing the inventory correlation indicators of goods. As Figure 4 shown, a method for analyzing inventory correlation indicators of a kind of goods provided by an embodiment of the present application includes:

[0133] S401. Analyze the value level of each kind of goods by using the sales forecast data of each kind of goods or the inventory data of each kind of goods.

[0134] It should be noted that the higher the value of the goods, the more necessary it is to ensure its inventory level, and the more goods need to be procured. Therefore, it affects the procurement quantity of goods. And the sales amount of goods can be analyzed through the sales forecast data of goods, so it can reflect the value of goods. And the value of goods is also reflected through the value of goods in inventory. Therefore, the value level of each kind of goods can be analyzed by using the sales forecast data of each kind of goods or the inventory data of each kind of goods.

[0135] Optionally, specifically, the ABC analysis method can be used. By analyzing the sales amount or inventory value of goods, the goods are classified into three categories: A (high value), B (medium value), and C (low value). Specifically, sort the goods in descending order according to the sales amount or inventory value of the goods, calculate the percentage of the cumulative sales amount or inventory value of each good in the total amount, and classify the goods into three categories: A, B, and C according to the cumulative percentage. For example, the top 20% of the goods are classified as category A, the next 30% as category B, and the remaining 50% as category C.

[0136] S402. Based on the sales forecast data of each kind of goods and the current inventory level of each kind of goods, determine the total sales amount, the beginning inventory level, and the ending inventory level of each kind of goods during the forecast time period.

[0137] Among them, the forecast time period is the time period of the sales forecast data of the forecasted goods.

[0138] It should be noted that the sales forecast data of the commodity can include the sales volume of each unit time within the target future time period. Therefore, the total sales amount of the commodity within the target future time period can be obtained by statistically analyzing the sales forecast data of the commodity and then combining it with the price of the commodity. And based on the current inventory of the commodity, the beginning inventory can be obtained, that is, the inventory at the starting time point of the forecast time period. And based on the current inventory of the commodity and the sales volume during the forecast time period, the ending inventory can be obtained, that is, the inventory at the last time point of the forecast time period.

[0139] S403. Calculate the turnover rate of each commodity by using the total sales amount, the beginning inventory, and the ending inventory of each commodity within the forecast time period.

[0140] Specifically, divide the total sales amount of the commodity within the target future time period by the sum of the beginning inventory and the ending inventory to obtain the turnover rate of the commodity.

[0141] Therefore, correspondingly, a specific implementation manner of step S303 includes:

[0142] Determine the purchase quantity for each commodity based on the optimal economic order quantity, value level, and turnover rate of the reorder point of each commodity.

[0143] That is, in the embodiments of the present application, it is necessary to comprehensively consider the optimal economic order quantity, value level, and turnover rate to determine the purchase quantity of the commodity. Optionally, specifically, for commodities with a relatively high value level and turnover rate, a relatively large quantity can be added to their optimal economic order quantity as the purchase quantity of the commodity to ensure the inventory and turnover of the commodity.

[0144] S106. Generate a commodity purchase recommendation form based on the purchase quantity of each commodity and the current commodity supply information of each supplier.

[0145] After determining the purchase quantity of each commodity, it is necessary to determine various information such as the suppliers of the purchased commodities, the quantity of commodities purchased from each supplier, and the purchase price. Therefore, it is necessary to generate specific purchase information based on the current commodity supply information of each supplier and the purchase quantity of each commodity. Optionally, the current commodity supply information can include the price, delivery cycle, and supply capacity of the commodities supplied by the supplier. Specifically, the suppliers can be screened according to the allocation rules and the purchase quantity of each supplier can be allocated.

[0146] Optionally, in another embodiment of the present application, a specific implementation manner of step S106 is as Figure 5 shown and includes the following steps:

[0147] S501. For each commodity, calculate the comprehensive score of each supplier according to the current delivery cycle, current supply capacity value, current application price, and supply phenotype information of each supplier of the commodity.

[0148] S502. According to the comprehensive scores of each supplier and the purchase quantity of the commodity, select the current supplier of the commodity and allocate the purchase quantity of the commodity to each current supplier.

[0149] Specifically, the higher the comprehensive score of the supplier, the higher the selection priority of the supplier. Therefore, according to the comprehensive scores of the suppliers, select one or more suppliers, and then allocate the purchase quantity of the commodity according to the proportion of the comprehensive score of each supplier in the total comprehensive score.

[0150] S503. Generate a purchase recommendation form for each commodity according to the current supply information and purchase quantity of each current supplier of each commodity.

[0151] Among them, the purchase recommendation form of the commodity may include information such as the type, quantity, unit price, supplier, and commodity classification of the commodity recommended for purchase.

[0152] Optionally, a total purchase recommendation form can be generated for all commodities to facilitate subsequent review and summary, etc. Of course, a purchase recommendation form can also be generated separately for each commodity.

[0153] S107. Generate a purchase order for the commodity according to the purchase recommendation form of the commodity.

[0154] It should be noted that after generating the purchase recommendation form, based on the purchase recommendation form, combined with supplier information and purchase strategies, data such as the purchase amount are summarized and information such as the purchase time is determined, and then these information are combined to generate a specific purchase order for the commodity, so as to obtain a standard purchase order for the commodity. Therefore, the generated purchase order for the commodity may include the commodity name, purchase quantity, unit price, total price, supplier information, estimated delivery date, etc.

[0155] Optionally, in another embodiment of the present application, after executing step S107, it may further include:

[0156] Respond to the user's modification operation to modify the information in the purchase order for the commodity, and when the user confirms the purchase order for the commodity, perform approval processing on the purchase order for the commodity through an automated approval process, and when the purchase order for the commodity passes the approval, send the purchase order for the commodity to each supplier.

[0157] To facilitate users to flexibly modify the information in the commodity purchase order according to current requirements, in the embodiment of the present application, the generated commodity purchase order is provided to the user for modification and confirmation. Moreover, for the convenience of auditing, after the user confirms, the commodity purchase order is submitted to the corresponding approver for approval through an automated approval process. After the commodity purchase order passes the approval process, the system will send the purchase order to the corresponding supplier.

[0158] The embodiment of the present application provides a method for generating an intelligent commodity purchase order, which collects the current commodity data of each commodity from various business systems. Then, based on the current commodity data, the current commodity feature data of each commodity is analyzed to facilitate subsequent analysis through a model. Next, the current commodity feature data of each commodity is input into the prediction model corresponding to its type, and the sales volume prediction data of each commodity is obtained through the prediction model, so as to accurately predict the future sales volume of the commodity through the prediction model, which is used to analyze the purchase quantity of the commodity, rather than using the historical sales situation as the future sales volume for analyzing the purchase quantity. Then, using the sales volume prediction data of each commodity and the inventory data of each commodity, the safety inventory of each commodity is calculated, so as to determine the inventory quantity that can ensure sales without causing inventory backlog. Therefore, the purchase quantity of each commodity can be determined according to the sales volume prediction data of each commodity and the safety inventory of each commodity. Finally, based on the purchase quantity of each commodity and the current commodity supply information of each supplier, a commodity purchase recommendation list is generated, and a commodity purchase order is generated according to the commodity purchase recommendation list. Thus, the accurate analysis of the commodity information is realized, the accurate future sales volume is obtained, and the accurate purchase quantity is further analyzed, which can not only ensure sales but also avoid backlog. Moreover, finally, the order is automatically generated according to the analyzed information, which can not only ensure the accuracy of the generated purchase order but also improve the efficiency.

[0159] Another embodiment of the present application provides an apparatus for generating an intelligent commodity purchase order, as Figure 6 shown, including:

[0160] A data acquisition unit 601, configured to collect the current commodity data of each commodity from various business systems.

[0161] A feature analysis unit 602, configured to analyze the current commodity feature data of each commodity based on the current commodity data.

[0162] A sales volume prediction unit 603, configured to input the current commodity feature data of each commodity into the prediction model corresponding to its type, and obtain the sales volume prediction data of each commodity through the prediction model.

[0163] The safety stock analysis unit 604 is used to calculate the safety stock quantity of each commodity by using the sales volume prediction data of each commodity and the inventory data of each commodity.

[0164] The purchase quantity determination unit 605 is used to determine the purchase quantity of each commodity according to the sales volume prediction data of each commodity and the safety stock quantity of each commodity.

[0165] The recommendation list generation unit 606 is used to generate a commodity purchase recommendation list based on the purchase quantity of each commodity and the current commodity supply information of each supplier.

[0166] The purchase order generation unit 607 is used to generate a commodity purchase order according to the commodity purchase recommendation list.

[0167] Optionally, in the commodity intelligent purchase order generation device provided in another embodiment of the present application, it further includes:

[0168] The preprocessing unit is used to preprocess the current commodity data of each commodity according to a preset data processing method.

[0169] Optionally, in the commodity intelligent purchase order generation device provided in another embodiment of the present application, the safety stock analysis unit includes:

[0170] The standard deviation determination unit is used to determine the standard deviation of the demand of the commodity within the prediction time period based on the sales volume prediction data of each commodity.

[0171] The safety stock quantity determination unit is used to calculate the safety stock quantity of each commodity respectively through the supply chain delivery cycle of each commodity, the standard normal distribution value corresponding to the service level, and the standard deviation of the demand.

[0172] Optionally, in the commodity intelligent purchase order generation device provided in another embodiment of the present application, the purchase quantity determination unit includes:

[0173] The reorder point determination unit is used to calculate the reorder point of each commodity respectively by using the safety stock quantity of each commodity, the sales volume prediction data of each commodity, and the supply chain delivery cycle of each commodity.

[0174] The order quantity calculation unit is used to calculate the optimal economic order quantity of the reorder point of each commodity according to the sales volume prediction data of each commodity and the order cost of each commodity.

[0175] The purchase quantity calculation unit is used to determine the purchase quantity of each commodity based on the optimal economic order quantity of the reorder point of each commodity.

[0176] Optionally, in the commodity intelligent purchase order generation device provided in another embodiment of the present application, it further includes:

[0177] A level analysis unit for analyzing the value level of each commodity by using the sales volume prediction data or inventory data of each commodity.

[0178] An index determination unit for determining the total sales amount, initial inventory quantity, and ending inventory quantity of each commodity within the prediction time period based on the sales volume prediction data of each commodity and the current inventory quantity of each commodity.

[0179] A turnover rate calculation unit for calculating the turnover rate of each commodity by using the total sales amount, initial inventory quantity, and ending inventory quantity of each commodity within the prediction time period respectively.

[0180] Among them, the purchase quantity calculation unit includes:

[0181] A purchase quantity calculation subunit for determining the purchase quantity of each commodity based on the optimal economic order quantity, value level, and turnover rate of the reorder point of each commodity.

[0182] Optionally, in the commodity intelligent purchase order generation device provided in another embodiment of the present application, the recommendation order generation unit includes:

[0183] A score calculation unit for calculating the comprehensive score of each supplier for each commodity according to the current delivery cycle, current supply capacity value, current application price, and supply phenotype information of each supplier of the commodity.

[0184] A selection unit for selecting the current supplier of the commodity and allocating the commodity purchase quantity of each current supplier of the commodity according to the comprehensive score of each supplier and the purchase quantity of the commodity.

[0185] A purchase recommendation order generation unit for generating a purchase recommendation order for each commodity according to the current supply information of each current supplier of each commodity and the commodity purchase quantity.

[0186] Optionally, in the commodity intelligent purchase order generation device provided in another embodiment of the present application, it further includes:

[0187] A modification unit for modifying the information in the commodity purchase order in response to the user's modification operation.

[0188] An approval unit for performing an approval process on the commodity purchase order through an automated approval process when the user confirms the commodity purchase order.

[0189] A sending unit for sending the commodity purchase order to each supplier when the commodity purchase order passes the approval.

[0190] It should be noted that for the specific working processes of the various units provided in the above embodiments of the present application, reference may be made correspondingly to the implementation processes of the corresponding steps in the above method embodiments, and details are not described herein again.

[0191] Another embodiment of the present application provides an electronic device, as Figure 7 shown, including:

[0192] a memory 701 and a processor 702.

[0193] Among them, the memory 701 is used to store programs.

[0194] The processor 702 is used to execute the programs stored in the memory 701. When the programs are executed, they are specifically used to implement the method for generating a commodity intelligent purchase order provided in any of the above embodiments.

[0195] Another embodiment of the present application provides a computer storage medium for storing computer programs. When the computer programs are executed by a processor, they are used to implement the method for generating a commodity intelligent purchase order provided in any of the above embodiments.

[0196] Computer storage media include both permanent and non-permanent, removable and non-removable media and can be implemented by any method or technology for information storage. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tapes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory media such as modulated data signals and carrier waves.

[0197] Those skilled in the art can further realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been generally described according to their functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Skilled professionals can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0198] The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for generating an intelligent purchase order of goods, characterized in that: include: Collect current product data for each product from various business systems; Analyze the current commodity feature data of each commodity based on the current commodity data; Inputting the current commodity feature data of each commodity into the prediction model corresponding to the type to which it belongs, and obtaining the sales forecast data of each commodity through prediction by the prediction model; Calculate the safety stock of each of the commodities using the sales forecast data and inventory data of each of the commodities; Determine the purchase quantity of each commodity according to the sales forecast data of each commodity and the safety inventory of each commodity; Generate a commodity purchase suggestion list based on the purchase quantity of each of the commodities and the current commodity supply information of each supplier; Generate a commodity purchase order according to the commodity purchase suggestion list.

2. The method according to claim 1, characterized in that After collecting the current commodity data of each commodity from each business system, the method further includes: The current commodity data of each of the commodities is pre-processed according to a preset data processing method.

3. The method according to claim 1, characterized in that The calculating the safety inventory of each commodity by using the sales forecast data of each commodity and the inventory data of each commodity includes: Based on the sales forecast data of each of the commodities, determining the standard deviation of the demand for the commodities within the forecast period; The safety stock quantity of each commodity is calculated by using the supply chain delivery cycle of each commodity, the standard normal distribution value corresponding to the service level and the standard deviation of the demand.

4. The method according to claim 1, characterized in that: Determining the purchase quantity of each commodity according to the sales forecast data of each commodity and the safety inventory of each commodity includes: The reorder point of each commodity is calculated by using the safety inventory of each commodity, the sales forecast data of each commodity and the supply chain delivery cycle of each commodity; Calculate the optimal economic order quantity for the reorder point of each of the commodities according to the sales forecast data of each of the commodities and the ordering cost of each of the commodities; The purchase quantity of each of the commodities is determined based on the optimal economic order quantity of the reorder point of each of the commodities.

5. The method according to claim 4, characterized in that Also includes: Analyzing the value level of each of the commodities using sales forecast data of each of the commodities or inventory data of each of the commodities; Based on the sales forecast data of each of the commodities and the current inventory of each of the commodities, determine the total sales, the beginning inventory and the ending inventory of each of the commodities in the forecast period; The turnover rate of each commodity is calculated by using the total sales, the beginning inventory and the ending inventory of each commodity in the forecast period; The optimal economic order quantity based on the reorder point of each of the commodities is determined as the purchase quantity of each of the commodities, including: The purchase quantity for each of the commodities is determined based on the optimal economic order quantity, value level and turnover rate of the reorder point of each of the commodities.

6. The method according to claim 1, characterized in that The method of generating a commodity purchase suggestion list based on the purchase quantity of each commodity and the current commodity supply information of each supplier includes: For each of the commodities, respectively, calculate a comprehensive score of each of the suppliers according to the current delivery cycle, current supply capacity value, current application price and supply phenotype information of each of the suppliers of the commodities; According to the comprehensive scores of each supplier and the purchase quantity of the commodity, the current supplier of the commodity is selected and the commodity purchase quantity of each current supplier of the commodity is allocated; A purchase suggestion list for each of the commodities is generated based on the current supply information of each current supplier of each of the commodities and the commodity purchase quantity.

7. The method according to claim 1, characterized in that After generating the commodity purchase order according to the commodity purchase suggestion list, the method further includes: In response to a modification operation by a user, modify the information in the commodity purchase order; After the user confirms the commodity purchase order, the commodity purchase order is approved through an automated approval process; When the commodity purchase order is approved, the commodity purchase order is sent to each supplier.

8. A device for generating an intelligent purchase order of goods, characterized in that: include: A data collection unit, used to collect current commodity data of each commodity from various business systems; A feature analysis unit, configured to analyze the current commodity feature data of each commodity based on the current commodity data; A sales forecasting unit, used to input the current commodity feature data of each commodity into a forecasting model corresponding to the type to which it belongs, and obtain sales forecast data of each commodity through forecasting by the forecasting model; A safety stock analysis unit, used to calculate the safety stock of each of the commodities by using the sales forecast data of each of the commodities and the inventory data of each of the commodities; A purchase quantity determination unit, used to determine the purchase quantity of each of the commodities according to the sales forecast data of each of the commodities and the safety inventory of each of the commodities; A suggestion sheet generating unit, used for generating a commodity purchase suggestion sheet based on the purchase quantity of each of the commodities and the current commodity supply information of each supplier; A purchase order generating unit is used to generate a commodity purchase order according to the commodity purchase suggestion list.

9. An electronic device, characterized in that: include: Memory and processor; Wherein, the memory is used to store programs; The processor is used to execute the program, and when the program is executed, it is specifically used to implement the method for generating an intelligent purchase order for goods as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that: Used to store a computer program, which, when executed by a processor, is used to implement the method for generating an intelligent purchase order for commodities as described in any one of claims 1 to 7.

Images