A method and device for determining safety stock

The probability distribution map of the total inventory of items is generated through the kernel density estimation calculation method. Combined with the out-of-stock demand, the shortcomings of the assumption of normal distribution in the existing technology are solved, and the calculation accuracy and service level of safe inventory are improved.

CN113780912BActive Publication Date: 2024-12-03BEIJING JINGDONG SHANGKE INFORMATION TECH CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202010997100.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-09-21
Publication Date
2024-12-03
Estimated Expiration
2040-09-21

AI Technical Summary

Technical Problem

In the prior art, when calculating safe inventory, normal distribution functions are often used to result in low calculation accuracy and inability to effectively match the irregular demand distribution of items.

Method used

The kernel density estimation calculation method is used to generate a probability distribution map of the total inventory of items based on historical outbound data, and combined with the probability of satisfaction of outbound demands in the prediction period, the total inventory of target items is determined and the safe inventory is calculated.

Benefits of technology

Improve the prediction accuracy of safe inventory, ensure that inventory meets outbound demand, reduce the risk of insufficient inventory, and thus improve service level.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113780912B_ABST
    Figure CN113780912B_ABST
Patent Text Reader

Abstract

The present invention discloses a method and device for determining safety stock, and relates to the field of computer technology. A specific implementation of the method includes: obtaining the outbound data of items in at least one historical period; determining the probability distribution graph corresponding to the total inventory of items using a kernel density estimation algorithm based on the outbound data of items in at least one historical period; determining the total inventory of target items in the forecast period based on the probability distribution graph; determining the safety stock in the forecast period based on the total inventory of target items and the turnover inventory in the forecast period. This implementation improves the calculation accuracy of the safety stock.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of warehousing logistics, and in particular to a method and device for determining safety stock. Background Art

[0002] Safety stock is the inventory prepared in advance to meet the demand that exceeds the forecast quantity within a set period (such as when a large number of sudden orders appear). If the safety stock quantity of an item is too large, it will lead to an increase in storage costs. If the safety stock quantity of an item is too small, it may lead to a decrease in service level. Therefore, accurately calculating the amount of safety stock is very important for controlling storage costs and ensuring service levels.

[0003] At present, the inverse function of the normal distribution function is usually used to calculate the amount of safety stock. However, since the value range of the normal distribution function is (-∞, +∞), and the inventory of an item cannot be a negative number, using the normal distribution function to calculate the safety stock of an item will result in a low accuracy rate in calculating the safety stock. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides a method and device for determining safety stock, which can improve the calculation accuracy of safety stock.

[0005] To achieve the above objective, according to one aspect of an embodiment of the present invention, a method for determining safety stock is provided.

[0006] A method for determining safety stock according to an embodiment of the present invention includes: obtaining data on the shipment of items in at least one historical period;

[0007] Determine a probability distribution graph corresponding to the total inventory of items using a kernel density estimation algorithm based on the outbound data of items in the at least one historical period;

[0008] Determine the total inventory of target items within the forecast period according to the probability distribution graph;

[0009] The safety stock within the forecast period is determined according to the total stock of the target item and the turnover stock within the forecast period.

[0010] Optionally, the total inventory of the target items is determined based on the probability distribution graph and the probability of satisfying the outbound demand within the forecast period.

[0011] Optionally, determining the total inventory of the target item according to the probability distribution graph and the probability of satisfying the outbound demand within the forecast period includes:

[0012] According to the probability distribution graph, first distribution probabilities corresponding to different total inventories of items are calculated respectively;

[0013] According to the size order of the outbound data in the at least one historical period indicated by the probability distribution graph, the first distribution probabilities corresponding to the different total inventories of items are accumulated in sequence to obtain the accumulated probabilities corresponding to the different total inventories of items;

[0014] A target cumulative probability whose cumulative probability is not less than the probability of satisfying the outbound demand and whose difference with the probability of satisfying the outbound demand is the smallest is determined, and the target total inventory is determined according to the target cumulative probability.

[0015] Optionally, the probability distribution graph further indicates a second distribution probability of the outbound data; and calculating the first distribution probabilities corresponding to different total inventories of items according to the probability distribution graph includes:

[0016] According to the size order, first intervals between the multiple outbound data of different sizes are respectively split to obtain multiple second intervals and second distribution probabilities respectively corresponding to the multiple second intervals;

[0017] The product of each second interval and the second distribution probability corresponding to itself is normalized, and the result of the normalization is used as the first distribution probability corresponding to the total inventory of items.

[0018] Optionally, determining the total inventory of the target items according to the target cumulative probability includes:

[0019] The outbound data corresponding to the last first distribution probability accumulated in the target cumulative probability is determined, and the outbound quantity indicated by the determined outbound data is used as the total inventory of the target items.

[0020] Optionally, the determining the probability distribution graph corresponding to the inventory of items by using a kernel density estimation algorithm according to the delivery data of items provided in the at least one historical period includes:

[0021] The bandwidth of the probability distribution diagram to be fitted is determined according to the number of the outbound data in the at least one historical period, and the outbound data in the at least one historical period is fitted with a Gaussian kernel according to the bandwidth to obtain the probability distribution diagram.

[0022] Optionally, the difference between the total inventory of the target item and the turnover inventory is used as the safety stock.

[0023] Optionally, the method further includes: allocating items to be replenished based on existing inventory and the safety stock.

[0024] To achieve the above objective, according to another aspect of an embodiment of the present invention, a device for determining safety stock is provided.

[0025] A device for determining safety stock in an embodiment of the present invention comprises: a data acquisition module, a probability determination module and an inventory determination module; wherein:

[0026] The data acquisition module is used to acquire the delivery data of the provided items in at least one historical period;

[0027] The probability determination module is used to determine the probability distribution diagram corresponding to the total inventory of items by using a kernel density estimation algorithm according to the outbound data of items in the at least one historical period;

[0028] The inventory determination module is used to determine the total inventory of target items within the forecast period according to the probability distribution diagram; and determine the safety inventory within the forecast period according to the total inventory of target items and the turnover inventory within the forecast period.

[0029] To achieve the above objective, according to another aspect of an embodiment of the present invention, an electronic device for determining safety stock is provided.

[0030] An electronic device for determining a safety stock according to an embodiment of the present invention comprises: one or more processors; a storage device for storing one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors implement a method for determining a safety stock according to an embodiment of the present invention.

[0031] To achieve the above objective, according to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided.

[0032] A computer-readable storage medium according to an embodiment of the present invention stores a computer program, and when the program is executed by a processor, a method for determining a safety stock according to an embodiment of the present invention is implemented.

[0033] One embodiment of the above invention has the following advantages or beneficial effects: by obtaining the outbound data of items in at least one historical period, and according to the obtained outbound data, using the kernel density estimation algorithm to determine the probability distribution map corresponding to the total inventory of items, then determining the total inventory of target items in the forecast period according to the probability distribution map, and then determining the safety stock in the forecast period according to the determined total inventory of target items and the turnover inventory in the forecast period. Since the kernel density estimation algorithm can generate the corresponding probability distribution according to the actual demand, it is not necessary to pre-assume that the actual demand obeys a known distribution (such as a normal distribution), that is, when the actual demand for items is irregularly distributed, this solution will not assume the irregular distribution as a normal distribution, but use the kernel density estimation algorithm to fit the probability distribution of the total inventory of items according to its actual irregular distribution. Therefore, compared with the prior art that assumes the actual demand for items in all cases as a normal distribution, this solution improves the fitting accuracy of the probability distribution of the actual demand for items, thereby improving the prediction accuracy of the safety stock. Moreover, the determined total inventory of target items matches the probability of satisfying the outbound delivery demand within the forecast period, for example, matches the cycle service level (CSL) within the forecast period. Then, the safety stock within the forecast period is determined based on the total inventory of target items, so that the determined safety stock can meet the outbound delivery demand (such as CSL) requirements, and then the items to be replenished are allocated based on the safety stock, which can reduce the risk of insufficient inventory of items within the preset period, thereby helping to improve the service level.

[0034] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The accompanying drawings are used to better understand the present invention and do not constitute an improper limitation of the present invention.

[0036] Figure 1 is a schematic diagram of main steps of a method for determining safety stock according to an embodiment of the present invention;

[0037] Figure 2 is a schematic diagram of a probability distribution diagram according to an embodiment of the present invention;

[0038] Figure 3 is a schematic diagram of main steps of another method for determining safety stock according to an embodiment of the present invention;

[0039] Figure 4 is a schematic diagram of main modules of a device for determining safety stock according to an embodiment of the present invention;

[0040] Figure 5 is an exemplary system architecture diagram to which embodiments of the present invention may be applied;

[0041] Figure 6 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0042] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0043] It should be pointed out that the embodiments of the present invention and the technical features therein may be combined with each other without conflict.

[0044] Figure 1 It is a schematic diagram of main steps of a method for determining safety stock according to an embodiment of the present invention.

[0045] like Figure 1 As shown, a method for determining safety stock in an embodiment of the present invention mainly includes the following steps:

[0046] Step S101: Obtaining the delivery data of items provided in at least one historical period.

[0047] The historical cycle can be a supply cycle in the past period of time. For example, when the supply cycle is one month, the outbound data of the items provided in the past month or several months can be obtained. The outbound data can be obtained based on the management data in the warehouse (outbound quantity and outbound time, etc.) or the sales data of the items (sales quantity and sales time, etc.). It can be understood that the obtained outbound data is multiple. For example, when obtaining the sales data of the items in the past six months, there is sales data generated every day in this six months, so the obtained sales data is about 180.

[0048] Step S102: according to the outbound data of the items provided in the at least one historical period, a probability distribution diagram corresponding to the total inventory of the items is determined by using a kernel density estimation algorithm.

[0049] The kernel density estimation algorithm belongs to a non-parametric algorithm. Since the non-parametric algorithm does not need to pre-set the specific form of the model and the error distribution in practical applications, it can obtain a wider nonlinear change. At the same time, when extracting samples to estimate the population, it does not need to rely on the distribution form of the population to which the sample belongs. It can be widely applied to different types of populations. Therefore, it is extremely useful for reducing deviations, improving prediction accuracy, and understanding the dynamic structure of sample sequences. In particular, when the population distribution is very irregular, the non-parametric algorithm can also provide a more accurate confidence interval. Based on this, in an embodiment of the present invention, a kernel density estimation algorithm is used to obtain the outbound data within at least one historical period as a sample to determine the probability distribution diagram of the total inventory of items. Therefore, even when the inventory distribution of items is irregular, the kernel density estimation algorithm as a non-parametric algorithm can more accurately determine the probabilities corresponding to the total inventory of different items.

[0050] When using the kernel density estimation algorithm to determine the probability distribution graph of the total inventory of items, the bandwidth of the probability distribution graph to be fitted can be determined based on the number of the outbound data in at least one historical period, and based on the bandwidth, a Gaussian kernel can be used to fit the outbound data in at least one historical period to obtain the probability distribution graph.

[0051] Specifically, the bandwidth of the probability distribution graph to be fitted can be determined by the following formula (1), and then the Gaussian kernel is used to fit the obtained multiple outbound data to determine the probability distribution graph corresponding to the total inventory of items. The expression of the Gaussian kernel can be shown in the following formula (2), and the kernel density estimation algorithm can be shown in the following formula (3):

[0052]

[0053] Among them, h represents the bandwidth of the probability distribution graph to be fitted, n represents the number of outbound data obtained in at least one historical period, Characterize the debugging parameters, It can be determined according to the principle of minimizing the mean square error.

[0054]

[0055]

[0056] Among them, a i represents the i-th outbound data, n is the number of outbound data obtained, h is the bandwidth, and y represents the number axis formed by arranging multiple outbound data in order of their size.

[0057] According to the above formulas (1) to (3), the kernel density estimation algorithm can be used to fit the obtained outbound data to generate a probability distribution diagram corresponding to the total inventory of items. The probability distribution diagram can be shown as Figure 2 As shown. Figure 2 In the figure, the horizontal axis (y) represents the number axis formed by arranging multiple outbound data in order of their size, and the vertical axis (f) represents the second distribution probability of the outbound data. Since the embodiment of the present invention determines the safety stock of the forecast period by the outbound data of the items in the historical period, the probability of the total inventory of the items in the forecast period can be further determined according to the distribution probability of the outbound data. That is, the first probability distribution of the total inventory of the items can be determined according to the second probability distribution of the outbound data. Then, Figure 2 It is not only a probability distribution diagram of outbound data, but also a probability distribution diagram of the total inventory of items. Since the outbound data of items in the historical period can reflect the actual demand for items in the historical period, the probability distribution diagram of the total inventory of items determined based on the outbound data can also reflect the actual demand for items, which is conducive to improving the accuracy of safety stock prediction.

[0058] For example, historical sales data of item A in the past month (a historical cycle) is obtained, where the number of item A sold each day from day 1 to day 10 is 100, then 100 appears 10 times in the historical sales data; the number of item A sold each day from day 11 to day 15 is 150, then 150 appears 5 times in the historical sales data; the number of item A sold each day from day 16 to day 28 is 80, then 80 appears 13 times in the historical sales data; the number of item A sold each day from day 29 to day 30 is 200, then 200 appears 2 times in the historical sales data. Therefore, in the probability distribution diagram, the horizontal axis represents the number axis formed by each historical sales data, that is, the arrangement of each historical sales data on the horizontal axis is 80, 100, 150 and 200 respectively, and the points on the curve represent the second distribution probabilities corresponding to the corresponding historical sales data, that is, the second distribution probability corresponding to 80 is 13 / 30, the second distribution probability corresponding to 100 is 10 / 30, the second distribution probability corresponding to 150 is 5 / 30, and the second distribution probability corresponding to 200 is 2 / 30. It can be understood that although this example uses sales data with discrete points and its corresponding second distribution probability is also a discrete point as an example, in the actual implementation process, the outbound data within multiple historical periods are generally obtained as the sample data of the kernel density estimation algorithm, and through the fitting of the kernel density estimation algorithm, the second distribution probability corresponding to each outbound data can be fitted into a continuous curve. That is to say, in the probability distribution diagram corresponding to the total inventory of items, the second distribution probabilities corresponding to multiple outbound data can form a continuous curve, and the formed continuous curve can be as follows Figure 2 As shown. Figure 2 In the figure, the horizontal axis is the number axis formed by arranging multiple outbound data in order of size. The point f(y i) is the second distribution probability corresponding to each outbound data.

[0059] Step S103: Determine the total inventory of target items within the forecast period according to the probability distribution graph.

[0060] In one embodiment of the present invention, when determining the total inventory of target items, the total inventory of target items can be determined based on the probability distribution diagram and the probability of meeting the outbound demand within the forecast period. The probability of meeting the outbound demand within the forecast period can be characterized by the customer service level or the cycle service level (CSL). CSL represents the probability that the existing inventory can meet customer demand within a cycle, which is a common indicator used to measure inventory and demand forecasting in supply chain management.

[0061] In an embodiment of the present invention, the first distribution probabilities corresponding to different total inventories of items can be calculated according to the probability distribution graph; the first distribution probabilities corresponding to the different total inventories of items are accumulated in turn according to the size order of the outbound data in the at least one historical period indicated by the probability distribution graph to obtain the cumulative probabilities corresponding to the different total inventories of items; the target cumulative probability is determined, the cumulative probability being not less than the probability of satisfying the outbound demand and having the smallest difference with the probability of satisfying the outbound demand, and the target total inventory of items is determined according to the target cumulative probability.

[0062] Specifically, when calculating the probabilities corresponding to the total inventory of different items according to the probability distribution diagram, the first intervals between the multiple outbound data of different sizes can be split according to the size order to obtain multiple second intervals and the second distribution probabilities corresponding to the multiple second intervals; the product of each second interval and the second distribution probability corresponding to itself is normalized, and the result of the normalization is used as the first distribution probability corresponding to the total inventory of items.

[0063] Since the probability distribution function determined by the kernel density estimation algorithm has no explicit expression, its corresponding inverse function cannot be obtained. Generally speaking, the cumulative probability can approximately represent the value of its inverse function. Therefore, in the embodiment of the present invention, the cumulative probability is used to determine the target total inventory.

[0064] As mentioned above, in Figure 2 In the probability distribution diagram shown in the figure, the horizontal axis (y) represents the number axis formed by arranging multiple outbound data in order of size, and the vertical axis f(y i ) represents the second distribution probability corresponding to each outbound data. Generally speaking, according to the second distribution probability f(y iThe area enclosed by the fitted curve and the coordinate axis can be approximated as the distribution probability of the total inventory of items. For example, the distribution probability corresponding to y2 is Figure 2 The distribution probability of z1 shaded part and y3 corresponding to y3 is shown as follows: Figure 2 The shaded part of z2 is shown, y i+1 The corresponding distribution probability is Figure 2 The z shown i The shaded area. Figure 2 After the values ​​of the shaded part shown are normalized, the first distribution probability of the corresponding total inventory of items can be obtained. However, since the probability distribution function determined by the kernel density estimation algorithm does not display an expression, the area of ​​the shaded part cannot be directly calculated. In an embodiment of the present invention, in order to improve the calculation accuracy of the first distribution probability of the total inventory of items, the first intervals between multiple outbound data are split. Still taking the historical sales data of the above-mentioned item A as outbound data as an example, each historical sales data is arranged in order of size as 80, 100, 150 and 200. If the first distribution probability of the total inventory of items is calculated directly based on the intervals between each historical sales data, it may be because the interval between each two historical sales data is large and the intervals between each historical sales data are of different sizes. Therefore, if the first distribution probability of the total inventory of items is calculated directly based on the intervals between each historical sales data, this may result in a low accuracy rate of the first distribution probability.

[0065] In order to improve the accuracy of the first distribution probability of the total inventory of items, in an embodiment of the present invention, the first intervals between multiple outbound data are split. For example, the historical sales data 80 and 100 correspond to Figure 2 y1 and y3 in . Then the first interval between every two historical sales data is split, such as splitting the first interval between 80-100, and obtaining y2 after splitting, then the size of the outbound data corresponding to y2 is 90. Similarly, the first interval between other historical sales data can be split to obtain multiple second intervals. For example, the first interval between 100-150 is split to obtain five second intervals of 100-110, 110-120, 120-130, 130-140 and 140-150. It can be understood that since the second interval is obtained by splitting the first interval, the number of second intervals is greater than the number of first intervals. In addition, in this example, for the sake of convenience, the second interval of 10 is taken as an example in this example. In actual applications, in order to ensure accuracy and improve the first distribution probability of the total inventory of items, the value of the second interval is generally less than 10; after taking the outbound data of multiple historical periods, the number of split second intervals is generally more than 100.

[0066] After splitting to obtain multiple second intervals, the f(yi ) as the second distribution probability corresponding to the second interval, and then use the following formula (4) to calculate the product of each second interval and its corresponding second distribution probability, and normalize the calculated product of the second interval and its corresponding second distribution probability according to the following formula (5) to obtain the first distribution probability corresponding to the total inventory of items. For example, the approximate rectangle enclosed by f(y1)y1y2f(y2) is taken as a first distribution probability of the total inventory of items.

[0067] z i =(y i+1 -y i )·f(y i ) (4)

[0068]

[0069] Among them, (y i+1 -y i ) represents the i-th second interval, f(y i ) represents the second distribution probability corresponding to the i-th second interval, Characterizes the first distribution probability of the total inventory of the i-th item.

[0070] As mentioned above, the cumulative probability is generally used to approximate the inverse function value of the kernel density estimation algorithm. Therefore, after obtaining the first distribution probability corresponding to the total inventory of items, the first distribution probabilities corresponding to the different total inventories of items can be accumulated in turn to obtain the cumulative probabilities corresponding to the different total inventories of items; determine the target cumulative probability whose cumulative probability is not less than the probability of meeting the outbound demand (such as CSL) and has the smallest difference with CSL, and determine the target total inventory of items based on the target cumulative probability.

[0071] For the convenience of calculation, the cumulative probability corresponding to the total inventory of each item can be calculated according to the following formula (6), and then the total inventory of items corresponding to the cumulative probability that is greater than the turnover service level CSL within the forecast period and closest to CSL can be determined based on the calculated cumulative probability, and the determined total inventory of items is used as the target total inventory of items.

[0072]

[0073] in, Characterize the first distribution probability of the total inventory of the i-th item, q i Represents the cumulative probability corresponding to the total inventory of the i-th item.

[0074] For example, when the turnover service level CSL in the forecast period is 0.9, there are multiple cumulative probabilities greater than CSL, namely 0.91, 0.92 and 0.93, then 0.91 is selected as the target cumulative probability, and the target total inventory of items is determined according to the number of items provided indicated by the outbound data corresponding to 0.91. In other words, when determining the target total inventory of items, the outbound data corresponding to the last first distribution probability accumulated in the target cumulative probability can be determined first, and then the outbound quantity indicated by the determined outbound data is used as the target total inventory of items. For example, 0.91 is the cumulative probability corresponding to y4, 0.92 is the cumulative probability corresponding to y5, and 0.93 is the cumulative probability corresponding to y6. Then, although 0.91, 0.92 and 0.93 are not less than CSL, 0.91 is the closest to CSL among the three cumulative probabilities, that is, the difference between 0.91 and CSL (0.9) is the smallest, so 0.91 is used as the target cumulative probability. Since 0.91 is the cumulative probability corresponding to y4, the last accumulated first distribution probability corresponding to 0.91 is the product of the second interval represented by y3-y4 and its corresponding second distribution probability. Then the outbound data corresponding to 0.91 is the outbound data corresponding to f(y4), and then the outbound quantity indicated by the outbound data is used as the total inventory of the target item. For example, when the outbound data is historical sales data, the outbound quantity indicated by the outbound data is the outbound quantity of items equal to the sales volume.

[0075] Step S104: Determine the safety stock within the forecast period according to the total stock of the target item and the turnover stock within the forecast period.

[0076] After determining the total inventory of the target item, the difference between the total inventory of the target item and the turnover inventory can be used as the safety stock based on the turnover inventory within the forecast period. The turnover inventory refers to the inventory required for normal turnover within the forecast period. For example, based on the current orders, it can be determined that within the 10 days of the forecast period, 10 items need to be provided every day, so the turnover inventory corresponding to the forecast period is 100. If the total target inventory is 150, then the safety stock is 50.

[0077] Furthermore, after determining the turnover inventory and safety inventory, the items to be replenished can be allocated based on the existing inventory, safety inventory and / or turnover inventory. When allocating the items to be replenished, the replenishment can be further carried out according to the existing replenishment strategy of the warehouse. For example, if the replenishment strategy is 100% replenishment based on safety inventory, then in the above example where the turnover inventory is 100 and the safety inventory is 50, if the current inventory is 120, then the number of items to be replenished is 30. If the replenishment strategy is 110% replenishment based on safety inventory, then the number of items to be replenished is 33.

[0078] According to the above embodiment, Figure 3 As shown, a method for determining safety stock provided by an embodiment of the present invention may include the following steps:

[0079] Step S301: Obtain the outbound data of items in at least one historical period.

[0080] Step S302: According to the outbound data of items in the at least one historical period, a probability distribution graph corresponding to the total inventory of items is determined using a kernel density estimation algorithm.

[0081] Step S303: Determine the total target item inventory that matches the turnover service level within the forecast period according to the probability distribution graph.

[0082] Step S304: Determine the safety stock within the forecast period according to the total stock of the target item and the turnover stock within the forecast period.

[0083] Step S305: Allocate items to be replenished according to the existing inventory and the safety inventory.

[0084] It can be seen from the method for determining safety stock according to the embodiment of the present invention that by obtaining the outbound data of items in at least one historical period, and according to the obtained outbound data, the probability distribution diagram corresponding to the total inventory of items is determined by using the kernel density estimation algorithm, and then the total inventory of target items in the forecast period is determined according to the probability distribution diagram, and then the safety stock in the forecast period is determined according to the determined total inventory of target items and the turnover inventory in the forecast period. Since the kernel density estimation algorithm can generate the corresponding probability distribution according to the actual demand, it is not necessary to pre-assume that the actual demand obeys a known distribution (such as a normal distribution), that is, when the actual demand for items is irregularly distributed, this solution will not assume the irregular distribution as a normal distribution, but use the kernel density estimation algorithm to fit the probability distribution of the total inventory of items according to its actual irregular distribution. Therefore, compared with the prior art that assumes the actual demand for items in all cases as a normal distribution, this solution improves the fitting accuracy of the probability distribution of the actual demand for items, thereby improving the prediction accuracy of the safety stock. Moreover, the determined total inventory of target items matches the probability of meeting the outbound demand within the forecast period, for example, matches the CSL within the forecast period. Then, the safety stock within the forecast period is determined based on the total inventory of target items, so that the determined safety stock can meet the outbound demand (such as CSL) requirements, and then the items to be replenished are allocated according to the safety stock, which can reduce the risk of insufficient inventory of items within the preset period, thereby helping to improve the service level.

[0085] Figure 4 4 is a schematic diagram of main modules of a device for determining safety stock according to an embodiment of the present invention.

[0086] like Figure 4 As shown, a device 400 for determining safety stock according to an embodiment of the present invention includes: a data acquisition module 401, a probability determination module 402 and an inventory determination module 403; wherein,

[0087] The data acquisition module 401 is used to acquire the outbound data of items in at least one historical period;

[0088] The probability determination module 402 is used to determine the probability distribution diagram corresponding to the total inventory of items by using a kernel density estimation algorithm according to the outbound data of items in the at least one historical period;

[0089] The inventory determination module 403 is used to determine the total inventory of target items within the forecast period according to the probability distribution diagram; and determine the safety inventory within the forecast period according to the total inventory of target items and the turnover inventory within the forecast period.

[0090] In one embodiment of the present invention, the inventory determination module 403 is used to determine the total inventory of the target item according to the probability distribution graph and the probability of satisfying the outbound demand within the forecast period.

[0091] In one embodiment of the present invention, the inventory determination module 403 is used to calculate the first distribution probabilities corresponding to different total inventories of items according to the probability distribution diagram; accumulate the first distribution probabilities corresponding to the different total inventories of items in sequence according to the size order of the outbound data in the at least one historical period indicated by the probability distribution diagram to obtain the cumulative probabilities corresponding to the different total inventories of items; determine the target cumulative probability whose cumulative probability is not less than the probability of satisfying the outbound demand and whose difference with the probability of satisfying the outbound demand is the smallest, and determine the target total inventory of items according to the target cumulative probability.

[0092] In one embodiment of the present invention, the inventory determination module 403 is used to split the first intervals between the multiple outbound data of different sizes according to the size order, obtain multiple second intervals and the second distribution probabilities corresponding to the multiple second intervals respectively; normalize the product of each second interval and the second distribution probability corresponding to itself, and use the result of the normalization as the first distribution probability corresponding to the total inventory of the items.

[0093] In one embodiment of the present invention, the inventory determination module 403 is used to determine the outbound data corresponding to the last of the first distribution probability accumulated in the target cumulative probability, and use the outbound quantity indicated by the determined outbound data as the total inventory of the target item.

[0094] In one embodiment of the present invention, the probability determination module 402 is used to determine the bandwidth of the probability distribution graph to be fitted according to the number of the outbound data in the at least one historical period, and to fit the outbound data in the at least one historical period using a Gaussian kernel according to the bandwidth to obtain the probability distribution graph.

[0095] In one embodiment of the present invention, the inventory determination module 403 is used to use the difference between the total inventory of the target item and the turnover inventory as the safety inventory.

[0096] In one embodiment of the present invention, the inventory determination module 403 is further used to allocate items to be replenished according to the existing inventory and the safety inventory.

[0097] According to a device for determining safety stock according to an embodiment of the present invention, it can be seen that by obtaining the outbound data of items in at least one historical period, and according to the obtained outbound data, using the kernel density estimation algorithm to determine the probability distribution diagram corresponding to the total inventory of items, then determining the total inventory of target items in the forecast period according to the probability distribution diagram, and then determining the safety stock in the forecast period according to the determined total inventory of target items and the turnover inventory in the forecast period. Since the kernel density estimation algorithm can generate the corresponding probability distribution according to the actual demand, it is not necessary to pre-assume that the actual demand obeys a known distribution (such as a normal distribution), that is, when the actual demand for items is irregularly distributed, this solution will not assume the irregular distribution as a normal distribution, but use the kernel density estimation algorithm to fit the probability distribution of the total inventory of items according to its actual irregular distribution. Therefore, compared with the prior art that assumes the actual demand for items in all cases as a normal distribution, this solution improves the fitting accuracy of the probability distribution of the actual demand for items, thereby improving the prediction accuracy of the safety stock. Moreover, the determined total inventory of target items matches the probability of meeting the outbound demand within the forecast period, for example, matches the CSL within the forecast period. Then, the safety stock within the forecast period is determined based on the total inventory of target items, so that the determined safety stock can meet the outbound demand (such as CSL) requirements, and then the items to be replenished are allocated according to the safety stock, which can reduce the risk of insufficient inventory of items within the preset period, thereby helping to improve the service level.

[0098] Figure 5 An exemplary system architecture 500 is shown to which the method for determining safety stock or the apparatus for determining safety stock according to the embodiments of the present invention can be applied.

[0099] like Figure 5As shown, the system architecture 600 may include terminal devices 501, 502, 503, a network 504, and a server 505. The network 504 is used to provide a medium for communication links between the terminal devices 501, 502, 503 and the server 505. The network 504 may include various connection types, such as wired, wireless communication links, or optical fiber cables, etc.

[0100] Users can use terminal devices 501, 502, 503 to interact with server 505 through network 504 to receive or send messages, etc.

[0101] The terminal devices 501 , 502 , and 503 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0102] The server 505 may be a server that provides various services, such as a backend management server that provides support for shopping websites browsed by users using the terminal devices 501, 502, and 503. The backend management server may analyze and process the received data such as product information query requests, and feed back the processing results (such as safety inventory information) to the terminal device.

[0103] It should be noted that the method for determining the safety stock provided in the embodiment of the present invention is generally executed by the server 505 , and accordingly, the device for determining the safety stock is generally disposed in the server 505 .

[0104] It should be understood that Figure 5 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to the implementation requirements.

[0105] Reference below Figure 6 , which shows a schematic diagram of the structure of a computer system 600 of a terminal device suitable for implementing an embodiment of the present invention. Figure 6 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0106] like Figure 6 As shown, the computer system 600 includes a central processing unit (CPU) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage part 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the system 600 are also stored. The CPU 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0107] The following components are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed into the storage section 608 as needed.

[0108] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the central processing unit (CPU) 601, the above-mentioned functions defined in the system of the present invention are executed.

[0109] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0110] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0111] The modules involved in the embodiments of the present invention may be implemented in software or hardware. The modules described may also be set in a processor, for example, it may be described as: a processor includes a data acquisition module, a probability determination module and an inventory determination module. The names of these modules do not constitute a limitation on the modules themselves in some cases, for example, the data acquisition module may also be described as a "module for acquiring the outbound data of items in at least one historical period".

[0112] As another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiment; or it may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device includes: obtaining the outbound data of items in at least one historical period; according to the outbound data of items in at least one historical period, using the kernel density estimation algorithm to determine the probability distribution map corresponding to the total inventory of items; according to the probability distribution map, determining the total inventory of target items in the forecast period; according to the total inventory of target items and the turnover inventory in the forecast period, determining the safety inventory in the forecast period.

[0113] According to the technical solution of the embodiment of the present invention, by obtaining the outbound data of the items in at least one historical period, and according to the obtained outbound data, using the kernel density estimation algorithm to determine the probability distribution diagram corresponding to the total inventory of the items, then according to the probability distribution diagram, determine the total inventory of the target items in the forecast period, and then according to the determined total inventory of the target items and the turnover inventory in the forecast period, determine the safety stock in the forecast period. Since the kernel density estimation algorithm can generate the corresponding probability distribution according to the actual demand, it is not necessary to pre-assume that the actual demand obeys a known distribution (such as a normal distribution), that is, when the actual demand for the items is irregularly distributed, this solution will not assume the irregular distribution as a normal distribution, but use the kernel density estimation algorithm to fit the probability distribution of the total inventory of the items according to its actual irregular distribution. Therefore, compared with the prior art that assumes the actual demand for the items in all cases as a normal distribution, this solution improves the fitting accuracy of the probability distribution of the actual demand for the items, thereby improving the prediction accuracy of the safety stock. Moreover, the determined total inventory of target items matches the probability of meeting the outbound demand within the forecast period, for example, matches the CSL within the forecast period. Then, the safety stock within the forecast period is determined based on the total inventory of target items, so that the determined safety stock can meet the outbound demand (such as CSL) requirements, and then the items to be replenished are allocated according to the safety stock, which can reduce the risk of insufficient inventory of items within the preset period, thereby helping to improve the service level.

[0114] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for determining safety stock, characterized in that: include: Obtain the outbound data of items in at least one historical period; Determine a probability distribution graph corresponding to the total inventory of items using a kernel density estimation algorithm based on the outbound data of items in the at least one historical period; Determining the target total inventory of items within the forecast period according to the probability distribution graph and the probability of satisfying the outbound demand within the forecast period includes: calculating first distribution probabilities corresponding to different total inventories of items according to the probability distribution graph; accumulating the first distribution probabilities corresponding to the different total inventories of items in sequence according to the size order of the outbound data within the at least one historical period indicated by the probability distribution graph to obtain cumulative probabilities corresponding to different total inventories of items; determining a target cumulative probability whose cumulative probability is not less than the probability of satisfying the outbound demand and whose difference with the probability of satisfying the outbound demand is the smallest, and determining the target total inventory of items according to the target cumulative probability; The safety stock within the forecast period is determined according to the total stock of the target item and the turnover stock within the forecast period.

2. The method according to claim 1, characterized in that The probability distribution graph further indicates the second distribution probability of the outbound data; and calculating the first distribution probabilities corresponding to different total inventories of items according to the probability distribution graph includes: According to the size order, first intervals between the multiple outbound data of different sizes are respectively split to obtain multiple second intervals and second distribution probabilities respectively corresponding to the multiple second intervals; The product of each second interval and the second distribution probability corresponding to itself is normalized, and the result of the normalization is used as the first distribution probability corresponding to the total inventory of items.

3. The method according to claim 1, characterized in that Determining the total inventory of the target items according to the target cumulative probability includes: The outbound data corresponding to the last first distribution probability accumulated in the target cumulative probability is determined, and the outbound quantity indicated by the determined outbound data is used as the total inventory of the target items.

4. The method according to claim 1, characterized in that: The method of determining the probability distribution graph corresponding to the inventory of items by using a kernel density estimation algorithm according to the outbound data of items in the at least one historical period includes: The bandwidth of the probability distribution diagram to be fitted is determined according to the number of the outbound data in the at least one historical period, and the outbound data in the at least one historical period is fitted with a Gaussian kernel according to the bandwidth to obtain the probability distribution diagram.

5. The method according to claim 1, characterized in that The difference between the total inventory of the target item and the turnover inventory is used as the safety stock.

6. The method according to any one of claims 1 to 5, characterized in that: Also includes: Arrange items to be replenished based on existing inventory and the safety stock.

7. A device for determining safety stock, characterized in that: include: Data acquisition module, probability determination module and inventory determination module; wherein, The data acquisition module is used to acquire the outbound data of items in at least one historical period; The probability determination module is used to determine the probability distribution diagram corresponding to the total inventory of items by using a kernel density estimation algorithm according to the outbound data of items in the at least one historical period; The inventory determination module is used to determine the target total inventory of items within the prediction period according to the probability distribution diagram and the probability of satisfying the outbound demand within the prediction period, including: calculating the first distribution probabilities corresponding to different total inventories of items according to the probability distribution diagram; accumulating the first distribution probabilities corresponding to the different total inventories of items in sequence according to the size order of the outbound data in the at least one historical period indicated by the probability distribution diagram to obtain the cumulative probabilities corresponding to the different total inventories of items; determining the target cumulative probability whose cumulative probability is not less than the probability of satisfying the outbound demand and whose difference with the probability of satisfying the outbound demand is the smallest, and determining the target total inventory of items according to the target cumulative probability; and determining the safety stock within the prediction period according to the target total inventory of items and the turnover inventory within the prediction period.

8. An electronic device for determining safety stock, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 6.

9. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.