CN114372735A

CN114372735A - Method and device for determining buffer stock quantity data in logistics supply chain

Info

Publication number: CN114372735A
Application number: CN202011094097.9A
Authority: CN
Inventors: 吕骥图; 许哲民; 柯俞嘉; 金虹希; 郭雨佳; 王晶; 张潆尹; 王本玉
Original assignee: Shanghai Shunrufenglai Technology Co ltd
Current assignee: Shanghai Shunrufenglai Technology Co ltd
Priority date: 2020-10-14
Filing date: 2020-10-14
Publication date: 2022-04-19

Abstract

The application provides a method and a device for determining buffer stock quantity data in a logistics supply chain, which are used for improving the prediction accuracy of a goods demand quantity prediction result to a certain extent. The method comprises the following steps: acquiring historical goods demand quantity data of a target node, wherein the target node is a node in a logistics supply chain, and the historical goods demand quantity data is used for indicating the demand quantity of the target node for target goods in a historical time period; inputting the historical goods demand quantity data into a goods demand quantity prediction model to obtain goods demand quantity prediction data, wherein the goods demand quantity prediction model is obtained by training an initial neural network model by adopting goods demand quantity data of different time periods, and the goods demand quantity data of different time periods are marked with distribution types of different nodes of different goods in different time periods; and determining the buffer inventory quantity data of the target node for the target goods according to the goods demand quantity prediction data and the current goods inventory quantity data.

Description

Method and device for determining buffer stock quantity data in logistics supply chain

Technical Field

The application relates to the field of logistics, in particular to a method and a device for determining buffer inventory quantity data in a logistics supply chain.

Background

In the field of logistics, with the continuous improvement and expansion of logistics services, a logistics company can provide traditional logistics services for transporting logistics pieces, and can also play an upstream role in a supply chain, and the logistics company can provide corresponding goods to downstream merchants through the logistics services of the logistics company, for example, provide commodities required for retail of the merchants such as logistics outlets, supermarkets, convenience stores and the like.

In the aspect of a supply chain, the inventory of goods can be considered from the aspect of demand, and the goods are convenient to stock. In the processing process, besides the traditional mode which can depend on manual prediction, the demand and the inventory can also be predicted through the analysis of big data, and the historical demand quantity related to goods is taken as time series data to be predicted through a time series model.

In the existing research of related technologies, the inventor finds that the demand prediction result given by the time sequence model is unstable and often does not conform to the actual situation, and the stock deployed according to the goods demand prediction result given by the time sequence model is easy to have two situations, too much stock or insufficient supply, and obviously, the precision of the goods demand prediction result is low.

Disclosure of Invention

The application provides a method and a device for determining buffer inventory quantity data in a logistics supply chain, which are used for improving the prediction precision of a goods demand quantity prediction result to a certain extent and further improving the precision of buffer inventory.

In a first aspect, the present application provides a method for determining buffer inventory quantity data in a logistics supply chain, where the method includes:

acquiring historical goods demand quantity data of a target node, wherein the target node is a node in a logistics supply chain, and the historical goods demand quantity data is used for indicating the demand quantity of the target node for target goods in a historical time period;

inputting historical goods demand quantity data into a goods demand quantity prediction model to predict the demand quantity of a target node for target goods in the current time period and obtain goods demand quantity prediction data, wherein the goods demand quantity prediction model is obtained by training an initial neural network model by adopting goods demand quantity data of different time periods, and the goods demand quantity data of different time periods are marked with distribution types of different goods at different nodes of different time periods;

and determining buffer inventory quantity data of the target node for the target goods according to the goods demand quantity prediction data and the current goods inventory quantity data, wherein the current goods inventory quantity data is used for indicating the inventory quantity of the target node for the target goods in the current time period, and the buffer inventory quantity data is used for indicating the standby inventory quantity which is higher than the inventory quantity corresponding to the current goods inventory quantity data.

With reference to the first aspect of the present application, in a first possible implementation manner of the first aspect of the present application, the obtaining historical cargo demand quantity data of the target node includes:

acquiring a plurality of cargo images of a target node in a historical time period, wherein the image content of the cargo image comprises a shelf layer in which cargos are placed, and the cargo image is obtained by on-site shooting of the target node;

identifying commodities contained in the images in the goods images to obtain commodity data, wherein the commodity data comprises commodity categories and commodity numbers of the identified commodities;

and confirming historical goods demand quantity data according to the goods data.

With reference to the first possible implementation manner of the first aspect of the present application, in a second possible implementation manner of the first aspect of the present application, the goods image identifier has a date, the goods data includes a plurality of sets of sub-goods data labeled by taking the date as a date unit, and the determining the historical goods demand quantity data according to the goods data includes:

extracting target sub-commodity data marked with target days from the commodity data;

according to the target sub-commodity data, determining the current-day inventory quantity data and the current-day sales quantity data of the target node for the target goods;

and determining historical goods demand quantity data according to the inventory quantity data on the day and the sales quantity data on the day.

With reference to the second possible implementation manner of the first aspect of the present application, in a third possible implementation manner of the first aspect of the present application, the determining the historical cargo demand quantity data according to the daily inventory quantity data and the daily sales quantity data includes:

determining the position type of the target node in the node network of the logistics supply chain according to the node identification of the target node, wherein the position type comprises an end node type or a non-end node type;

and determining historical goods demand quantity data according to the position type, the current-day inventory quantity data and the current-day sales quantity data.

With reference to the third possible implementation manner of the first aspect of the present application, in a fourth possible implementation manner of the first aspect of the present application, determining the historical cargo demand quantity data according to the location type, the current-day inventory quantity data, and the current-day sales quantity data includes:

when the position type is the end node type, determining whether the amount of the inventory quantity data on the day is larger than zero;

if the quantity of the goods in the current day is larger than zero, determining the quantity of the goods sold in the current day as the historical goods demand quantity data of the target node on the target day;

if the current day inventory quantity data is equal to zero, extracting a target day with preset characteristics equal to the target day by the target node according to the current day inventory quantity data and the current day sales data, wherein the amount of the current day inventory quantity data on the target day is larger than zero;

and determining the average sales volume data of the historical goods demand volume data of the target day as the historical goods demand volume data of the target node on the target day.

With reference to the third possible implementation manner of the first aspect of the present application, in a fifth possible implementation manner of the first aspect of the present application, the determining the historical cargo demand quantity data according to the location type, the current-day inventory quantity data, and the current-day sales quantity data includes:

when the position type is a non-end node type, determining a downstream node with a downstream identifier, wherein the downstream identifier is used for identifying a downstream node belonging to a target node in a framework in a logistics supply chain;

and counting the order quantity data of the downstream nodes on the target day, and determining the historical goods demand quantity data of the target nodes on the target day according to the order quantity data, the inventory quantity data on the current day and the sales quantity data on the current day.

With reference to the fourth or fifth possible implementation manner of the first aspect of the present application, in a sixth possible implementation manner of the first aspect of the present application, the distribution types include a normal distribution type, a gamma distribution type, a poisson distribution type, and a negative distribution type that respectively satisfy a preset fluctuation type, the preset fluctuation type is obtained by eliminating three fluctuation types, namely, an extremely low frequency fluctuation type, an extremely high frequency fluctuation type, and an extremely small fluctuation type, from a fluctuation type set according to a duty ratio of a zero value and a fluctuation amplitude, and the preset fluctuation type includes a high frequency stable fluctuation type, a low frequency stable fluctuation type, a high frequency fluctuation type, and a low frequency fluctuation type.

In a second aspect, the present application provides an apparatus for determining buffer inventory quantity data in a logistics supply chain, the apparatus comprising:

the receiving and sending unit is used for acquiring historical goods demand quantity data of a target node, wherein the target node is a node in a logistics supply chain, and the historical goods demand quantity data is used for indicating the demand quantity of the target node for target goods in a historical time period;

the processing unit is used for inputting the historical goods demand quantity data into a goods demand quantity prediction model so as to predict the demand quantity of a target node for target goods in the current time period and obtain goods demand quantity prediction data, wherein the goods demand quantity prediction model is obtained by training an initial neural network model by adopting goods demand quantity data of different time periods, and the goods demand quantity data of different time periods are marked with distribution types of different nodes of different goods in different time periods; and determining buffer inventory quantity data of the target node for the target goods according to the goods demand quantity prediction data and the current goods inventory quantity data, wherein the current goods inventory quantity data is used for indicating the inventory quantity of the target node for the target goods in the current time period, and the buffer inventory data is used for indicating the standby inventory quantity which is higher than the inventory quantity corresponding to the current goods inventory quantity data.

With reference to the second aspect of the present application, in a first possible implementation manner of the second aspect of the present application, the obtaining unit is specifically configured to:

the processing unit is specifically used for identifying the commodities contained in the image in the goods image to obtain commodity data, wherein the commodity data comprises the commodity category and the commodity number of the identified commodities; and confirming historical goods demand quantity data according to the goods data.

With reference to the first possible implementation manner of the second aspect of the present application, in a second possible implementation manner of the second aspect of the present application, the goods image identifier has a date, the goods data includes multiple sets of sub-goods data labeled by taking the date as a date unit, and the processing unit is specifically configured to:

With reference to the second possible implementation manner of the second aspect of the present application, in a third possible implementation manner of the second aspect of the present application, the processing unit is specifically configured to:

With reference to the third possible implementation manner of the second aspect of the present application, in a fourth possible implementation manner of the second aspect of the present application, the processing unit is specifically configured to:

With reference to the third possible implementation manner of the second aspect of the present application, in a fifth possible implementation manner of the second aspect of the present application, the processing unit is specifically configured to:

With reference to the fourth or fifth possible implementation manner of the second aspect of the present application, in a sixth possible implementation manner of the second aspect of the present application, the distribution types include a normal distribution type, a gamma distribution type, a poisson distribution type, and a negative distribution type that respectively satisfy a preset fluctuation type, the preset fluctuation type is obtained by eliminating three fluctuation types, namely, an extremely low frequency fluctuation type, an extremely high frequency fluctuation type, and an extremely small fluctuation type, from a fluctuation type set according to a duty ratio of a zero value and a fluctuation amplitude, and the preset fluctuation type includes a high frequency stable fluctuation type, a low frequency stable fluctuation type, a high frequency fluctuation type, and a low frequency fluctuation type.

In a third aspect, the present application further provides a device for determining data of the amount of buffer stock in a logistics supply chain, including a processor and a memory, where the memory stores a computer program, and the processor executes the steps in any one of the methods provided in the first aspect of the present application when calling the computer program in the memory.

In a fourth aspect, the present application further provides a computer-readable storage medium storing a plurality of instructions adapted to be loaded by a processor to perform the steps of any of the methods provided in the first aspect of the present application.

From the above, the present application has the following advantageous effects:

on one hand, the method is characterized in that a cargo demand quantity prediction model is trained, the model is obtained by training an initial neural network model by adopting cargo demand quantity data of different time periods, and the distribution types of different nodes of different cargos in different time periods are marked on the cargo demand quantity data, so that when the trained model inputs historical cargo demand quantity data of a target node and predicts the cargo demand quantity, the distribution types of the cargo demand quantity of the target node in different time periods can be concerned, and the specific distribution type corresponding to the target node is determined, so that the demand quantity of the target node to the target cargo in the current time period can be predicted more accurately based on the distribution types;

secondly, because better and accurate goods demand quantity prediction data is obtained, the method and the device can determine corresponding buffer inventory quantity data based on the goods demand quantity prediction data and by combining the current goods inventory quantity data of the target node, wherein the current goods inventory quantity data is used for indicating the inventory quantity of the target node for the target goods in the current time period, and the buffer inventory quantity data is used for indicating the reserve inventory quantity which is higher than the inventory quantity corresponding to the current goods inventory quantity data, so that when the target node carries out the stock in the current time period, a better and accurate and stable stock result can be obtained and used as a supply chain node in a logistics supply chain, and the target node can work and play a role more stably in the supply chain operation.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

Fig. 1 is a schematic flow chart illustrating a method for determining buffer inventory quantity data in a logistics supply chain according to the present application;

FIG. 2 is a schematic flow chart illustrating the process of acquiring historical cargo demand quantity data according to the present application;

FIG. 3 is a schematic view of another process for determining historical cargo demand quantity data according to the present application;

FIG. 4 is a schematic view of another process for determining historical cargo demand quantity data according to the present application;

FIG. 5 is a schematic view of another process for determining historical cargo demand quantity data according to the present application;

FIG. 6 is a schematic diagram of a scenario of the present application for a wave classification process;

FIG. 7 is a schematic diagram of a structure of a device for determining the amount of buffer stock in a logistics supply chain according to the present application;

fig. 8 is a schematic structural diagram of a device for determining buffer inventory quantity data in a logistics supply chain according to the present application.

Detailed Description

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

In the description that follows, specific embodiments of the present application will be described with reference to steps and symbols executed by one or more computers, unless otherwise indicated. Accordingly, these steps and operations will be referred to, several times, as being performed by a computer, the computer performing operations involving a processing unit of the computer in electronic signals representing data in a structured form. This operation transforms the data or maintains it at locations in the computer's memory system, which may be reconfigured or otherwise altered in a manner well known to those skilled in the art. The data maintains a data structure that is a physical location of the memory that has particular characteristics defined by the data format. However, while the principles of the application have been described in language specific to above, it is not intended to be limited to the specific form set forth herein, and it will be recognized by those of ordinary skill in the art that various of the steps and operations described below may be implemented in hardware.

The principles of the present application may be employed in numerous other general-purpose or special-purpose computing, communication environments or configurations. Examples of well known computing systems, environments, and configurations that may be suitable for use with the application include, but are not limited to, hand-held telephones, personal computers, servers, multiprocessor systems, microcomputer-based systems, mainframe-based computers, and distributed computing environments that include any of the above systems or devices.

The terms "first", "second", and "third", etc. in this application are used to distinguish between different objects and not to describe a particular order. Furthermore, the terms "include" and "have," as well as any variations thereof, are intended to cover non-exclusive inclusions.

First, before the present application is introduced, the relevant contents of the present application with respect to the application background will be described.

The method and the device for determining the buffer stock quantity data in the logistics supply chain and the computer-readable storage medium can be applied to determining equipment of the buffer stock quantity data, and are used for improving the prediction accuracy of the goods demand quantity prediction result to a certain extent so as to improve the accuracy of the buffer stock.

In the present application, the device for determining the buffer inventory quantity data in the logistics supply chain may be understood as a server device, a physical host, or a User Equipment (UE) and other hardware devices with data processing capability, where the UE may specifically be a terminal device such as a smart phone, a tablet computer, a notebook computer, a palm computer, a desktop computer, or a Personal Digital Assistant (PDA).

The logistics service can be specifically an express delivery service, and the logistics supply chain is a supply chain realized by a logistics company through the logistics service, and the logistics supply chain is composed of different supply chain nodes, such as a logistics node of the logistics company for transporting goods in the logistics supply chain, a third party node for transporting goods in the logistics supply chain, which is added to the logistics supply chain, a logistics node of the logistics company for selling goods in the logistics supply chain, and a third party node for selling goods in the logistics supply chain.

Next, a method for determining the buffer stock quantity data in the logistics supply chain provided by the present application is described.

First, referring to fig. 1, fig. 1 is a schematic flow chart illustrating a method for determining buffer inventory quantity data in a logistics supply chain according to the present application, where the method for determining buffer inventory quantity data in a logistics supply chain specifically includes:

step S101, acquiring historical goods demand quantity data of a target node, wherein the target node is a node in a logistics supply chain, and the historical goods demand quantity data is used for indicating the demand quantity of the target node for target goods in a historical time period;

step S102, inputting historical goods demand quantity data into a goods demand quantity prediction model to predict the demand quantity of a target node for target goods in the current time period and obtain goods demand quantity prediction data, wherein the goods demand quantity prediction model is obtained by training an initial neural network model by adopting goods demand quantity data of different time periods, and the goods demand quantity data of different time periods are marked with distribution types of different nodes of different goods in different time periods;

step S103, according to the forecast data of the quantity of the required goods and the inventory quantity data of the current goods, determining the buffer inventory quantity data of the target node for the target goods, wherein the inventory quantity data of the current goods is used for indicating the inventory quantity of the target node for the target goods in the current time period, and the buffer inventory quantity data is a standby inventory quantity which is higher than the inventory quantity corresponding to the inventory quantity data of the current goods.

As can be seen from the above embodiment shown in fig. 1, in one aspect, the present application trains a cargo demand quantity prediction model, where the model is obtained by training an initial neural network model using cargo demand quantity data of different time periods, and the cargo demand quantity data is marked with distribution types of different nodes of different cargos in different time periods, so that when inputting historical cargo demand quantity data of a target node to perform cargo demand quantity prediction, the trained model can focus on the distribution types of the cargo demand quantity of the target node in different time periods, and determine a specific distribution type corresponding to the target node, so as to better and accurately predict the demand quantity of the target node for the target cargo in a current time period based on the distribution types;

The following proceeds to a detailed description of the various steps of the embodiment shown in fig. 1:

in the application, the forecast of the quantity of the goods required by the target node may be triggered by a worker on a determining device for buffering inventory quantity data in the logistics supply chain, or may be triggered by a device such as UE accessing the determining device for buffering inventory quantity data in the logistics supply chain and triggering, or may be automatically triggered by the determining device for buffering inventory quantity data in the logistics supply chain according to a preset policy, and the specific triggering condition may be adjusted according to actual needs, which is not limited herein.

For example, a worker of an a express company may access a server on a smartphone through an X applet on a B Application (APP), and initiate a demand quantity prediction request for a target node, and the server triggers demand quantity prediction for the target node according to the request; for another example, the server automatically predicts the quantity of the required goods of each node in the logistics supply chain every day at regular time according to the nodes identified in the node list of the logistics supply chain.

Historical cargo demand quantity data for indicating the quantity of demand of the corresponding node for the corresponding cargo in a historical time period, wherein the historical time period and the corresponding cargo can be determined according to actual needs, for example, the historical time period can be a time period of time span of past year, past quarter, past month, past week and the like, and for example, the cargo can be the cargo which is all involved in the node, the cargo forwarded by the node, the commodity sold by the node in the season and the like.

The acquisition of the historical cargo demand quantity data can be mainly obtained through the following modes:

first, configuration by staff

It can be understood that the historical cargo quantity data can be manually configured by related staff, and specifically, the staff can manually configure the required historical cargo quantity data by aiming at the determination method of the buffer inventory quantity data in the logistics supply chain; or, the staff may also configure a related interface on the device that needs to store the historical quantity data of the goods required by the present application, and when the determining device for buffering the inventory quantity data in the logistics supply chain triggers the determining method for buffering the inventory quantity data in the logistics supply chain, the historical quantity data of the goods required by the staff configured by the staff may be retrieved from the devices through the interface.

Second, it is processed by the determining device of the buffer stock quantity data

The goods requirements can be obtained by a paper questionnaire survey issued or an on-line questionnaire survey pushed to the UE, and the staff of the node or the customer of the node fills the goods requirements; alternatively, related supply chain order data, such as related incoming record data, outgoing record data, selling record data, production record data, etc., may be called to determine the quantity of the goods required according to the actual order data of the goods.

In an exemplary implementation, referring to a flowchart of the present application for acquiring the historical quantity data of goods required shown in fig. 2, the acquiring process of the historical quantity data of goods required in the present application may include the following steps S201 to S203:

step S201, acquiring a plurality of goods images of a target node in a historical time period, wherein the image content of the goods images comprises a goods shelf layer in which goods are placed, and the goods images are obtained by on-site shooting of the target node;

in the application, the acquisition of the historical goods demand quantity data can be realized by combining an image recognition mode.

On the site of the node, a shelf layer on the site of the node can be shot by deploying a fixed camera or by holding a shooting device, so that the condition of monitoring the goods entering and exiting the site of the node is achieved, and due to the adoption of an image recognition mode, richer and more direct data can be left in the content, and a better restoration effect is achieved on the site of the node.

Step S202, identifying commodities contained in the image in the goods image to obtain commodity data, wherein the commodity data comprises commodity types and commodity numbers of the identified commodities;

in practical applications, the image features of the product corresponding to the product can be identified by means of an Artificial Intelligence (AI) technique, so as to realize the identification of the product.

The method comprises the steps of taking commodity images, commodity shelf images and other images containing commodities as a training set of a model, marking commodities contained in image contents of the images in the training set for the images in the training set, sequentially inputting the images in the training set into an initial neural network model after the training set is obtained, conducting forward propagation, calculating a loss function according to a commodity recognition result output by the neural network model, conducting backward propagation according to the loss function, adjusting parameters of the neural network model, achieving the purpose of training the neural network model through repeated forward and backward propagation, and when requirements of training times, commodity recognition accuracy and the like are met, the trained neural network model is a commodity recognition model which can be put into practical application and used for recognizing the commodities contained in the images from the commodity images.

The identification result of the goods, that is, the goods data, may include the goods category of the goods identified from the goods image and the corresponding goods number.

Different commodities can be distinguished according to preset commodity identifications, for example, the commodity identification corresponding to the commodity A is 1, the commodity number of the commodity A is represented by 1 × 5, and the commodity identification corresponding to the commodity C is 6.

Further, the commodity data can be presented in a grid graph manner. It can be understood that the commodity is usually regularly placed on the shelf, one-to-one grids can be generated according to the image position and the occupied space of each identified commodity, and each grid is marked with a corresponding commodity identifier, so that commodity data with richer space data can be obtained, and the use of related data processing is facilitated.

Step S203, the historical goods demand quantity data is confirmed according to the goods data.

After the identified commodity data is obtained, the commodity data can be confirmed as the commodity required quantity data in the application, the commodity required quantity is confirmed through image identification according to the actual order data of the commodity, the actual order data is confirmed through the image monitoring mode, rich image data support can be provided, in addition, the in-out situation of the actual commodity on the site of the target node can be accurately restored, the commodity data serves as historical commodity required quantity data, and the actual required quantity situation can be completely reflected when the situation that supply is not in short supply is not generated.

Of course, in consideration of the situation of supply and demand shortage which may exist in practical application, the commodity data can be obtained and then appropriately amplified by combining the empirical adjustment coefficient, so that the historical cargo demand quantity data can include the quantity of the cargo demand which is not met when the situation of supply and demand shortage exists.

Taking the goods image as the actual order data as an example, the goods image may be marked with a date, and the processed goods data may include a plurality of sets of sub-goods data marked in units of dates.

In another exemplary implementation manner, the determining process of historical required goods quantity data after obtaining the commodity data from the goods image through image recognition, as shown in fig. 3, another flow chart for determining historical required goods quantity data according to the present application may specifically include:

step S301, extracting target sub-commodity data marked with target days from the commodity data;

firstly, in the commodity data identified in the front, the preset field carries the corresponding date, and the sub-commodity data of a specific day can be screened from the commodity data according to the date identified by the preset field.

The specific day, or the target day, may be adjusted according to actual needs, for example, individual dates may be manually specified, or by default, each day may be selected as the target day one by one, or individual dates may be selected according to other preset selection strategies, which is not limited herein.

Step S302, according to the target sub-commodity data, determining the current-day inventory quantity data and the current-day sales quantity data of the target node for the target goods;

in the commodity data, the current-day inventory and the current-day sales volume corresponding to the current day can be carried in the preset field, so that the current-day inventory quantity data and the current-day sales volume data corresponding to the sub-commodity data of the specific day can also be extracted.

Step S303, determining historical goods demand quantity data according to the current-day inventory quantity data and the current-day sales quantity data.

After the current inventory and the current sales are obtained, the historical goods demand quantity data can be obtained by processing according to the current inventory and the current sales.

In practical application, the specific processing strategy of the historical cargo demand quantity data can be combined with the network position of the node in the logistics supply chain to perform corresponding configuration.

For example, in yet another exemplary implementation, determining historical quantity of demand data based on the current day inventory and the current day sales includes:

It is understood that each node in the logistics supply chain can be configured with a corresponding node identifier to identify a location type, wherein the nodes can be preferably selected to be divided into two types, namely an end node and a non-end node, the end node is a node pointing to the sale of goods by an end consumer, and the non-end node is an important undertaker of the channel function of the goods, the nodes in the logistics supply chain are subjected to two classification processing according to the two types of the node locations, and the two goods demand modes of 'stocking-selling' and 'stocking-storing-distributing' are distinguished to perform the determination processing of the corresponding goods demand quantity data.

For the determination process of the historical cargo quantity data of the end node, for example, referring to another flow chart shown in fig. 4 for determining the historical cargo quantity data of the present application, the determination process may include:

step S401, when the position type is the end node type, determining whether the amount of the inventory quantity data on the day is larger than zero;

first, it may be determined whether the end node's current day inventory is greater than zero.

It should be understood that, in the present application, the amount of the inventory of the day, or the quantity data of the inventory of the day, may refer to the final inventory of the day, or may refer to the inventory within a specified time period of the day.

Step S402, if the quantity of sales data on the current day is larger than zero, determining the quantity of sales data on the current day as historical goods demand quantity data of the target node on the target day;

the supply chain end node is a node directly facing the client and the final consumer, and when the end node does not have a backorder and does not have an order deferral mechanism, the sales amount on the day can be used as the required quantity of goods on the day.

Step S403, if the current date is equal to zero, extracting a target date with preset characteristics equal to the target date from the target node according to the current date inventory quantity data and the current date sales data, wherein the amount of the current date inventory quantity data of the target date is larger than zero;

if the sales volume is smaller than the demand, the method and the system can perform customization scheme of the supply chain solution in consideration of the lost sales opportunity.

In the application, for the situation that the goods demand is not met, the real goods demand quantity according to the goods demand quantity can be reduced, and the reduction of the real goods demand quantity of the target day can be assisted by combining the standard days which have the same preset characteristics with the target day and have no sales quantity smaller than the demand.

The equivalent preset characteristics are used for indicating common characteristics of the target day and other dates in the aspect of the cargo demand, such as characteristics of similar time points, similar cargo demand quantity trends and the like on the basis of the inventory quantity data of the day and the sales quantity data of the day.

Step S404, determining the average sales volume data of the historical goods demand volume data of the target day as the historical goods demand volume data of the target node on the target day.

After determining the calibration days having the same preset specialization as the target day, the average sales data of the calibration days can be combined as the historical cargo demand quantity data of the target day.

For example, for any day d, if the inventory is 0, the stock shortage may occur, the sales amount sales is less than or equal to the demand, and the demand reduction is performed:

a. searching the last sales volume of the date d, and recording the occurrence time t of the sales volume;

b. searching dates without out-of-stock within a certain time period (which can be adjusted to be 1 month, 3 months, half year and the like according to the industry) before the date d, wherein the set of the dates is S;

c. selecting dates of the same type as the date d from the non-shortage dates S (the same type division rule can be adjusted according to the industry characteristics, such as the same type is weekday/Monday), and obtaining a non-shortage date set S-of the same type as the date d;

d. calculating the average value m of sales quantity occurring after the time t in the non-shortage date set S-;

e. the demand of date d is sales + m.

On the other hand, for the determination process of the historical cargo demand quantity data of the non-end node, for example, referring to another flow chart shown in fig. 5 for determining the historical cargo demand quantity data of the present application, the process may include:

step S501, when the position type is a non-terminal node type, determining a downstream node with a downstream identifier, wherein the downstream identifier is used for identifying a downstream node belonging to a target node in a framework in a logistics supply chain;

it will be appreciated that non-end nodes are not directly facing customers, and their demand is derived from orders from downstream nodes. The order quantity of the downstream node usually considers the fluctuation of future sales volume, and makes more orders or fewer orders, so that the order quantity is not equal to the sales volume, but the non-terminal node does not have real information of the downstream sales volume. In the application, a supply chain solution is established for the real demand quantity of the non-end node, and when the target node is the non-end node, the real demand quantity of goods of the target node can be determined in an auxiliary mode by means of nodes located at the downstream of the target node.

In the application, corresponding upstream and downstream identifications can be configured for each node or related nodes with upstream and downstream relations in the logistics supply chain, and the relevant nodes with upstream and downstream relations can be identified by the identifications, so that the downstream node corresponding to the target node can be determined according to the upstream and downstream identifications.

Step S502, counting the order quantity data of the downstream nodes on the target day, and determining the historical goods demand quantity data of the target nodes on the target day according to the order quantity data, the inventory quantity data on the day and the sales quantity data on the day.

After determining the downstream nodes corresponding to the target node, historical cargo demand quantity data of the target node can be determined by combining the historical cargo demand quantity data of the downstream nodes.

The following are exemplary:

1) dividing all nodes into end node sets S according to network relation and whether to directly face clients_EAnd a set of non-end nodes S_P；

2)S_EThe node in (1) is processed by restoring the real required quantity of the end node as described above, and the obtained result is used as S_EThe real required number of middle nodes, all S_ENode joining restored node set S_FPerforming the following steps;

3) get S_POf a node S_P(i) The set of downstream nodes is Sc_p＝iWhen Sc is used_p＝iAll already exist in S_FWhen it is stated S_P(i) Has completed the restoration, can start to pair S_P(i) Reduction, the flow is as follows:

a. to S_P(i) Set of downstream nodes Sc_p＝iEach of which determines their respective direction S_P(i) Order cycle and order time point;

b. summing the demand in the ordering period of the downstream node, summarizing the sum to the ordering time point, and using the sum as the downstream node pair S_P(i) The amount of demand and the time of occurrence.

c. Will S_P(i) Move out of S_PAdding S_FIn (1).

Repeat 3) until S_FContaining all nodes.

After the order quantity data of the target node is determined, the inventory quantity data of the current day and the sales quantity data of the current day can be compared and corrected so as to restore the actual cargo demand quantity of the target node serving as a non-terminal node.

In the application, the forecasting process of the cargo demand quantity data is realized by a cargo demand quantity forecasting model, the associated historical quantity of cargo demand data may be configured and the corresponding quantity of cargo demand data for the future time period noted, and, in addition, for the historical cargo demand quantity data, in the application, the distribution types of different cargos at different time periods and different nodes are also marked, so that the analysis of the cargo demand quantity in combination with the distribution types in the application can be realized, and as can be understood, according to different data characteristics and distribution characteristics, different distribution types of the demand quantity conditions of the goods are distinguished, and training the model through the historical cargo demand quantity data, so that in the training process of the model, the model can focus on the distribution types of the cargo demand quantities of the nodes in different time periods, the model can accurately fit the development trend of the cargo demand quantity situation by combining the distribution types.

In the training process, configured and labeled historical goods demand quantity data is used as a training set of the model, an initial neural network model is sequentially input, forward propagation is carried out, a loss function is calculated according to goods demand quantity prediction results output by the neural network model, backward propagation is carried out according to the loss function, parameters of the neural network model are adjusted, the purpose of training the neural network model is achieved through multiple times of forward and backward propagation, when requirements such as training times, goods demand quantity prediction accuracy and the like are met, the trained neural network model is a goods demand quantity prediction model, and the model can be used for being put into practical application and used for predicting goods demand quantity of related time periods according to the input historical goods demand quantity data.

As another exemplary implementation manner, in the present application, the distribution types of the cargo demand development trend may include a normal distribution type, a gamma distribution type, a poisson distribution type, and a negative binomial distribution type, which respectively satisfy the preset fluctuation types.

The preset fluctuation type is obtained by removing three fluctuation types, namely an extreme low-frequency fluctuation type, an extreme high-frequency fluctuation type and an extreme small fluctuation type from a fluctuation type set according to the ratio of zero values and the fluctuation amplitude, and the preset fluctuation type comprises a high-frequency stable fluctuation type, a low-frequency stable fluctuation type, a high-frequency fluctuation type and a low-frequency fluctuation type.

For example, in the present application, demand characteristic parameters, which may represent fluctuation and occurrence frequency of the development trend of the quantity of the cargo demand, may be extracted from the data of the quantity of the cargo demand. For example, the demand characteristic parameters may include:

n: the number of non-zero values, i.e., the number of times a non-zero demand occurs;

p: average interval, number of data points/number of non-zero data points;

m: the mean of non-zero value requirements;

cv: the coefficient of variation, standard deviation/mean, is divided into coefficient of variation cv including non-zero values and coefficient of variation cv _ nz not including non-zero values, cv is used for judging extreme fluctuation, and cv _ nz is used for judging other cases.

Configuring a corresponding classification threshold according to the obtained demand characteristic parameters, and classifying the fluctuation types:

extreme low frequency ripple type: in the analysis time period, the number of the non-zero values of the data is extremely small, and is not more than 3, so that the statistical significance on the calculation of the demand parameters is not achieved;

extreme small fluctuation type: the non-zero requirements are small, and the mean value is less than or equal to 1;

extreme fluctuation type: the data fluctuation is severe, and the coefficient of variation cv is more than or equal to 5;

high-frequency stabilization type: zero-value data occupies a small percentage, and the data is stable;

low-frequency stabilization type: zero values and zero values account for more, and data is stable;

high-frequency fluctuation type: zero-value data account for less data, and data fluctuation is large;

low frequency fluctuation type: the zero-value data accounts for a large proportion, and the data fluctuation is large.

For example, referring to a scene diagram of the present invention shown in fig. 6, a required number n is 3, a non-zero mean value m is 1, a coefficient of variation cv is 5, and an average interval p is 1.32 as a specific classification threshold, so as to classify the fluctuation types, where cv is used for determining the extreme fluctuation type, and cv _ nz is used in other cases.

After the fluctuation types are determined, three fluctuation types of an extreme low-frequency fluctuation type, an extreme high-frequency fluctuation type and an extreme small fluctuation type can be eliminated, and the three kinds of cargo demand quantity data with small fluctuation characteristics are eliminated for subsequent distribution type screening.

In the subsequent determination process of the classification type, the optimal fitting distribution of the cargo demand quantity data and the distribution parameters thereof can be determined by means of a maximum likelihood estimation method and the like, and in the application, the main distribution type comprises normal distribution N (mu, sigma)²) Four distribution types, gamma distribution G (α, λ), poisson distribution P (λ), and negative binomial distribution NB (r, P).

After the forecast data of the quantity of the required goods is obtained through the forecast of the quantity of the required goods forecast model, the corresponding buffer stock quantity data can be determined according to the forecast data of the quantity of the required goods and the current stock quantity data of the target node, and the buffer stock quantity data is used for indicating the spare stock quantity of the stock quantity which is higher than the stock quantity corresponding to the current stock quantity data, so that in practical application, the condition of increasing the quantity of the required goods occurs in the case of increasing the quantity of the required goods by newly-added customers and the like meeting the basic goods demand, and the condition of supplying and short of supply is avoided.

After the quantity data of the current goods inventory and the quantity prediction data of the goods demand are obtained, the difference value of the corresponding goods quantity can be determined, if the difference value is positive, the current goods inventory is sufficient, the prediction result of the quantity of the goods demand can be met, and if the difference value is negative, the goods inventory is insufficient, and the goods need to be supplemented.

For example, an amplification factor may be configured to appropriately amplify an amount corresponding to the demanded quantity prediction data of the goods, for example, the amplification factor k is equal to 1.1 or k is equal to 1.2, the amount of the current stock quantity data of the goods is subtracted from the amount, and an obtained amount difference value D may be used as an amount corresponding to the buffer stock quantity data, and if the amount difference value D is positive, replenishment processing may be performed based on the value difference value D; if negative, the goods may be diverted to other logistics nodes without processing or with other processing, such as a diversion process in the supply chain network.

Or, a buffer amount may be configured, where the buffer amount is used to indicate a cargo amount higher than an amount corresponding to the cargo demand amount prediction data, and after a difference between the amount corresponding to the cargo demand amount prediction data and an amount corresponding to the current stock data of the cargo is determined, the difference is added to the buffer amount, and the obtained sum may be used as an amount corresponding to the buffer stock amount data.

Of course, the corresponding buffer inventory quantity data may be determined based on the forecast data of the quantity of the demand for the goods and the current inventory quantity data of the goods in different ways, and the goods may be correspondingly inventory-processed according to the buffer inventory quantity data, which may be adjusted according to actual needs, and is not limited herein.

In order to better implement the method for determining the buffer stock quantity data in the logistics supply chain provided by the application, the application also provides a device for determining the buffer stock quantity data in the logistics supply chain.

Referring to fig. 7, fig. 7 is a schematic structural diagram of a device for determining buffer inventory quantity data in a logistics supply chain according to the present application, in which the device 700 for determining buffer inventory quantity data in a logistics supply chain specifically includes the following structure:

the receiving and sending unit 701 is configured to obtain historical cargo demand quantity data of a target node, where the target node is a node in a logistics supply chain, and the historical cargo demand quantity data is used to indicate a demand quantity of the target node for a target cargo in a historical time period;

the processing unit 702 is configured to input historical cargo demand quantity data into a cargo demand quantity prediction model, so as to predict a demand quantity of a target node for a target cargo in a current time period, and obtain cargo demand quantity prediction data, where the cargo demand quantity prediction model is obtained by training an initial neural network model by using cargo demand quantity data of different time periods, and the cargo demand quantity data of different time periods are labeled with distribution types of different cargo at different nodes of different time periods; and determining buffer inventory quantity data of the target node for the target goods according to the goods demand quantity prediction data and the current goods inventory quantity data, wherein the current goods inventory quantity data is used for indicating the inventory quantity of the target node for the target goods in the current time period, and the buffer inventory quantity data is used for indicating the standby inventory quantity which is higher than the inventory quantity corresponding to the current goods inventory quantity data.

In an exemplary implementation manner, the obtaining unit 701 is specifically configured to:

the processing unit 702 is specifically configured to:

identifying commodities contained in the images in the goods images to obtain commodity data, wherein the commodity data comprises commodity categories and commodity numbers of the identified commodities; and confirming historical goods demand quantity data according to the goods data.

In another exemplary implementation manner, the goods image is identified with a date, the goods data includes multiple sets of sub-goods data labeled by taking the date as a date unit, and the processing unit 702 is specifically configured to:

In another exemplary implementation manner, the processing unit 702 is specifically configured to:

With reference to the third possible implementation manner of the second aspect of the present application, in a fourth possible implementation manner of the second aspect of the present application, the processing unit 702 is specifically configured to:

and counting the order quantity data of the downstream nodes on the target day, and determining the historical goods quantity demand data of the target nodes on the target day according to the order quantity data, the inventory quantity data on the current day and the sales data on the current day.

In yet another exemplary implementation manner, the distribution types include a normal distribution type, a gamma distribution type, a poisson distribution type, and a negative binomial distribution type that respectively satisfy preset fluctuation types, the preset fluctuation types are obtained by eliminating three fluctuation types, namely an extremely low-frequency fluctuation type, an extremely high-frequency fluctuation type, and an extremely small fluctuation type, from a fluctuation type set according to a duty ratio of a zero value and a fluctuation amplitude, and the preset fluctuation types include a high-frequency stable fluctuation type, a low-frequency stable fluctuation type, a high-frequency fluctuation type, and a low-frequency fluctuation type.

Referring to fig. 8, fig. 8 shows a schematic structural diagram of a device for determining buffer stock quantity data in a logistics supply chain according to the present application, specifically, the device for determining buffer stock quantity data in a logistics supply chain according to the present application includes a processor 801, a memory 802, and an input/output device 803, where the processor 801 is configured to implement, when executing a computer program stored in the memory 802, the steps of the method for determining buffer stock quantity data in a logistics supply chain according to any embodiment corresponding to fig. 1 to fig. 6; alternatively, when the processor 801 is configured to execute the computer program stored in the memory 802, the functions of the units in the embodiment corresponding to fig. 7 are implemented, for example, the hardware structure corresponding to the transceiver unit 701 in fig. 7 is the input/output device 803, the hardware structure corresponding to the processing unit 702 is the processor 801, and the memory 802 is configured to store the computer program required by the processor 801 to execute the method for determining the buffer inventory amount data in the logistics supply chain in any embodiment corresponding to fig. 1 to fig. 6.

Illustratively, a computer program may be partitioned into one or more modules/units, which are stored in the memory 802 and executed by the processor 801 to accomplish the present application. One or more modules/units may be a series of computer program instruction segments capable of performing certain functions, the instruction segments being used to describe the execution of a computer program in a computer device.

The determining device for the buffer inventory data in the logistics supply chain may include, but is not limited to, a processor 801, a memory 802, and an input/output device 803. Those skilled in the art will understand that the illustration is only an example of the determining device of the buffering inventory quantity data in the logistics supply chain, and does not constitute a limitation to the determining device of the buffering inventory quantity data in the logistics supply chain, and may include more or less components than those shown, or combine some components, or different components, for example, the determining device of the buffering inventory quantity data in the logistics supply chain may also be a network access device, a bus, etc., and the processor 801, the memory 802, the input and output device 803, and the network access device, etc. are connected through the bus.

The Processor 801 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. The general purpose processor may be a microprocessor or the processor may be any conventional processor or the like, the processor being a control center of the equipment for determining the buffer inventory quantity data in the logistics supply chain, and various interfaces and lines are used to connect the various parts of the whole equipment.

The memory 802 may be used to store computer programs and/or modules, and the processor 801 may implement various functions of the computer device by running or executing the computer programs and/or modules stored in the memory 802 and invoking data stored in the memory 802. The memory 802 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data (such as audio data, video data, etc.) created from the use of the determination device of the data of the buffer stock amount in the logistics supply chain, and the like. In addition, the memory may include high speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), at least one magnetic disk storage device, a Flash memory device, or other volatile solid state storage device.

The processor 801, when executing the computer program stored in the memory 802, may specifically implement the following functions:

It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the above-described specific working processes of the determining apparatus and the device for determining the buffer stock quantity data in the logistics supply chain and the corresponding units thereof may refer to the description of the determining method for the buffer stock quantity data in the logistics supply chain in any embodiment corresponding to fig. 1 to fig. 6, and are not repeated herein.

It will be understood by those skilled in the art that all or part of the steps of the methods of the above embodiments may be performed by instructions or by associated hardware controlled by the instructions, which may be stored in a computer readable storage medium and loaded and executed by a processor.

For this reason, the present application provides a computer-readable storage medium, in which a plurality of instructions are stored, where the instructions can be loaded by a processor to execute the steps in the method for determining the buffer inventory quantity data in the logistics supply chain according to any embodiment of fig. 1 to 6, and specific operations can refer to the description of the method for determining the buffer inventory quantity data in the logistics supply chain according to any embodiment of fig. 1 to 6, which is not repeated herein.

Wherein the computer-readable storage medium may include: read Only Memory (ROM), Random Access Memory (RAM), magnetic or optical disks, and the like.

Since the instructions stored in the computer-readable storage medium can execute the steps in the method for determining the data of the buffer stock quantity in the logistics supply chain according to any embodiment of fig. 1 to 6, the beneficial effects that can be achieved by the method for determining the data of the buffer stock quantity in the logistics supply chain according to any embodiment of fig. 1 to 6 can be achieved, for details, see the foregoing description, and are not repeated herein.

The method, the apparatus, the device and the computer-readable storage medium for determining the buffer inventory quantity data in the logistics supply chain provided by the present application are described in detail above, and a specific example is applied in the present application to explain the principle and the implementation of the present application, and the description of the above example is only used to help understanding the method and the core idea of the present application; meanwhile, for those skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.

Claims

1. A method for determining buffer inventory quantity data in a logistics supply chain, the method comprising:

inputting the historical goods demand quantity data into a goods demand quantity prediction model to predict the demand quantity of the target node for the target goods in the current time period and obtain goods demand quantity prediction data, wherein the goods demand quantity prediction model is obtained by training an initial neural network model by adopting goods demand quantity data of different time periods, and the goods demand quantity data of different time periods are marked with distribution types of different goods at different time periods and different nodes;

and determining buffer inventory quantity data of the target node for the target goods according to the goods demand quantity prediction data and current goods inventory quantity data, wherein the current goods inventory quantity data is used for indicating the inventory quantity of the target node for the target goods in the current time period, and the buffer inventory quantity data is used for indicating the standby inventory quantity which is higher than the inventory quantity corresponding to the current goods inventory quantity data.

2. The method of claim 1, wherein the obtaining historical freight demand quantity data for the target node comprises:

acquiring a plurality of cargo images of the target node in the historical time period, wherein the image content of the cargo image comprises a shelf layer in which cargos are placed, and the cargo image is obtained by on-site shooting of the target node;

and confirming the historical goods demand quantity data according to the commodity data.

3. The method of claim 2, wherein the item image identifies a date, the item data includes a plurality of sets of sub-item data labeled in units of dates, and the confirming the historical item demand quantity data from the item data includes:

according to the target sub-commodity data, determining the current-day inventory quantity data and the current-day sales quantity data of the target node for the target commodity;

and determining the historical goods demand quantity data according to the inventory quantity data on the day and the sales quantity data on the day.

4. The method of claim 3, wherein said determining said historical quantity of needed goods data based on said quantity of inventory data on the day and said sales data on the day comprises:

and determining the historical goods demand quantity data according to the position type, the current-day inventory quantity data and the current-day sales quantity data.

5. The method of claim 4, wherein said determining the historical quantity of demand data based on the location type, the quantity of inventory data on the day, and the sales data on the day comprises:

if the current daily sales data are larger than zero, determining the current daily sales data as historical goods demand quantity data of the target node on the target day;

if the current day inventory quantity data is equal to zero, extracting a target day with the same preset characteristics of the target node and the target day according to the current day inventory quantity data and the current day sales data, wherein the amount of the current day inventory quantity data of the target day is larger than zero;

6. The method of claim 4, wherein said determining the historical quantity of demand data based on the location type, the quantity of inventory data on the day, and the sales data on the day comprises:

when the location type is a non-end node type, determining a downstream node with a downstream identifier, wherein the downstream identifier is used for identifying a downstream node belonging to the target node in a framework in the logistics supply chain;

and counting the order quantity data of the downstream node on the target day, and determining the historical goods demand quantity data of the target node on the target day according to the order quantity data, the inventory quantity data on the current day and the sales quantity data on the current day.

7. The method according to claim 5 or 6, wherein the distribution types include a normal distribution type, a gamma distribution type, a Poisson distribution type and a negative binomial distribution type which respectively satisfy preset fluctuation types, the preset fluctuation types are obtained by eliminating three fluctuation types of an extreme low-frequency fluctuation type, an extreme high-frequency fluctuation type and an extreme small fluctuation type according to a proportion of zero values and a fluctuation amplitude in a fluctuation type set, and the preset fluctuation types include a high-frequency stable fluctuation type, a low-frequency stable fluctuation type, a high-frequency fluctuation type and a low-frequency fluctuation type.

8. An apparatus for determining buffer stock quantity data in a logistics supply chain, the apparatus comprising:

the processing unit is used for inputting the historical cargo demand quantity data into a cargo demand quantity prediction model so as to predict the demand quantity of the target node for the target cargo in the current time period and obtain cargo demand quantity prediction data, wherein the cargo demand quantity prediction model is obtained by training an initial neural network model by adopting cargo demand data of different time periods, and the cargo demand data of different time periods are marked with distribution types of different cargo in different time periods and different nodes; and determining buffer inventory quantity data of the target node for the target goods according to the goods demand quantity prediction data and current goods inventory quantity data, wherein the current goods inventory quantity data is used for indicating the inventory quantity of the target node for the target goods in the current time period, and the buffer inventory quantity data is used for indicating the standby inventory quantity which is higher than the inventory quantity corresponding to the current goods inventory data.

9. A device for determining buffer stock quantity data in a logistics supply chain, characterized by comprising a processor and a memory, wherein the memory stores a computer program, and the processor executes the method according to any one of claims 1 to 7 when calling the computer program in the memory.

10. A computer-readable storage medium storing a plurality of instructions adapted to be loaded by a processor to perform the method of any one of claims 1 to 7.