Purchase order management method and device for chain owner enterprise, and product
By using the SARIMA model and greedy algorithm in the purchase order management of chain owner enterprises, the problem of large deviation in the prediction of out-of-stock volume in the existing technology is solved, and more accurate out-of-stock volume prediction and actual out-of-stock volume reduction are achieved.
Patent Information
- Application Number
- CN202510050576.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-13
- Publication Date
- 2025-05-30
AI Technical Summary
The purchase order out-of-stock forecasting method used in the prior art for chain-owners has an underfit problem, resulting in large prediction deviations and affecting decision-making accuracy.
The SARIMA model is used in combination with the greedy algorithm, and by obtaining purchase order parameters and historical out-of-stock records, the SARIMA model is trained to predict out-of-stock volume of purchase orders, and to reduce the actual out-of-stock volume by adjusting the order parameters.
It improves the accuracy of forecasting out-of-stock volume of purchase orders, reduces actual out-of-stock volume, and enhances the procurement decision-making capabilities of chain owners.
Smart Images
Figure CN120069757A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the technical field of data processing, and particularly to a procurement order management method, device, readable storage medium, and product for the leading enterprise in the supply chain. Background Art
[0002] As the leading enterprise in the supply chain, its core responsibility is to monitor the inventory, production activities, and quality inspection of suppliers in real time. By deeply analyzing these key production information, the enterprise can dynamically evaluate the supply stability of suppliers, ensure the timeliness of supply, and the compliance of product quality. The purpose of doing this is to reduce procurement costs through refined management, and ultimately achieve the improvement of supply chain efficiency and the optimization of enterprise operating benefits. The existing methods for predicting the shortage quantity simply predict the future shortage quantity based on the shortage history through a neural network model. Therefore, there are often underfitting problems, resulting in a large deviation between the predicted shortage quantity and the actual shortage quantity, which may cause decision-making errors for the leading enterprise in the supply chain. Summary of the Invention
[0003] The purpose of the embodiments of this application is to provide a procurement order management method, device, readable storage medium, and product for the leading enterprise in the supply chain, so as to solve the technical problem of how to predict the shortage quantity of procurement orders in the prior art.
[0004] To achieve the above purpose, the first aspect of this application provides a procurement order management method for the leading enterprise in the supply chain, including:
[0005] Obtain the parameters of the procurement order to be predicted, where the parameters of the procurement order to be predicted include: the expected arrival time, the type of goods to be purchased, the identity of the supplier of the goods to be purchased, and the required quantity of the goods to be purchased;
[0006] According to the parameters of the procurement order to be predicted, determine the predicted shortage quantity of the procurement order to be predicted through the SARIMA model;
[0007] Adjust the procurement order to be predicted according to the predicted shortage quantity to reduce the actual shortage quantity of the procurement order to be predicted;
[0008] Among them, the training set for training the SARIMA model includes multiple procurement shortage records generated after allocating multiple supplier inventory lists and / or multiple supplier production lists to multiple procurement orders according to the greedy algorithm. The procurement shortage records include the identity of the supplier, the type of goods, the required arrival time, the required quantity, and the shortage quantity.
[0009] In the embodiments of the present application, the SARIMA model is obtained through the following steps: obtaining multiple purchase out-of-stock records; sorting the multiple purchase out-of-stock records according to the required arrival time to obtain an out-of-stock time series, where for any required arrival time, there is one or more purchase out-of-stock records corresponding to it; obtaining a preset SARIMA model according to the autocorrelation function graph and partial autocorrelation function graph of the out-of-stock time series; iteratively inputting the multiple purchase out-of-stock records into the preset SARIMA model one by one according to the required arrival time in chronological order to obtain the SARIMA model; where each iteration includes: inputting the supplier identity, goods type, required arrival time, and required quantity of goods of the purchase out-of-stock record into the preset SARIMA model to obtain the iterative out-of-stock quantity; adjusting the variable parameters in the preset SARIMA model according to the iterative out-of-stock quantity and the out-of-stock quantity of the purchase out-of-stock record.
[0010] In the embodiments of the present application, the parameters of the purchase order to be predicted further include: the expected average supply quantity of the supplier; the steps for obtaining the SARIMA model further include: obtaining the historical supply quantity mean value of the goods type by the supplier identity corresponding to each purchase out-of-stock record; iteratively inputting the multiple purchase out-of-stock records into the preset SARIMA model one by one according to the required arrival time in chronological order to obtain the SARIMA model, including: iteratively inputting the multiple purchase out-of-stock records and the corresponding historical supply quantity mean value into the preset SARIMA model according to the required arrival time in chronological order to obtain the SARIMA model.
[0011] In the embodiments of the present application, the purchase order management method for the chain master enterprise further includes: obtaining a purchase order sequence, a supplier inventory list sequence, and a supplier production list sequence, where the purchase orders in the purchase order sequence include the goods type, the supplier identity of the goods type, the required arrival time, and the required quantity of goods, the inventory lists in the supplier inventory list sequence include the supplier identity, the goods type, and the in-stock quantity, and the production lists in the supplier production list sequence include the supplier identity, the goods type, the goods off-line time, and the production quantity; selecting each purchase order in the purchase order sequence one by one according to the required arrival time; allocating the inventory list and / or production list that can meet the required arrival time to the currently selected purchase order through the greedy algorithm to update the purchase order sequence, the supplier inventory list sequence, and the supplier production list sequence until all the purchase orders in the purchase order sequence are allocated to obtain multiple purchase out-of-stock records.
[0012] In the embodiments of the present application, the inventory orders and / or production orders that can meet the required arrival time are allocated to the current purchase order through a greedy algorithm, including: determining, in the supplier inventory order sequence and the supplier production order sequence, the inventory orders and / or production orders that match the goods type of the current purchase order and the supplier identity, so as to obtain multiple allocation methods for the current purchase order; selecting, through the greedy algorithm, the allocation method that can meet the required arrival time from the multiple allocation methods, and allocating the corresponding inventory orders and / or production orders to the current purchase order.
[0013] In the embodiments of the present application, selecting, through the greedy algorithm, the allocation method that can meet the required arrival time from the multiple allocation methods includes: in the case where the inventory quantity of the allocation method is greater than or equal to the required quantity of the current purchase order, determining the allocation method as the one that can meet the required arrival time; in the case where the inventory quantity of each of the multiple allocation methods is less than the required quantity of the current purchase order, selecting, from the multiple allocation methods, the allocation method whose sum of the inventory quantity and the production quantity is greater than or equal to the required quantity of the current purchase order and the goods off-line time is earlier than the required arrival time as the one that can meet the required arrival time; in the case where the inventory quantity of each of the multiple allocation methods is less than the required quantity of the current purchase order and the goods off-line time is earlier than the required arrival time, and the sum of the inventory quantity and the production quantity is less than the required quantity of the current purchase order, determining the allocation method whose sum of the inventory quantity and the production quantity is closest to the required quantity of the current purchase order as the one that can meet the required arrival time.
[0014] In the embodiments of the present application, the number of current purchase orders is multiple, and the supplier identities of the multiple current purchase orders are different, or the goods types of the multiple current purchase orders are different; allocating, through the greedy algorithm, the inventory orders and / or production orders that can meet the required arrival time to the current purchase order further includes: respectively determining, in the supplier inventory order sequence and the supplier production order sequence, the inventory orders and / or production orders that match the goods types of the multiple current purchase orders and the supplier identities, so as to obtain multiple allocation methods for each of the multiple current purchase orders; concurrently selecting, through the greedy algorithm, the allocation methods that can meet the required arrival time from the multiple allocation methods for each of the multiple current purchase orders, and allocating the corresponding inventory orders and / or production orders to each current purchase order.
[0015] In the embodiment of the present application, the procurement order management method for the chain master enterprise further includes: after obtaining multiple procurement out-of-stock records, classifying the multiple procurement out-of-stock records according to the goods type and the supplier identity to obtain multiple sets of procurement out-of-stock records; analyzing the seasonal characteristics of the multiple procurement out-of-stock records in each set of procurement out-of-stock records in chronological order according to the out-of-stock quantity and the required arrival time of the procurement out-of-stock records; merging the sets of procurement out-of-stock records with the same seasonal characteristics to obtain a training set for SARIMA model training; in the case where the number of training sets is multiple, each training set corresponds to the training of one SARIMA model.
[0016] The second aspect of the present application provides a procurement order management device for a chain master enterprise, including: a memory configured to store instructions; and a processor configured to call instructions from the memory and capable of implementing the procurement order management method for the chain master enterprise provided in any one of the above embodiments when executing the instructions.
[0017] The third aspect of the present application provides a machine-readable storage medium, on which instructions are stored, and the instructions are used to cause a machine to execute the procurement order management method for the chain master enterprise provided in any one of the above embodiments.
[0018] The fourth aspect of the present application provides a computer program product, including a computer program, and the computer program realizes the procurement order management method for the chain master enterprise provided in any one of the above embodiments when executed by a processor.
[0019] Through the above technical solution, the seasonal characteristics of procurement orders, inventory orders, and production orders in terms of time will be combined and inherited in the time series of procurement out-of-stock records through the greedy algorithm, so as to reflect seasonality in the time series of procurement out-of-stock records. This manifestation of seasonal characteristics is captured by the SARIMA model, so as to predict the out-of-stock quantity of the subsequent procurement orders to be predicted; moreover, since each procurement out-of-stock record in the training set includes the supplier identity and the goods type, the trained SARIMA model can output the predicted out-of-stock quantity of the procurement order according to the goods type of the procurement goods and the supplier identity of the procurement goods in the procurement order to be predicted.
[0020] Other features and advantages of the embodiments of the present application will be described in detail in the subsequent specific implementation part. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] The drawings are used to provide a further understanding of the embodiments of the present application, and constitute a part of the specification. Together with the following specific implementation, they are used to explain the embodiments of the present application, but do not constitute a limitation to the embodiments of the present application. In the drawings:
[0022] Figure 1Schematically shows a flowchart of a procurement order management method for a chain master enterprise according to an embodiment of the present application;
[0023] Figure 2 Schematically shows a flowchart of a method for training a SARIMA model for predicting a procurement order to be predicted according to an embodiment of the present application;
[0024] Figure 3 Schematically shows a structural block diagram of a procurement order management device for a chain master enterprise according to an embodiment of the present application. Detailed implementation manners
[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. It should be understood that the specific implementation manners described herein are only for explaining and interpreting the embodiments of the present application, and are not used to limit the embodiments of the present application. Based on the embodiments in 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.
[0026] It should be noted that the acquisition, transmission, storage, use, processing, etc. of data in the technical solutions of the present application all comply with the relevant regulations of national laws and regulations. In the embodiments of the present application, some industry-existing solutions such as certain software, components, models, etc. may be mentioned. They should be regarded as exemplary, and their purpose is only to illustrate the feasibility in the implementation of the technical solutions of the present application, but it does not mean that the applicant has already or necessarily used this solution.
[0027] It should be noted that if there are descriptions involving "first", "second", etc. in the embodiments of the present application, the descriptions of "first", "second", etc. are only for descriptive purposes, and cannot be understood as indicating or implying their relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one such feature. In addition, the technical solutions between various embodiments can be combined with each other, but it must be based on the fact that those of ordinary skill in the art can implement them. When the combination of technical solutions appears to be contradictory or unable to be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by the present application.
[0028] SARIMA (Seasonal Autoregressive Integrated Moving Average) model.
[0029] The chain leader enterprise refers to the leading enterprise in the industrial chain. Generally, the chain leader enterprise occupies a dominant position in the entire industrial chain, has a strong direct or indirect influence on the resources and applications of most enterprises in the entire industrial chain, highly concerns the value realization of the entire industrial chain, and shoulders the heavy responsibility of improving the performance of the entire industrial chain as the core enterprise.
[0030] In the procurement order management method of the chain leader enterprise, one of the particularly important functions is to monitor the procurement order and adjust the procurement order in a timely manner based on predicting whether there will be a shortage of the procurement order in the future and the shortage quantity when there is a shortage, so as to avoid the occurrence of the actual shortage of the procurement order. The existing procurement order management methods generally use a deep neural network to iteratively train the system-related data as the model training set to obtain a prediction model. Although the model based on the deep neural network has high generalization, it has insufficient grasp of business relevance and accuracy in actual use. Moreover, the training of the neural network model requires a large data set and computing resources, and there are often problems such as insufficient training set and underfitting of the model in actual use. Therefore, it is difficult to meet the needs of small and medium-sized enterprises for predicting the shortage of orders.
[0031] Through the analysis of the procurement order and its related factors, it is found that the procurement orders of enterprises, as well as the inventory orders and production orders of suppliers, generally have periodicity, and this periodicity often reflects seasonal cycles, such as quarterly, monthly or weekly. Based on such seasonal rules, the embodiments of the present application provide a procurement order management method for chain leader enterprises, which captures the seasonal cycle characteristics of the shortage quantity of procurement orders by using the SARIMA model, so as to realize the prediction of the shortage quantity according to the business characteristics of the procurement orders.
[0032] Figure 1 Schematically shows a flowchart of a procurement order management method for chain leader enterprises according to an embodiment of the present application. As Figure 1 shown, the embodiments of the present application provide a procurement order management method for chain leader enterprises, and the method may include the following steps:
[0033] S102. Obtain the parameters of the procurement order to be predicted, and the parameters of the procurement order to be predicted include: the expected arrival time, the type of goods to be purchased, the identity of the supplier of the goods to be purchased, and the required quantity of the goods to be purchased;
[0034] S104. Determine the predicted shortage quantity of the procurement order to be predicted through the SARIMA model according to the parameters of the procurement order to be predicted;
[0035] S106. Adjust the procurement order to be predicted according to the predicted shortage quantity to reduce the actual shortage quantity of the procurement order to be predicted;
[0036] Among them, the training set for SARIMA model training includes multiple purchase shortage records generated after allocating multiple suppliers' inventory lists and / or multiple suppliers' production lists to multiple purchase orders according to the greedy algorithm. The purchase shortage records include supplier identity, goods type, required arrival time, required quantity of goods, and shortage quantity.
[0037] The purchase order management method for the chain master enterprise provided by the embodiments of the present application predicts the shortage quantity of purchase orders through the SARIMA model. The prediction of the shortage quantity depends on the following purchase order parameters: the required arrival time of the purchase order, the goods type of the purchased goods, the supplier identity of the purchased goods, and the required quantity of the purchased goods. The purchase shortage records included in the training set for the SARIMA model training are generated based on the greedy algorithm by allocating multiple suppliers' inventory lists and / or multiple suppliers' production lists to multiple purchase orders. Since one or more of the purchase orders, inventory lists, and production lists have seasonality in time, and the basic principle of the greedy algorithm is to obtain the current optimal solution, in the process of allocating purchase orders, according to the greedy algorithm, the inventory lists and / or production lists that can meet the requirements of the purchase orders will be allocated to the purchase orders. Therefore, the seasonal characteristics of the purchase orders, inventory lists, and production lists in time will be combined and inherited in the time series of the purchase shortage records, so that seasonality is reflected in the time series of the purchase shortage records. This manifestation of the seasonal characteristics is captured by the SARIMA model, so as to predict the shortage quantity for the subsequent purchase orders to be predicted. Moreover, since each purchase shortage record in the training set includes the supplier identity and the goods type, the trained SARIMA model can output the predicted shortage quantity of the purchase order corresponding to the goods type of the purchased goods and the supplier identity of the purchased goods of the purchase order to be predicted.
[0038] It can be understood that the supplier identity in step S102 can be, for example, the supplier number recorded by the chain master enterprise. The adjustment content for the purchase order to be predicted in step S106 may include: the supplier identity of the purchase order to be predicted, the required quantity of the purchased goods, and the expected required arrival time.
[0039] As an example, when the required quantity of the purchase order to be predicted cannot be satisfied by the current supplier at the expected required arrival time, the supplier identity in the purchase order to be predicted can be modified so that other suppliers who can meet the required quantity at the expected required arrival time can supply the goods.
[0040] As Figure 2 As shown, in some embodiments of the present application, to implement the SARIMA model to predict the shortage quantity of the purchase order to be predicted, the SARIMA model is obtained through the following steps:
[0041] S202. Obtain multiple purchase shortage records;
[0042] S204. Sort multiple purchase out-of-stock records according to the required arrival time to obtain an out-of-stock time series, where there is one or more purchase out-of-stock records corresponding to any required arrival time;
[0043] S206. Obtain a preset SARIMA model based on the autocorrelation function graph and partial autocorrelation function graph of the out-of-stock time series;
[0044] Iterate the preset SARIMA model by inputting multiple purchase out-of-stock records into the preset SARIMA model one by one in chronological order according to the required arrival time to obtain a SARIMA model, including:
[0045] S208. Input the purchase out-of-stock records into the preset SARIMA model in chronological order according to the required arrival time to iterate the preset SARIMA model;
[0046] Among them, each iteration includes:
[0047] S210. Input the supplier identity, goods type, required arrival time, and required quantity of goods of the purchase out-of-stock record into the preset SARIMA model to obtain an iterative out-of-stock quantity;
[0048] S212. Adjust the variable parameters in the preset SARIMA model according to the iterative out-of-stock quantity and the out-of-stock quantity of the purchase out-of-stock record;
[0049] S214. After each iteration, determine whether all the multiple purchase out-of-stock records sorted according to the required arrival time have been input. If not, return to step S208 to execute. If all have been input, obtain a SARIMA model.
[0050] Through the above training steps, it is possible to integrate purchase out-of-stock records including different goods types and different supplier identities into the same out-of-stock time series, and determine the parameters in the preset SARIMA model and train the preset SARIMA model based on this out-of-stock time series. In each iteration process of training the preset SARIMA model, the supplier identity, goods type, required arrival time, and required quantity of goods of the current purchase out-of-stock record will be input into the preset SARIMA model. Subsequently, the preset SARIMA model will output an iterative out-of-stock quantity of the current purchase out-of-stock record. According to the difference between the iterative out-of-stock quantity and the actual out-of-stock quantity of the current purchase out-of-stock record, the adjustable parameters in the SARIMA model can be adjusted, so that the iterative out-of-stock quantity and the actual out-of-stock quantity are closer. After all the multiple purchase out-of-stock records in the out-of-stock time series have been input, the iteration of the preset SARIMA model is completed, and thus a SARIMA model for predicting the out-of-stock quantity is obtained.
[0051] Understandably, in step S204, after sorting multiple purchase out-of-stock records according to the required arrival time, when there are multiple purchase out-of-stock records corresponding to any required arrival time, the differences between the multiple purchase out-of-stock records corresponding to the required arrival time may include different supplier identities and different types of goods.
[0052] Understandably, in step S206, the autocorrelation function graph is used to determine the correlation of the out-of-stock time series with itself at different time lags. The partial autocorrelation function graph is used to determine the direct correlation between the out-of-stock quantity at multiple selected times after the out-of-stock time series and the out-of-stock quantity at the current time, and excludes the influence of other out-of-stock quantities between the two.
[0053] Specifically, step S206 may include:
[0054] Obtain the out-of-stock time series;
[0055] In the case where the out-of-stock time series is not stationary, difference the out-of-stock time series to obtain a stationary out-of-stock time series and determine the number of differences;
[0056] Plot the autocorrelation function graph and partial autocorrelation function graph of the stationary out-of-stock time series, and determine the non-seasonal parameters p and q in the preset SARIMA model based on the autocorrelation function graph and partial autocorrelation function graph; where p can be the first significant lag point of the partial autocorrelation function graph, and q can be the first significant lag point of the autocorrelation function graph.
[0057] Plot the seasonal autocorrelation function graph and seasonal partial autocorrelation function graph of the stationary out-of-stock time series, and determine the seasonal parameters P and Q and the seasonal period s in the preset SARIMA model based on the seasonal autocorrelation function graph and seasonal partial autocorrelation function graph; where P is the first significant lag point of the seasonal partial autocorrelation function graph, and Q is the first significant lag point of the seasonal autocorrelation function graph;
[0058] According to the non-seasonal parameters p and q, and the seasonal parameters P, Q, and s, the parameter initialization of the preset SARIMA model can be determined, thereby obtaining the preset SARIMA model.
[0059] In some embodiments of the present application, the parameter of the purchase order to be predicted further includes: the expected average supply quantity of the supplier;
[0060] The steps of obtaining the SARIMA model further include:
[0061] Obtain the historical average supply quantity of the goods type by the supplier identity corresponding to each purchase out-of-stock record;
[0062] Iterate the preset SARIMA model by inputting multiple purchase out-of-stock records one by one into the preset SARIMA model in chronological order according to the required arrival time, so as to obtain the SARIMA model, including:
[0063] Input multiple purchase out-of-stock records and the corresponding historical average supply volume into the preset SARIMA model in chronological order according to the required arrival time, and iterate the preset SARIMA model to obtain the SARIMA model.
[0064] The historical average supply volume can reflect the average supply volume of a specific type of goods by a supplier during a certain period in history. The historical supply volume can be used as a derived variable and input into the preset SARIMA model to train the model to obtain the SARIMA model. By using the historical average supply volume as a derived variable to train the preset SARIMA model, the finally obtained SARIMA model can capture the law between the historical average supply volume and the out-of-stock volume, thereby improving the accuracy of the SARIMA model in predicting the out-of-stock volume. When predicting the out-of-stock volume through the SARIMA model, the expected average supply volume of the corresponding purchase order to be predicted can be combined with other parameters of the purchase order to be predicted and input into the SARIMA model to realize the prediction of the out-of-stock volume of the purchase order to be predicted by the SARIMA model.
[0065] It can be understood that the historical supply volume in the historical average supply volume may include the average inventory volume, and / or average production volume, and / or average production volume of a specific type of goods of a certain supplier within a preset period. The preset period can be, for example, weekly, monthly, or a period set by the staff after analyzing the historical supply volume to associate the periodic law of the historical average supply volume with the periodic law of the out-of-stock volume.
[0066] Since the quality inspection pass rate will affect the actual supply volume of the supplier, when the average quality inspection pass rate is included in the historical average supply volume, the average quality inspection pass rate can be multiplied by the average production volume to obtain the actual average production volume. Of course, in some embodiments of the present application, the average inventory volume, the average production volume, and the average production volume can be used as three types of derived variables respectively.
[0067] In order to predict the out-of-stock volume of the purchase order to be predicted, it is necessary to first obtain the purchase out-of-stock records for training the preset SARIMA model.
[0068] Therefore, in some embodiments of the present application, the purchase order management method for the chain owner enterprise may further include the following steps of allocating goods through the greedy algorithm to obtain the purchase out-of-stock records:
[0069] Obtain a purchase order sequence, a supplier inventory list sequence, and a supplier production order sequence. The purchase orders in the purchase order sequence include the goods type, the supplier identity of the goods type, the required arrival time, and the required quantity of goods. The inventory lists in the supplier inventory list sequence include the supplier identity, the goods type, and the inventory quantity of goods. The production orders in the supplier production order sequence include the supplier identity, the goods type, the time when the goods come off the production line, and the production quantity of goods;
[0070] Select each purchase order in the purchase order sequence one by one according to the required arrival time;
[0071] Use the greedy algorithm to allocate the inventory lists and / or production orders that can meet the required arrival time to the currently selected purchase order, so as to update the purchase order sequence, the supplier inventory list sequence, and the supplier production order sequence until all the purchase orders in the purchase order sequence are allocated, so as to obtain multiple purchase shortage records.
[0072] Based on the above steps, use the greedy algorithm to allocate goods to the purchase orders that currently need to be allocated goods in the purchase order sequence, and follow the basic principles of the greedy algorithm to allocate the inventory lists in the supplier inventory list sequence that can best meet the current purchase order and / or the production orders in the supplier production order sequence that can best meet the current purchase order to the current purchase order. Thus, based on the time sequence in the purchase order sequence, the time sequence characteristics of the purchase order sequence, the supplier inventory list sequence, and the supplier production order sequence are merged into the time sequence of the purchase shortage record sequence.
[0073] In some embodiments of the present application, using the greedy algorithm to allocate the inventory lists and / or production orders that can meet the required arrival time to the current purchase order includes:
[0074] In the supplier inventory list sequence and the supplier production order sequence, determine the inventory lists and / or production orders that match the goods type of the current purchase order and the supplier identity to obtain multiple allocation methods for the current purchase order;
[0075] Use the greedy algorithm to select the allocation method that can meet the required arrival time from multiple allocation methods, and allocate the corresponding inventory lists and / or production orders to the current purchase order.
[0076] By obtaining multiple allocation methods for the current purchase order, all the allocable methods of the current purchase order can be determined on the premise of the known supplier inventory list sequence and supplier production order sequence, so as to subsequently determine, based on the greedy algorithm, the allocation method that can meet the required arrival time among the multiple allocation methods, and select the allocation method that can meet the required quantity of goods as the goods allocation method for the current purchase order.
[0077] In some embodiments of the present application, a greedy algorithm is used to select an allocation method that can meet the required arrival time from multiple allocation methods, including:
[0078] When the inventory quantity of the allocation method is greater than or equal to the required quantity of the current purchase order, the allocation method is determined as the allocation method that can meet the required arrival time;
[0079] When the inventory quantities of multiple allocation methods are all less than the required quantity of the current purchase order, among the multiple allocation methods, an allocation method whose sum of inventory quantity and production quantity is greater than or equal to the required quantity of the current purchase order and whose goods off-line time is earlier than the required arrival time is determined as the allocation method that can meet the required arrival time;
[0080] Among the allocation methods where the inventory quantities of multiple allocation methods are all less than the required quantity of the current purchase order and the goods off-line time is earlier than the required arrival time, when the sum of inventory quantity and production quantity is less than the required quantity of the current purchase order, the allocation method with the sum of inventory quantity and production quantity closest to the required quantity of the current purchase order is determined as the allocation method that can meet the required arrival time.
[0081] Since the goods in the inventory list are spot goods and there is no need to consider the constraint of the goods off-line time on the required arrival time, the above-mentioned allocation method based on the greedy algorithm preferentially selects the inventory list with the inventory quantity greater than or equal to the required quantity of the current purchase order and allocates it to the current purchase order, thereby completing the determination of the allocation method for the current purchase order. Since among the multiple allocation methods, there are an allocation method of allocating the inventory list alone to the current purchase order, an allocation method of allocating the inventory list and the production list in combination to the current purchase order, and an allocation method of allocating the production list alone to the current purchase order. Therefore, in the case where there is no inventory list that can meet the current purchase order among the multiple allocation methods, the allocation method of combining the inventory list and the production list can be preferentially selected to determine the allocation method of the current purchase order. Thus, on a certain basis, part of the required quantity of the current purchase order is satisfied by the inventory list, the out-of-stock risk is reduced, and the remaining part of the required quantity of the current purchase order is satisfied by the production list with the goods off-line time earlier than the required arrival time. In the case where multiple allocation methods cannot fully meet the required quantity of the current purchase order, the allocation method with the sum of inventory quantity and production quantity closest to the required quantity of the current purchase order is determined as the allocation method that can meet the required arrival time, and at the same time, a non-zero out-of-stock quantity of the current purchase order can be obtained.
[0082] In some embodiments of the present application, since the quality inspection pass rates of the goods from different suppliers are different. Therefore, in the supplier inventory list sequence and the supplier production list sequence, determine the inventory list and / or production list that match the goods type of the current purchase order and the supplier identity to obtain multiple allocation methods for the current purchase order, including:
[0083] Determine the production quantity of the current allocation method according to the production list in the current allocation method;
[0084] Multiply the production quantity of the current allocation method by the average quality inspection pass rate of this supplier to obtain the actual production quantity of the current allocation method.
[0085] Thus, it is possible to determine whether the current allocation method can meet the required quantity of the current purchase order according to the actual production quantity in the current allocation method.
[0086] As an example, the process of selecting the allocation method that can meet the required arrival time from multiple allocation methods through the greedy algorithm can be as follows:
[0087] First, obtain the purchase order sequence, the supplier inventory list sequence, and the supplier production list sequence.
[0088] Table 1 Purchase order sequence issued by the chain owner enterprise
[0089] Purchase order number Supplier code Material number Required arrival time Required quantity ORDER001 0000001 M001 2024-06-01 70 ORDER002 0000001 M001 2024-06-02 50 ORDER003 0000001 M002 2024-06-03 60 ORDER004 0000002 M002 2024-06-01 80
[0090] As shown in Table 1, the purchase order sequence issued by the chain owner enterprise may include multiple purchase orders, and each purchase order is distinguished by a purchase order number. The supplier identity of the purchase order is distinguished and recorded through the supplier code, and the goods type is recorded by the material number. The required arrival time can be, for example, year / month / day, and the required quantity is the unit quantity of the goods.
[0091] Table 2 Supplier inventory list sequence
[0092] Stock order number Supplier code Material number Stock quantity S001 0000001 M001 80 S002 0000001 M002 80 S003 0000002 M002 40
[0093] As shown in Table 2, the supplier inventory list sequence may include multiple inventory lists, and each inventory list is distinguished by an inventory list number. The supplier identity of the inventory list is distinguished and recorded through the supplier code, and the goods type is recorded by the material number. The inventory quantity is the unit quantity of the goods.
[0094] Table 3 Supplier production list sequence
[0095] Production order number Supplier code Material number Goods off-line time Production quantity P001 0000001 M001 2024-05-30 30 P002 0000001 M002 2024-06-03 10 P003 0000002 M002 2024-06-01 20
[0096] As shown in Table 3, the supplier production order sequence may include multiple production orders, and each production order is distinguished by a production order number. The supplier identity of the production order is distinguished and recorded through the supplier code, and the goods type is recorded by the material number. The goods off-line time can be, for example, year / month / day, and the production quantity of the goods is the unit quantity of the goods.
[0097] After obtaining the purchase order sequence, the supplier inventory order sequence, and the supplier production order sequence, goods are allocated to the purchase orders. First, for example, allocate the purchase order with the purchase order number ORDER001. The corresponding supplier identity of ORDER001 is 0000001, and the average quality inspection pass rate of the goods of supplier 0000001 can be, for example, 0.95.
[0098] Based on the purchase order sequence, the supplier inventory order sequence, and the supplier production order sequence, among the multiple allocation methods of ORDER001, allocating the inventory order with the inventory order number S001 to the ORDER001 purchase order can satisfy the ORDER001 purchase order.
[0099] Correspondingly, the shortage quantity of ORDER001 is:
[0100] ORDER001 shortage quantity = ORDER001 required quantity - inventory quantity of S001 allocated to the ORDER001 purchase order = 70 - 70 = 0.
[0101] Subsequently, update the purchase order sequence and the supplier inventory order sequence to obtain the current inventory quantity of the S001 inventory order as 10.
[0102] In the process of subsequently allocating goods to the ORDER002 purchase order, among the multiple allocation methods of the ORDER002 purchase order, preferentially select the allocation method including the S001 inventory order. For example, one allocation method includes: the combination of the S001 inventory order and the P001 production order. Correspondingly, the shortage quantity of ORDER002 is:
[0103] ORDER002 shortage quantity = ORDER002 required quantity - inventory quantity of S001 allocated to the ORDER002 purchase order - production quantity of P001 * average quality inspection pass rate of supplier 0000001 = 50 - 10 - 30 * 0.95 = 12.
[0104] In some embodiments of the present application, to improve the allocation efficiency, the number of current purchase orders can be multiple, and the supplier identities of the multiple current purchase orders are different, or the goods types of the multiple current purchase orders are different. By using the greedy algorithm, the inventory orders and / or production orders that can meet the required arrival time are allocated to the current purchase orders, and it may further include:
[0105] In the supplier inventory list sequence and the supplier production list sequence, determine the inventory list and / or production list that match the goods types of multiple current purchase orders and the supplier identities respectively, so as to obtain multiple allocation methods for each of the multiple current purchase orders;
[0106] In parallel, select the allocation methods that can meet the required arrival time from the multiple allocation methods for each of the multiple current purchase orders through the greedy algorithm, and allocate the corresponding inventory list and / or production list to each current purchase order.
[0107] If the supplier identities or the goods types are different among the purchase orders, then allocating one of the purchase orders will not affect the subsequent allocation results of another purchase order with different supplier identities or goods types. Therefore, multiple current purchase orders with different supplier identities or goods types can be allocated in parallel through the greedy algorithm to improve the allocation efficiency.
[0108] Specifically, selecting the allocation methods that can meet the required arrival time from the multiple allocation methods for each of the multiple current purchase orders in parallel through the greedy algorithm can be implemented through the ForkJoinPool framework.
[0109] Based on the above steps, multiple purchase orders in Table 1 can be gradually allocated one by one.
[0110] In some embodiments of the present application, after each purchase order in the purchase order sequence is allocated and multiple purchase out-of-stock records are obtained, the purchase order management method for the chain master enterprise may further include:
[0111] Classify the multiple purchase out-of-stock records according to the goods types and supplier identities to obtain multiple purchase out-of-stock record sets;
[0112] Analyze the seasonal characteristics in time series of the multiple purchase out-of-stock records in each purchase out-of-stock record set according to the out-of-stock quantity and required arrival time of the purchase out-of-stock records;
[0113] Merge the purchase out-of-stock record sets with the same seasonal characteristics to obtain the training set for SARIMA model training;
[0114] When the number of training sets is multiple, each training set is correspondingly used for the training of a SARIMA model.
[0115] Based on the above steps, a preset SARIMA model can be trained through the purchase out-of-stock record sets with the same seasonal characteristics, without distinguishing different goods types or supplier identities, so that the SARIMA model can predict the out-of-stock quantity of the to-be-predicted purchase orders with different goods types or supplier identities.
[0116] Such asFigure 3 As shown in Figure 3 , an embodiment of the present application further provides a procurement order management device for a chain master enterprise, including: a memory 310 and a processor 320. The memory is configured to store instructions. The processor is configured to call instructions from the memory and, when executing the instructions, can implement the procurement order management method for the chain master enterprise provided in any of the above embodiments.
[0117] An embodiment of the present application further provides a readable storage medium, on which instructions are stored, and the instructions are used to cause a machine to execute the procurement order management method for the chain master enterprise provided in any of the above embodiments.
[0118] An embodiment of the present application further provides a computer program product, including a computer program, and when the computer program is executed by a processor, it implements the procurement order management method for the chain master enterprise provided in any of the above embodiments.
[0119] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) containing computer-usable program code.
[0120] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of flows and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.
[0121] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured product including an instruction device, and the instruction device implements the functions specified in Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.
[0122] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 steps of the functions specified in one block or multiple blocks.
[0123] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and memory.
[0124] The memory may include non-permanent memory in the form of computer-readable media, random access memory (RAM) and / or non-volatile memory such as read-only memory (ROM) or flash memory (flash RAM). The memory is an example of computer-readable media.
[0125] Computer-readable media includes permanent and non-permanent, removable and non-removable media and can store information by any method or technology. 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, disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transitory media such as modulated data signals and carrier waves.
[0126] It should also be noted that the term "comprising", "including" or any other variation thereof is intended to cover non-exclusive inclusion, so that a process, method, commodity or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or elements inherent to such process, method, commodity 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, commodity or device comprising the element.
[0127] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, various changes and modifications can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.
Claims
1. A purchase order management method for a chain leader enterprise, characterized in that: include: Acquire parameters of the purchase order to be predicted, wherein the parameters of the purchase order to be predicted include: expected arrival time, type of purchased goods, identity of the supplier of the purchased goods, and required quantity of the purchased goods; Determine the predicted out-of-stock quantity of the purchase order to be predicted by using the SARIMA model according to the purchase order parameters to be predicted; Adjusting the purchase order to be predicted according to the predicted stock-out quantity to reduce the actual stock-out quantity of the purchase order to be predicted; Among them, the training set used for training the SARIMA model includes multiple purchase out-of-stock records generated after allocating multiple supplier inventory orders and / or multiple supplier production orders to multiple purchase orders according to a greedy algorithm, and the purchase out-of-stock records include supplier identity, goods type, required arrival time, required quantity, and out-of-stock quantity.
2. The method according to claim 1, characterized in that The SARIMA model is obtained by the following steps: Obtaining a plurality of the purchase out-of-stock records; Sorting the plurality of purchase out-of-stock records according to the required arrival time to obtain an out-of-stock time sequence, wherein one or more purchase out-of-stock records correspond to any of the required arrival times; According to the autocorrelation function graph and the partial autocorrelation function graph of the out-of-stock time series, a preset SARIMA model is obtained; Inputting the plurality of purchase out-of-stock records into the preset SARIMA model one by one in chronological order according to the required arrival time, iterating the preset SARIMA model, so as to obtain the SARIMA model; Each iteration includes: Inputting the supplier identity, the goods type, the required arrival time, and the required quantity of the purchase out-of-stock record into the preset SARIMA model to obtain an iterative out-of-stock quantity; The variable parameters in the preset SARIMA model are adjusted according to the iterative stock-out quantity and the stock-out quantity of the procurement stock-out record.
3. The method according to claim 2, characterized in that The purchase order parameters to be predicted also include: the expected average supply quantity of the supplier; The step of obtaining the SARIMA model also includes: Obtaining the average historical supply quantity of the commodity type by the supplier identity corresponding to each of the purchase out-of-stock records; The step of inputting the plurality of purchase out-of-stock records into the preset SARIMA model one by one in chronological order according to the required arrival time, iterating the preset SARIMA model to obtain the SARIMA model comprises: According to the required arrival time, a plurality of the purchase out-of-stock records and the corresponding historical supply quantity averages are input into the preset SARIMA model in chronological order, and the preset SARIMA model is iterated to obtain the SARIMA model.
4. The method according to claim 1, characterized in that: Also includes: Acquire a purchase order sequence, a supplier inventory order sequence, and a supplier production order sequence, wherein the purchase order in the purchase order sequence includes the type of goods, the supplier identity of the type of goods, the required arrival time, and the required quantity of goods, the inventory order in the supplier inventory order sequence includes the supplier identity, the type of goods, and the inventory quantity, and the production order in the supplier production order sequence includes the supplier identity, the type of goods, the time when the goods are offline, and the production quantity; Selecting each of the purchase orders in the purchase order sequence one by one according to the required arrival time; The inventory order and / or the production order that can meet the required arrival time is allocated to the selected current purchase order through a greedy algorithm to update the purchase order sequence, the supplier inventory order sequence and the supplier production order sequence until all the purchase orders in the purchase order sequence are allocated to obtain a plurality of purchase out-of-stock records.
5. The method according to claim 4, characterized in that The allocating the inventory order and / or the production order that can meet the required arrival time to the current purchase order by using a greedy algorithm includes: In the supplier inventory order sequence and the supplier production order sequence, determine the inventory order and / or the production order that matches the goods type of the current purchase order and matches the identity of the supplier, so as to obtain multiple allocation methods for the current purchase order; A distribution method that can meet the required arrival time is selected from the multiple distribution methods through a greedy algorithm, and the corresponding inventory order and / or the production order is distributed to the current purchase order.
6. The method according to claim 5, characterized in that The method of selecting an allocation method that can meet the required arrival time from the multiple allocation methods by using a greedy algorithm includes: When the inventory quantity of the allocation method is greater than or equal to the required quantity of the current purchase order, the allocation method is determined as the allocation method that can meet the required arrival time; In the case that the inventory quantity of each of the multiple allocation methods is less than the required quantity of the current purchase order, an allocation method in which the sum of the inventory quantity and the production quantity is greater than or equal to the required quantity of the current purchase order and the goods offline time is earlier than the required arrival time is selected from the multiple allocation methods and is determined as the allocation method that can meet the required arrival time; In the case where the inventory quantity of each of the multiple allocation methods is less than the demand quantity of the current purchase order, and among the allocation methods in which the goods are offline before the required arrival time, the sum of the inventory quantity and the production quantity is less than the demand quantity of the current purchase order, the allocation method in which the sum of the inventory quantity and the production quantity is closest to the demand quantity of the current purchase order is determined as the allocation method that can meet the required arrival time.
7. The method according to claim 5, characterized in that There are multiple current purchase orders, and the suppliers of the multiple current purchase orders are of different identities, or the types of goods of the multiple current purchase orders are of different types; The step of allocating the inventory order and / or the production order that can meet the required delivery time to the current purchase order by a greedy algorithm further includes: In the supplier inventory order sequence and the supplier production order sequence, the inventory orders and / or the production orders that match the goods type of the multiple current purchase orders and match the supplier identity are respectively determined to obtain multiple allocation methods for each of the multiple current purchase orders; In parallel, a greedy algorithm is used to select an allocation method that can meet the required arrival time from among the multiple allocation methods of the multiple current purchase orders, and the corresponding inventory order and / or the production order is allocated to each of the current purchase orders.
8. The method according to claim 4, characterized in that Also includes: After obtaining the plurality of purchase out-of-stock records, classifying the plurality of purchase out-of-stock records according to the type of goods and the identity of the supplier to obtain a plurality of purchase out-of-stock record sets; Analyzing seasonal characteristics of multiple purchase out-of-stock records in each of the purchase out-of-stock record sets in time series according to the out-of-stock quantity and the required arrival time of the purchase out-of-stock record; Merging the purchase out-of-stock record sets with the same seasonal characteristics to obtain a training set for the SARIMA model training; When there are multiple training sets, each training set corresponds to the training of one SARIMA model.
9. A purchase order management device for a chain leader enterprise, characterized in that: include: a memory configured to store instructions; as well as A processor is configured to call the instructions from the memory and implement the purchase order management method for a chain leader enterprise according to any one of claims 1 to 8 when executing the instructions.
10. A machine-readable storage medium, characterized in that: The machine-readable storage medium stores instructions for causing a machine to execute a purchase order management method for a chain leader enterprise according to any one of claims 1 to 8.
11. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the computer program implements the purchase order management method for a chain leader enterprise according to any one of claims 1 to 8.