CN113469397A

CN113469397A - Intelligent supply chain system and server platform

Info

Publication number: CN113469397A
Application number: CN202010241091.3A
Authority: CN
Inventors: 黎杰臻; 左滨; 戴芸
Original assignee: Hitachi Ltd
Current assignee: Hitachi Ltd
Priority date: 2020-03-31
Filing date: 2020-03-31
Publication date: 2021-10-01

Abstract

The invention provides an intelligent supply chain system and a server platform. An intelligent supply chain system for managing supply chains of a plurality of commodities, comprising: a commodity classification device for classifying a plurality of commodities based on the history data; a sales prediction device that predicts sales for each of a plurality of types of commodities based on historical data and the classification result of the commodity classification device for the plurality of types of commodities; and an intelligent replenishment device which generates a replenishment decision by applying an automatic replenishment model based on a prediction result of the sales prediction device, wherein the intelligent replenishment device comprises a mixed replenishment part which generates a periodic and quantitative replenishment decision based on a point prediction result and a section prediction result of the sales prediction device. Therefore, three important blocks of commodity classification, sales prediction and intelligent replenishment can be effectively fused, so that the service becomes accurate, the supply chain becomes transparent, flexible and agile, and all functions are more cooperated.

Description

Intelligent supply chain system and server platform

Technical Field

The invention relates to an intelligent supply chain system and a server platform, in particular to a layout mode of an intelligent supply chain.

Background

In recent years, there is a strong need for effective management of the supply chain between manufacturers, suppliers, and sales stores. Specifically, in supply chain management, the commodities need to be scientifically classified and located, and then corresponding selling and inventory strategies are adopted for different classes of commodities. Secondly, more accurate sales prediction is needed to be used as a driver of the commodity quantity, a demand plan taking the sales prediction as a core can support the planning of each link, and the operation cost can be reduced by several times when the prediction accuracy is improved by 1%. And finally, applying the automatic replenishment model to a specific scene to optimize an inventory structure so as to continuously keep the inventory at a healthy level, positioning all SKUs (stock keeping units) meeting the replenishment condition under the condition of warehousing, obtaining a recommendation of replenishment quantity according to the replenishment task running model, and finally generating a replenishment decision to be delivered to production.

However, the prior art does not provide a supply chain management scheme that satisfies the above requirements, and an intelligent supply chain layout method that can satisfy the above requirements is urgently needed.

Moreover, with the rapid development of economy, the influence of logistics cost on enterprise profit is continuously increased, and in the case of gradual optimization of management optimization, the space for cost reduction is less and less, so that the control of logistics cost becomes an important ring. According to the market reflection, logistics has a great promotion space in the aspect of reducing the cost, and the inventory cost accounts for a significant part of the logistics cost. Therefore, inventory management is an important part of enterprise management cost, which directly affects customer satisfaction, and whether there is enough inventory to meet customer needs is also an important factor for increasing sales volume.

The setting of the stock directly affects the stock management cost, and too high stock setting brings the stock stay, thereby bringing unnecessary stock cost and possible scrapping cost, and leading the enterprise to have fund overstock and having bad influence on the operation condition of the enterprise. If the inventory setting is too low, the customer order cannot be satisfied, so that the sales opportunity is lost, the sales performance of the enterprise is directly influenced, and the reputation of the enterprise is possibly influenced.

At present, a single method is mostly adopted for management in the inventory management strategy, the flexibility is low, and the adaptive scene is single. Therefore, the search for new inventory control methods is urgent.

Disclosure of Invention

The invention aims to solve the technical problems in the prior art, and aims to provide an intelligent supply chain system and a server platform which can effectively integrate three important blocks of commodity classification, sales prediction and intelligent replenishment, so that the service becomes accurate, the supply chain becomes transparent, flexible and agile, and all functions are more cooperative.

One embodiment of the present invention provides an intelligent supply chain system for managing supply chains of a plurality of types of commodities, including: a commodity classification device for classifying a plurality of commodities based on the history data; a sales prediction device that predicts sales for each of a plurality of types of commodities based on historical data and the classification result of the commodity classification device for the plurality of types of commodities; and an intelligent replenishment device which generates a replenishment decision by applying an automatic replenishment model based on a prediction result of the sales prediction device, wherein the intelligent replenishment device comprises a mixed replenishment part which generates a periodic and quantitative replenishment decision based on a point prediction result and a section prediction result of the sales prediction device.

Therefore, three important blocks of commodity classification, sales prediction and intelligent replenishment can be effectively fused, so that the service becomes accurate, the supply chain becomes transparent, flexible and agile, and all functions are more cooperated.

In the above-described intelligent supply chain system, the hybrid replenishment part may include: a replenishment cycle calculation unit that calculates a replenishment cycle as an optimal economic replenishment cycle based on the annual sales amount prediction amount and the optimal economic replenishment amount in the prediction result; a safety stock calculation unit that calculates a safety stock amount based on the sales volume prediction and the safety factor corresponding to the actual prediction error and the service level; the replenishment quantity calculation unit is used for calculating the replenishment quantity as the actual replenishment quantity based on the predicted quantity of the sales in the replenishment lead period, the predicted quantity of the sales in the replenishment period, the safe stock and the current stock; and a replenishment plan generating unit that generates a replenishment plan based on the optimum economic replenishment cycle and the replenishment quantity calculated by the replenishment quantity calculating unit as an actual replenishment quantity.

Thus, during the replenishment period, regular replenishment is performed using the optimum economic replenishment period, and the actual replenishment quantity is calculated from the stock quantity in the replenishment period, so that it is possible to control the cost well and reduce the traffic when managing the replenishment business.

In the aforementioned intelligent supply chain system, the hybrid replenishment part may further include: the restocking point calculating unit calculates a restocking point based on the sales forecast quantity in the advance period of restocking and the safe stock; and a replenishment triggering determination unit that compares the current stock quantity with the replenishment point and determines whether to trigger a replenishment operation based on a result of the comparison.

By comparing the restocking point with the current stock quantity, the condition that the goods are restocked in advance in the replenishment period of the regular replenishment possibly can be emergently dealt with under the condition that the risk of goods shortage caused by fluctuation of the demand occurs, so that the fluctuation of the urgent demand can be dealt with under the condition of controlling the cost, and a perfect replenishment strategy is provided for enterprises.

In the above-described intelligent supply chain system, the replenishment trigger determination unit may determine that a replenishment operation is triggered when the current stock amount is equal to or less than the replenishment point, and the replenishment plan generation unit may generate a replenishment plan based on the replenishment quantity calculated by the replenishment quantity calculation unit as an actual replenishment quantity.

Therefore, under the condition of the shortage risk caused by the fluctuation of the demand, the emergency demand fluctuation can be dealt with under the condition of controlling the cost, a perfect replenishment strategy is provided for enterprises, the replenishment frequency and the replenishment quantity are reasonably optimized, the logistics resources are fully utilized, and the purpose of reducing the cost is achieved.

In the above-described intelligent supply chain system, the replenishment cycle calculation unit may calculate the optimal economic replenishment quantity based on an economic replenishment quantity model that is an EOQ model, based on a cost per order in a cycle, a storage cost per commodity in a cycle, and a demand in a cycle based on the probability density section prediction result.

The EOQ model is used to obtain the optimal economic replenishment period, thereby controlling the cost and reducing the amount of business when the replenishment business is managed.

An embodiment of the present invention provides a server platform that manages supply chains of a plurality of types of commodities, including a processor, a memory, and an interface, and that is capable of data communication with a client device via the interface, wherein the processor executes a program stored in the memory to perform: a commodity classification process for classifying a plurality of commodities based on historical data; a sales prediction process of performing sales prediction for each commodity based on historical data and a classification result of the commodity classification process for a plurality of commodities; and intelligent replenishment processing, wherein an automatic replenishment model is applied to generate a replenishment decision based on a prediction result of the sales prediction processing, in the replenishment intelligent processing, a periodic and quantitative replenishment decision is generated based on a point prediction result and an interval prediction result obtained by the sales prediction processing, and the result of the sales prediction processing and/or the replenishment decision are/is sent to the client device through the interface.

The above-described embodiments and effects of the intelligent supply chain system according to the present invention can be realized by the server platform, the method executed by the intelligent supply chain system, the program for causing the computer to execute the method, or the recording medium storing the program.

Drawings

Fig. 1 is a conceptual diagram illustrating an application object to which the smart supply chain system according to the first embodiment of the present invention can be applied.

Fig. 2 is a functional block diagram of an intelligent supply chain system according to a first embodiment of the present invention.

Fig. 3 is a block diagram showing an example of a hardware configuration of the intelligent supply chain system according to the first embodiment of the present invention.

Fig. 4 is a block diagram showing another example of the hardware configuration of the intelligent supply chain system according to the first embodiment of the present invention.

FIG. 5 is a flowchart illustrating a method performed by the intelligent supply chain system according to the first embodiment of the present invention.

Fig. 6 is a functional block diagram of an intelligent supply chain system according to a second embodiment of the present invention.

FIG. 7 is a flowchart illustrating a method performed by the intelligent supply chain system according to a second embodiment of the present invention.

Fig. 8 is a functional block diagram showing the intelligent replenishment device 30 in the intelligent supply chain system 100 according to the second embodiment of the present invention.

Fig. 9 is a schematic diagram illustrating the establishment of an optimal economic replenishment quantity EOQ based on an EOQ model utilized in an intelligent supply chain system according to a second embodiment of the present invention.

Fig. 10 is a flowchart illustrating a hybrid replenishment method for an intelligent capture device in an intelligent supply chain system according to a second embodiment of the present invention.

Fig. 11 is a flowchart illustrating a hybrid replenishment method of an intelligent capture device in an intelligent supply chain system according to a second embodiment of the present invention.

FIG. 12 is a flowchart showing an application example (specific example 1) in the case where the intelligent supply chain system of the present invention is applied to a manufacturer.

Detailed Description

The present invention will be described in more detail below with reference to the accompanying drawings, embodiments, and specific examples. The following description is only an example for the convenience of understanding the present invention and is not intended to limit the scope of the present invention. In the embodiments, the components of the apparatus and the system may be changed, deleted, or added according to the actual situation, and the steps of the method may be changed, deleted, added, or changed in order according to the actual situation.

(first embodiment)

First, an application object to which the intelligent supply chain system 100 of the present embodiment can be applied will be described. Fig. 1 is a conceptual diagram illustrating an application object to which the smart supply chain system according to the first embodiment of the present invention can be applied. The application objects of the intelligent supply chain system 100 include at least one of a sales store, a supplier and a manufacturer, wherein the relationship between the sales store, the supplier and the manufacturer is shown in fig. 1, and the arrows indicate the flow of goods. Three sales stores, namely, a sales store 1, a sales store 2, and a sales store 3, three manufacturers, namely, a supplier 1, a supplier 2, and a manufacturer 1, a manufacturer 2, and a manufacturer 3, are shown here, but the number and relationship of the sales stores, the suppliers, and the manufacturers are not limited thereto, and a large number of sales stores, suppliers, manufacturers, and more complicated relationships may actually exist.

The sales stores need to monitor sales volume of each commodity, so as to make sales and inventory strategies. Since a supplier takes goods from different manufacturers and supplies the taken goods to a plurality of sales stores, it is necessary to order the manufacturers and arrange a delivery plan for the sales stores at the next stage. The manufacturer produces the goods to supply to a plurality of different suppliers, so that the suppliers supply the goods to the sales stores, the life cycle of each kind of goods is different, and the demand of the goods in the future needs to be predicted so as to guide the future production plan.

In addition, since "ex-warehouse" for a certain customer is equivalent to "sales" for a supplier/manufacturer, the "ex-warehouse" in the supplier/manufacturer and "sales" in a sales store are collectively expressed as "sales" in this specification. For example, "sales time" includes not only sales time in a sales store but also ex-warehouse time in a supplier/manufacturer, "sales amount" includes not only sales amount in a sales store but also ex-warehouse amount in a supplier/manufacturer.

In response to at least one of the above needs, the present embodiment provides a smart supply chain system 100. Fig. 2 is a functional block diagram of the intelligent supply chain system 100 according to the first embodiment of the present invention. As shown in fig. 2, the intelligent supply chain system 100 includes a product sorting device 10, a sales predicting device 20, and an intelligent replenishment device 30, and manages supply chains of a plurality of products as shown in fig. 1, for example. Each device may be implemented as a functional module by executing an application program stored in a memory by a processor, or may be implemented as a hardware configuration independent of each other. With regard to specific implementations, specific examples will be described hereinafter.

As shown in fig. 2, the product sorting apparatus 10 sorts a plurality of products based on history data. For example, the product classification apparatus 10 additionally considers the temporal characteristics of the product in addition to the conventional ABC classification method. For example, a latest sales time interval is added as an analysis dimension, and a plurality of kinds of commodities are classified as an R-ABC classification method with ABC classification results based on sales data as history data and the latest sales time interval as two dimensions. Therefore, the ABC classification is further refined, and more valuable commodities can be selected from the traditional ABC classification. Further, in order to further clarify the sales characteristics of the products, the product classification device 10 may calculate a product characteristic analysis index for each product based on the history data. The article characteristic analysis index includes, for example, at least one of the following: demand probability, sales frequency, variation coefficient, variation amplitude, customer concentration, sales tempo, distribution interval and the like. Thus, based on the classification result (preliminary classification result) of the improved R-ABC classification method, multidimensional subdivision is performed using the calculated characteristic analysis indexes of each product, as a final classification result. Further, based on the result of the division, a commodity requiring confirmation of the result of the demand prediction by the salesperson is extracted. Therefore, the accuracy of demand forecast can be improved by effectively combining the business experience of the salesperson.

The sales prediction apparatus 20 predicts sales for each product based on the history data and the classification result of the product classification apparatus 10 for a plurality of types of products. For example, the sales predicting device 20 predicts sales for each product for a predetermined time (for example, 1 day, 1 week, 1 month, 1 year, etc.) for the product extracted by the product sorting device 10, for example, for which the demand predicting result needs to be confirmed, based on the sorting result of the product sorting device 10.

Specifically, the sales prediction apparatus 20 may construct a long/short term memory network point prediction model corresponding to a prediction time (for example, 1 day, 1 week, or 1 year) by using, for example, point prediction based on a long/short term memory network, perform data collection, data processing, model construction, model evaluation, and model deployment, and output a prediction result corresponding to the prediction time. Alternatively, the sales prediction apparatus 20 may construct a probability density section prediction model corresponding to a prediction time (for example, 1 month) by using probability density section prediction by deep learning (for example, deep ar method), perform data collection, data processing, model construction, and prediction result processing, and output a prediction result corresponding to the prediction time.

Further, the sales prediction apparatus 20 may build a plurality of prediction models corresponding to the respective prediction times among the plurality of prediction times by using a hybrid prediction combining the point prediction and the section prediction, and output prediction results corresponding to the respective prediction times. For example, the sales prediction apparatus 20 may build a long-term memory network point prediction model corresponding to 1 day, a probability density interval prediction model corresponding to 1 month, and a long-term memory network point prediction model corresponding to 1 year, and output prediction results corresponding to 1 day, 1 month, and 1 year, respectively. By combining the classification result of the product, the time to be predicted, and other factors, a more accurate prediction result corresponding to the required prediction time can be obtained by using the most appropriate prediction method.

The intelligent replenishment device 30 applies an automatic replenishment model based on the prediction result of the sales prediction device 20 and the commodity characteristic analysis index to generate a replenishment decision. For example, the intelligent replenishment device 30 creates different replenishment strategies for the characteristics of different products, and creates a "regular and irregular", a "regular and regular", and an "irregular and irregular" replenishment plan.

Specifically, the intelligent replenishment device 30 can generate a "regular and irregular" or "irregular and irregular" replenishment plan using a replenishment model based on point prediction. Alternatively, the intelligent replenishment device 30 may generate a "regular and quantitative" replenishment plan using a mixed replenishment model based on the point prediction and the section prediction. Alternatively, the intelligent replenishment device 30 may generate a "non-regular and non-regular" replenishment plan using a replenishment model based on section prediction. In this way, a replenishment plan is generated by using the most appropriate replenishment model according to factors such as different replenishment periods, and a replenishment decision more satisfying the demand can be provided.

Fig. 2 shows an example in which the intelligent supply chain system 100 is connected to an external database 200, and history data is acquired from the external database 200. However, the present embodiment is not limited thereto, and the database 200 may be stored in a built-in storage device of the smart supply chain system 100. In addition, the connection between the intelligent supply chain system 100 and the external database 200 may be various wired or wireless connections, which is not limited herein.

According to the intelligent supply chain system 100 of the embodiment, supply chain management is not only supply chain, but also depends on big data and information systems, three important blocks of commodity classification, sales prediction and intelligent replenishment are effectively integrated, internal and external data elements are integrated, and all systems coordinate and act together under the guidance of information, so that service energy is condensed to the maximum extent, and service capacity is released orderly. Therefore, the service can be accurate, the supply chain is transparent, flexible and agile, and the functions are more coordinated.

Two specific examples of the hardware configuration of the intelligent supply chain system 100 according to the present embodiment will be described below. Fig. 3 is a block diagram showing an example of a hardware configuration of the intelligent supply chain system 100 according to the first embodiment of the present invention. As shown in fig. 3, the intelligent supply chain system 100 implements the functional modules shown in fig. 2 through a computer system having a processor 110, a memory 120, an interface 130, an input device 140, and a display portion 150. The processor 110, the memory 120, the interface 130, the input device 140, and the display unit 150 are connected to each other via a bus 160.

Specifically, the processor 110 is, for example, a CPU, a microprocessor, or the like, and executes an application program stored in the memory 120 to realize the functions of each device of the smart supply chain system 100. The interface 130 is, for example, a communication interface, and is capable of data communication with the database 200. The input device 140 is an input device such as a keyboard, a mouse, a microphone, etc. for a user to input an instruction. The display unit 150 is, for example, a liquid crystal display, and can display a screen related to the processing procedure and the result of the smart supply chain system 100.

The intelligent supply chain system 100 based on the hardware configuration of this example can be installed in, for example, a sales store, a supplier, and a manufacturer as application targets, and used by the sales store, the supplier, and the manufacturer as users.

Fig. 4 is a block diagram showing another example of the hardware configuration of the intelligent supply chain system 100 according to the first embodiment of the present invention. As shown in FIG. 4, the intelligent supply chain system 100 includes a server platform 100A and a client device 100B that are capable of data communication with each other. The server platform 100A has a processor 110A, a memory 120A, and an interface 130A connected to each other via a bus 160A, and the client device 100B has a processor 110B, a memory 120B, an interface 130B, an input device 140, and a display section 150 connected to each other via a bus 160B.

Specifically, the processor 110A of the server platform 100A is, for example, a CPU, a microprocessor, or the like, and executes an application program stored in the memory 120A to realize the functions of each device of the intelligent supply chain system 100. The interface 130A of the server platform 100A is, for example, a communication interface, and is capable of data communication with the interface 130B of the client device 100B and/or the database 200.

The processor 110B of the client device 100B is, for example, a CPU, a microprocessor, or the like, and executes an application program stored in the memory 120B to implement the function of a user interface. The input device 140 is, for example, an instruction input device such as a keyboard, a mouse, or a microphone. The display unit 150 is, for example, a liquid crystal display, and can display a screen related to the processing procedure and the result of the smart supply chain system 100.

In the intelligent supply chain system 100 based on the hardware configuration of this example, the server platform 100A may be provided at a service provider, and the client device 100B may be provided at a sales store, a vendor, or a manufacturer, which is an application target. Thus, the service provider provides the aforementioned functions of the intelligent supply chain system 100 to the sales store, the supplier, and the manufacturer as the user.

The following describes a method performed by the intelligent supply chain system 100 of the present embodiment. Fig. 5 is a flowchart illustrating a method executed by the intelligent supply chain system 100 according to the first embodiment of the present invention. Here, an example of the method executed by the intelligent supply chain system 100 according to the present embodiment is specifically described, but the method executed by the intelligent supply chain system 100 according to the present embodiment is not limited to the following example.

As shown in fig. 5, in step S1, the article sorting apparatus 100 sorts the articles so as to adopt corresponding sales and stock policies for different categories of articles. Specifically, based on the R-ABC classification method, the products are subdivided in combination with each product characteristic analysis index, and the products requiring the confirmation of the demand prediction result by the salesperson are selected from the subdivided products.

In step S2, the sales predicting apparatus 20 performs sales prediction to make a demand plan. As described above, the sales prediction apparatus 20 may use the point prediction based on the long-and-short term memory network or the probability density interval prediction based on the DeepAR.

In step S3, the intelligent replenishment device 30 applies the automatic replenishment model to a specific scenario to generate a replenishment decision. As described above, the intelligent replenishment device 30 may use a replenishment model based on point prediction, a mixed replenishment model based on point prediction and section prediction, or a replenishment model based on section prediction.

(second embodiment)

Fig. 6 is a functional block diagram of an intelligent supply chain system according to a second embodiment of the present invention. As shown in fig. 6, the sales prediction apparatus 20 of the present embodiment includes a point prediction unit 21 and a section prediction unit 22. The point prediction unit 21 performs point prediction based on the long/short term memory network, and the section prediction unit 22 performs section prediction based on the probability density. The intelligent replenishment device 30 includes a prediction result collection unit 31 and a mixed replenishment unit 32. The prediction result collection unit 31 acquires the prediction result from the predicted sales device 20. The mixed replenishment unit 32 generates a regular and quantitative replenishment decision based on the prediction result from the sales prediction apparatus 20 acquired by the collection unit 301, specifically, using the prediction results based on the point prediction and the section prediction.

Fig. 7 is a flowchart illustrating a method executed by the intelligent supply chain system 100 according to a second embodiment of the present invention. Here, an example of the method executed by the intelligent supply chain system 100 according to the present embodiment is specifically described, but the method executed by the intelligent supply chain system 100 according to the present embodiment is not limited to the following example.

As shown in fig. 7, in step S1, the article sorting apparatus 100 sorts the articles so as to adopt corresponding sales and stock policies for different categories of articles. Specifically, based on the R-ABC classification method, the products are subdivided in combination with each product characteristic analysis index, and the products requiring the confirmation of the demand prediction result by the salesperson are selected from the subdivided products.

In step S3, the intelligent replenishment device 30 applies the hybrid replenishment model to a specific scenario to generate a replenishment decision, and generates a replenishment plan in combination with the actual inventory. Specifically, the mixed replenishment unit 32 of the intelligent replenishment device 30 generates a regular and quantitative replenishment plan by using a mixed replenishment model based on the prediction result from the predicted sales device 20 acquired by the prediction result collection unit 31. The hybrid replenishment model is based on point prediction and interval prediction. The regular and quantitative mixed replenishment model based on point prediction and interval prediction comprises regular replenishment and quantitative replenishment.

Fig. 8 is a functional block diagram showing the intelligent replenishment device 30 in the intelligent supply chain system 100 according to the second embodiment of the present invention. As shown in fig. 8, the intelligent replenishment device 30 includes a prediction result collection unit 31 and a mixed replenishment unit 32. The prediction result collection unit 31 acquires the prediction result from the predicted sales device 20. The mixed replenishment unit 32 generates a regular and quantitative replenishment strategy based on the prediction results based on the point prediction and the section prediction among the prediction results from the sales prediction device 20.

The mixed replenishment part 32 includes: a replenishment cycle calculation unit 321 that calculates a replenishment cycle as an optimal economic replenishment cycle Tr based on the annual sales prediction amount Qy and the optimal economic replenishment amount EOQ in the prediction results acquired by the prediction result collection unit 31; a safety stock calculation unit 322 that calculates a safety stock amount based on the sales volume prediction and the safety factor corresponding to the actual prediction error and the service level, specifically, based on the monthly demand standard deviation, the replenishment cycle, and the safety factor corresponding to the service level; the restocking point calculation unit 323 calculates a restocking point based on the sales predicted amount in the replenishment lead period and the safety stock amount calculated by the safety stock calculation unit 322; the replenishment quantity calculation unit 324 calculates the replenishment quantity as the actual replenishment quantity based on the predicted quantity of sales in the replenishment lead period, the predicted quantity of sales in the replenishment period, and the above-mentioned safe stock quantity and current stock quantity; the replenishment triggering determination unit 325 determines the current stock quantity and the replenishment point calculated by the replenishment point calculation unit 323; and a replenishment plan generating unit 326 that generates a replenishment plan based on the optimum economic replenishment cycle Tr calculated by the replenishment cycle calculating unit 321 and the replenishment quantity Qr calculated by the replenishment quantity calculating unit 324 as the actual replenishment quantity.

Fig. 9 is a diagram for explaining the establishment of the optimal economic replenishment quantity EOQ based on the EOQ model. Fig. 9 (a) shows the relationship between holding cost and order cost, and the EOQ model, i.e., the economic order lot model, can balance holding cost and order placement cost. As shown in fig. 9 (B), according to model EOQ, when the cost of the next order is equal to the cost of ownership, the total cost is the lowest, and the order quantity is the optimal economic replenishment quantity EOQ.

The EOQ model-based ordering method can be applied to, for example, a case where a product is mass-produced or purchased, or a case where the ordering cost and the holding cost are known.

The EOQ model is explained in detail below.

The total cost TC is the purchase cost (purchase cost) + lower order cost (order cost) + holding cost (holding cost) + short goods cost (short cost)

Namely:

wherein, C_eFor each order in the cycleCost, C_tThe keeping cost of each commodity in the period, D is the demand in the period, Q is the replenishment quantity, c is the variable cost, generally referred to as unit purchase cost, CsE [ Units reports)]The cost of out-of-stock.

If one wants to solve the minimum cost: min { tc (q) }, the first derivative is first set to 0 to obtain the extreme point:

since the first derivative can only be solved for verification as an extremum (which may be a maximum or minimum) and for verification as a minimum, the second derivative of the function is required:

when C is present_tD, Q are all positive, the second derivative is positive, so Q is minimal, so the optimal economic order is:

the replenishment cycle calculation means 321 in the present embodiment calculates Q as the optimum economic replenishment quantity EOQ in the normal case from the equation (1) based on the annual sales quantity prediction quantity Qy in the prediction result obtained by the prediction result collection unit 31 and the economic order lot quantity model which is the EOQ model, and obtains the replenishment cycle in the normal case as the optimum economic replenishment cycle Tr from the annual sales quantity prediction quantity Qy and the calculated optimum economic replenishment quantity EOQ.

As described above, in formula (1), C_eCost per order in a cycle, C_tD is an in-cycle demand amount, which is based on a point prediction or probability density interval prediction result, for example, by selecting a confidence interval of an in-cycle sales amount prediction by a user through an external input, and selecting a use demand lower limit X or a demand upper limit Y or a demand median M as the in-cycle demand amount D.

The optimal economic replenishment period Tr can be calculated by the following formula (2):

tr 365/(Qy/best economic replenishment EOQ) (formula 2)

Wherein Tr is the optimal economic replenishment period, and Qy is the annual sales forecast.

During the replenishment period, if no shortage or special condition occurs, the replenishment is performed at regular intervals. The regular replenishment is performed in accordance with the optimum economic replenishment cycle Tr calculated by the replenishment cycle calculation unit 321.

However, in consideration of the risk of stock shortage due to fluctuation in demand in actual management, in this case, it is necessary to replenish the stock in advance in a replenishment cycle of regular replenishment, and in this case, replenishment can be performed in combination with a quantitative replenishment logic.

In the quantitative replenishment logic, the safety stock calculation unit 322 calculates the safety stock SS based on the monthly demand standard deviation, the replenishment cycle, and the safety factor Z corresponding to the service level, thereby coping with an error of the sales amount prediction from the actual sales amount by setting the safety stock. The safe stock SS is calculated by the following calculation method:

safety stock (standard deviation of monthly demand/30 replenishment cycles) safety factor corresponding to service level.

The restocking point calculation unit 323 calculates a restocking point ROP based on the predicted amount of sales within the replenishment lead period and the safety stock SS calculated by the safety stock calculation unit 322. The restocking point ROP may be calculated by:

and the re-replenishing point is the sales forecast plus the safe stock in the replenishing lead period.

The replenishment quantity calculation unit 324 calculates the replenishment quantity Qr based on the predicted quantity of sales in the replenishment lead period, the predicted quantity of sales in the replenishment period, and the above-described safe stock quantity and current stock quantity. The replenishment quantity Qr can be calculated by the following calculation:

and (3) the quantity of the replenishment is 2, the quantity of the sales predicted in the replenishment lead period, the quantity of the sales predicted in the replenishment period, the safety stock and the current stock.

The replenishment trigger determination unit 325 determines the current stock quantity and the replenishment point ROP calculated by the replenishment point calculation unit 323, and when it is determined that the current stock quantity is larger than the replenishment point ROP, the replenishment plan generation unit 326 generates a replenishment plan based on the optimum economic replenishment cycle Tr calculated by the replenishment cycle calculation unit 321 and the replenishment quantity Qr calculated as the actual replenishment quantity by the replenishment quantity calculation unit 324.

When the current stock quantity is determined to be equal to or less than the restocking point ROP, the restocking trigger determination unit 325 determines that the restocking operation is triggered, and it is necessary to perform a restocking operation in advance in a restocking cycle of the regular restocking, and the restocking plan generation unit 326 generates a restocking plan based on the restocking amount Qr that is the actual restocking amount calculated by the restocking amount calculation unit 324.

Therefore, by adopting the optimal economic replenishment period calculated based on the EOQ model to perform regular replenishment, under the condition that the forecast is accurate and no shortage occurs, the cost can be well controlled, and the traffic volume when the replenishment business is managed is reduced. However, in actual management, the risk of shortage due to fluctuation in demand may occur, and in some cases, replenishment may be performed in advance in a replenishment cycle of regular replenishment, and at this time, replenishment may be performed in accordance with a quantitative replenishment logic. Therefore, by means of sales forecast and a model combining regular replenishment and quantitative replenishment, the system can deal with the fluctuation of emergency demands under the condition of controlling cost, and provides a perfect replenishment strategy for enterprises.

In step S101, the prediction result collection unit 31 acquires the prediction result from the predicted sales device 20.

In step S102, the replenishment cycle calculation unit 321 calculates the replenishment cycle as the optimum economic replenishment cycle Tr based on the annual quantity prediction amount Qy and the optimum economic replenishment amount EOQ in the prediction results acquired by the prediction result collection unit 31.

In step S103, the safety stock calculation unit 322 calculates the safety stock amount based on the monthly demand standard deviation, the replenishment cycle, and the safety factor corresponding to the service level.

In step S104, the replenishment quantity calculation unit 324 calculates the replenishment quantity based on the predicted quantity of sales in the replenishment lead period, the predicted quantity of sales in the replenishment period, and the above-described safe stock quantity and current stock quantity.

In step S105, the replenishment plan generating unit 326 generates a replenishment plan as a regular replenishment plan based on the optimum economic replenishment cycle Tr calculated by the replenishment cycle calculating unit 321 and the replenishment quantity Qr calculated as the actual replenishment quantity by the replenishment quantity calculating unit 324.

Fig. 11 is a flowchart illustrating a hybrid replenishment method of an intelligent capture device in an intelligent supply chain system according to a second embodiment of the present invention. In step S301, the prediction result collection unit 31 acquires the prediction result from the predicted sales device 20.

In step S302, the replenishment cycle calculation unit 321 calculates a replenishment cycle as an optimum economic replenishment cycle Tr based on the annual quantity forecast Qy and the optimum economic replenishment quantity EOQ.

In step S303, the safety stock calculation unit 322 calculates the safety stock amount based on the standard deviation of the monthly demand, the replenishment cycle, and the safety factor corresponding to the service level.

In step S304, the restocking point calculation unit 323 calculates a restocking point based on the predicted quantity of sales in the replenishment lead period and the safe stock quantity; the replenishment quantity calculation unit 324 calculates the replenishment quantity based on the predicted quantity of sales in the replenishment lead period, the predicted quantity of sales in the replenishment period, and the above-described safe stock quantity and current stock quantity.

In step S305, the replenishment trigger determination unit 325 determines the current stock quantity and the replenishment point ROP, and if it determines that the current stock quantity is larger than the replenishment point ROP, the flow proceeds to step S306, and the replenishment plan generation unit 326 generates a replenishment plan as a regular replenishment plan based on the optimal economic replenishment cycle Tr and the replenishment quantity Qr that is the actual replenishment quantity.

When it is determined in step S305 that the current stock quantity is equal to or less than the restocking point ROP, the restocking trigger determination unit 325 determines that the restocking operation is triggered and it is necessary to perform an advance restocking within the restocking cycle of the regular restocking, and the flow proceeds to step S307, where the restocking plan generation unit 326 generates a restocking plan as a regular and quantitative restocking plan based on the restocking amount Qr that is the actual restocking amount calculated by the restocking amount calculation unit 324.

(specific examples)

One specific example of the present invention is described below. The specific example can be applied to the hardware configuration shown in fig. 3, and can also be applied to the hardware configuration shown in fig. 4.

This example is used to illustrate the application of the model to the vendor. Fig. 12 is a flowchart showing an application example in the case where the intelligent supply chain system 100 of the present invention is applied to a supplier.

As shown in fig. 12, in step S201, the product classification apparatus 10 classifies the products by the R-ABC classification method.

The article classification device 10 classifies articles by the R-ABC classification method. Specifically, suppose that the sales amounts of the respective commodities in the past three months of a certain sales store are ranked from large to small in total, the percentage of the sales amount of each commodity in the total sales amount is calculated, and then the cumulative percentage is calculated. According to the definition of ABC classification, the cumulative proportion is 0% -80%, and is class A, the cumulative proportion is 80% -90%, and is class B, and the cumulative proportion is 90% -100%, and is class C. Wherein, the proportion here can be adjusted according to the actual conditions.

However, the time characteristic of a commodity is not considered in the conventional ABC analysis method, and for example, although the total sales volume of a certain commodity is 100, the importance of the commodity is different if the 100 sales records include recent sales records or sales records before 2 months. The dimension of the 'nearest sale interval' is added, so that the problem can be well solved. On the basis of ABC classification, commodities are classified into three levels according to the time dimension according to the latest sales interval of each commodity. Thus, the commercial products were classified into categories a1, a2, A3, B1, B2, B3, C1, C2, and C3. The number of stages into which the product is divided in the time dimension is not limited to this, and may be set as appropriate according to the actual situation.

Further, the values of the following commodity characteristic analysis indexes are calculated as follows. With respect to demand probabilities, how much probability over a unit period will result in sales performance is calculated. The frequency of sales is determined by counting sales results on a weekly basis and counting the number of weeks in which actual results are present as the frequency. As for the variation coefficient, the magnitude of the demand fluctuation is judged using the variation coefficient. Regarding the fluctuation range, the fluctuation of the total amount of demand in a predetermined period is checked. Regarding customer concentration, it is confirmed whether the sales in a specified period are concentrated on a few customers. Regarding the sales tempo, the time interval from each sales record to the next sales record is counted in units of commodities. The distribution interval is counted in units of the product, and the earliest sale date of the product and the time interval until the current analysis time point are counted, and the longer the interval, the longer the distribution time, and the shorter the interval, the shorter the distribution time.

Thus, according to the commodity characteristic analysis index, commodities which are high in sales frequency and have relatively sufficient data samples are extracted based on the evaluation rule, and the commodities are suitable for predicting the future demand through data analysis and machine learning.

Next, in step S202, the sales predicting apparatus 20 acquires the historical data of the suppliers supplying the different stores, and the attributes thereof are as follows:

firstly, data exploration is carried out, and the exploration mode may include: fill the negative sales sample (indicating return) with 0; if the sales volume of a certain commodity on a certain day is not recorded in the historical data, filling the sales volume with 0; price misses may be filled in using the mean or sales volume of the last day. And continuously exploring the correlation between the information obtained in the data collection stage and the sales volume, calculating the correlation between the sales volume and each attribute, and finding out that the correlation between each attribute and the sales volume meets the requirement of model construction.

Secondly, data aggregation is carried out, and the data are aggregated into training samples according to different time granularities. The week prediction and the month prediction are carried out at the same time. According to historical data, daily prediction needs to construct a daily output sample by taking (product ID, store ID and date) as a main key, namely after aggregation, each product of each store in each day has a unique sample, and aggregation attributes including daily sales, daily output price, average value/standard deviation/maximum/minimum value of sales of the previous three days are constructed; the month interval prediction needs to calculate the month ID in a divided manner, the first month ID is 1, and the number is increased in such a manner. And (product ID, store ID and month ID) is used as a main key to construct a monthly shipment volume sample, namely after aggregation, each product of each store in each month has a unique sample. Simultaneously constructing the aggregate attributes includes: the total monthly sales, the mean monthly shipment price, the variance monthly shipment price, the maximum monthly shipment price, the minimum monthly shipment price, the number of shipment times in this month, whether to reduce the price in this month, whether to increase the price in this month, and the number of price changes in this month. The annual prediction needs to construct a daily shipment sample by taking (product ID, store ID and year ID) as a main key, namely after aggregation, each product of each store in each year has a unique sample, and the aggregation attributes are constructed to include characteristics such as annual total sales, annual shipment price mean/variance/maximum/minimum value, annual shipment frequency and the like.

The daily prediction model predicts the shipment volume of the 5 days later by using the historical data of the previous 14 days, so that training samples are generated by using a time sliding window, namely one training sample comprises the shipment volume of the previous 14 months, store information and relevant characteristics constructed in the previous step, the shipment volume of the 5 days later is predicted, and the interval of the time sliding window is 1 day.

The monthly interval prediction model predicts the shipment volume situation of 2 months later by using historical data of the previous 5 months, so that training samples are generated by using a time sliding window, namely one training sample comprises the shipment volume of the previous 5 months, store information and related characteristics constructed in the previous step, the shipment volume of 2 months later is predicted, and the interval of the time sliding window is 1 month.

The annual forecasting model forecasts the shipment volume of 1 year later by using historical data of the previous 3 years, so training samples are generated by using a time sliding window, namely one training sample comprises the shipment volume of the previous 3 years, store information and relevant characteristics constructed in the previous step, the shipment volume of the next 1 year is forecasted, and the interval of the time sliding window is 1 year.

Then, model construction is performed. And (3) constructing a long-term and short-term memory network model by utilizing keras for daily prediction and annual prediction, and correcting the output of the model by using a relu function to ensure the characteristic that the commodity demand cannot be less than 0. And designing a biased loss function to improve the prediction accuracy, which is defined as

Wherein y _ truth is a real value, y _ pred is a prediction result, w is a preset weight, and w > 1.

Training a model by using an Adam optimization method, wherein relevant parameters are as follows: the learning rate is equal to 0.001, the number of samples in a batch is 32, and the number of training rounds is 15.

The moon prediction utilizes a GluONTS library to construct a DeepAR model, training parameters are set as epoch being 100, learning _ rate being 1e-3, output distribution is set as Gaussian distribution, and an RNN framework is LSTM.

In the process of predicting the result, a full-automatic processing program is established in the steps, the full-automatic processing program comprises a data processing system, a model rolling training system and a result storage and output system, and the model is put into production.

And the prediction result is output in a proper form such as a webpage, a report form and the like after being correspondingly post-processed according to the difference of the prediction subjects. In this example, a prediction result table is constructed in the database, and the structure and the representation are shown in table 1 and table 2, for example.

Table 1:

table 2:

next, in step S203, the intelligent replenishment device 30 generates a replenishment plan based on the prediction result in step S202. For suppliers, it is generally necessary to meet the order of customers as much as possible, but there may be a case where the market feedback is delayed without directly contacting the final market, so that there is a need for a certain risk coping capability for stock, and the cost is controlled to optimize the operation.

For example, for the supplier replenishment scenario, the annual forecast, monthly forecast, and daily forecast data are used to define the service level coefficient as 95%, the current stock quantity is 45, the order lead period is 5 days, the order cost per time is 50 yuan, and the holding cost per commodity is 30 yuan per commodity. The forecast median M is selected for the daily stock quantity and the demand quantity in the period.

(1) Calculating an optimal economic replenishment period based on the EOQ model

C_eCost per order in a cycle, C_tD is the storage cost per commodity in the cycle, and D is the demand (predicted median M) in the cycle.

Taking the above commercial product 09 as an example (annual sales forecast of 3985 using annual forecast):

the optimal economic replenishment period is 365/(annual sales forecast/optimal economic replenishment) is 365/(3985/69) is 6

(2) Safety stock preparation

The security stock of the commercial product 09 is (standard deviation of monthly demand/30 × 6) × service level coefficient is (23/30 × 6) × 1.68 ═ 8

(3) Re-replenishment spot customization

The restocking point of the product 09 is 33+8 and 41, respectively, the sales forecast amount in the replenishment lead period and the safety stock

(4) Make-up quantity

The replenishment quantity of the commercial product 09 at 2020.3.1 is 2 × the predicted quantity of sales in the replenishment lead period + the predicted quantity of sales in the replenishment period + the stock of safety — the current stock is 2 × 33+ (325/30 × 6) +8-45 is 94

Taking the data in the "prediction result table" as an example, the replenishment plan is shown in table 3.

Table 3:

meaning that when the above-mentioned commercial product 09 is in 2020.3.1-2020.4.1 cycle, regular replenishment is performed every 6 days, and if the stock is not full 41 in the replenishment period, replenishment is started, 2020.3.1 is performed for 94 pieces for the first time.

The embodiments and specific examples of the present invention have been described above with reference to the accompanying drawings. The above-described embodiments and specific examples are merely specific examples of the present invention and are not intended to limit the scope of the present invention. Those skilled in the art can modify the embodiments and specific examples based on the technical idea of the present invention, and various modifications, combinations, and appropriate omissions of the elements can be made, and the embodiments obtained thereby are also included in the scope of the present invention. For example, the above embodiments and specific examples may be combined with each other, and the combined embodiments are also included in the scope of the present invention.

Claims

1. An intelligent supply chain system for managing supply chains of a plurality of kinds of commodities, comprising:

a commodity classification device for classifying a plurality of commodities based on the history data;

a sales prediction device that predicts sales for each of a plurality of types of commodities based on historical data and the classification result of the commodity classification device for the plurality of types of commodities; and

an intelligent replenishment device which applies an automatic replenishment model based on the prediction result of the sales prediction device to generate a replenishment decision,

the intelligent replenishment device comprises a mixed replenishment part, and the mixed replenishment part generates a periodic and quantitative replenishment decision based on the prediction results of the point prediction and the interval prediction of the sales prediction device.

2. The intelligent supply chain system of claim 1,

the mixed replenishment part includes:

a replenishment cycle calculation unit that calculates a replenishment cycle as an optimal economic replenishment cycle based on the annual sales amount prediction amount and the optimal economic replenishment amount in the prediction result;

a safety stock calculation unit that calculates a safety stock amount based on the sales volume prediction and the safety factor corresponding to the actual prediction error and the service level;

the replenishment quantity calculation unit is used for calculating the replenishment quantity as the actual replenishment quantity based on the predicted quantity of the sales in the replenishment lead period, the predicted quantity of the sales in the replenishment period, the safe stock and the current stock; and

and a replenishment plan generating unit that generates a replenishment plan based on the optimum economic replenishment cycle and the replenishment quantity calculated by the replenishment quantity calculating unit as an actual replenishment quantity.

3. The intelligent supply chain system of claim 2,

the mixed replenishment part further includes:

the restocking point calculating unit calculates a restocking point based on the predicted sales amount in the advance period of restocking and the safe stock; and

and the replenishment triggering and judging unit is used for comparing the current inventory with the replenishment point and judging whether to trigger replenishment operation or not based on the comparison result.

4. The intelligent supply chain system of claim 3,

when the current stock amount is equal to or less than the restocking point, the restocking trigger determination unit determines that a restocking operation is triggered, and the restocking plan generation unit generates a restocking plan based on the restocking amount calculated by the restocking amount calculation unit as an actual restocking amount.

5. The intelligent supply chain system as in claims 2-4,

the replenishment cycle calculation unit calculates the optimal economic replenishment quantity based on an EOQ model, namely an economic replenishment quantity model, according to the order cost of each commodity in the cycle, the keeping cost of each commodity in the cycle and the demand quantity in the cycle based on the probability density interval prediction result.

6. A server platform for managing a supply chain of a plurality of commodities, comprising a processor, a memory and an interface, capable of data communication with a client device via the interface,

the processor executes the program stored in the memory to perform the following processing:

a commodity classification process for classifying a plurality of commodities based on historical data;

a sales prediction process of performing sales prediction for each commodity based on historical data and a classification result of the commodity classification process for a plurality of commodities; and

an intelligent replenishment process for applying an automatic replenishment model based on a prediction result of the sales prediction process to generate a replenishment decision,

in the replenishment intelligent processing, a regular and quantitative replenishment decision is generated based on the point prediction and interval prediction results obtained by the sales prediction processing,

the results of the sales prediction process and/or the replenishment decision are sent to the client device via the interface.