Material demand forecasting method, device, computer equipment and storage medium

Abstract

The present application provides a material demand forecasting method, device, computer equipment and storage medium, the method comprising: obtaining historical work data of a target user within a preset time period, the historical work data comprising work time information, the number of goods received and the number of materials used; determining the user type of the target user according to the work time information in the historical work data; analyzing the historical work data based on the user type of the target user to obtain the predicted number of goods received by the target user; obtaining the predicted material demand of the target user according to the predicted number of goods received, the number of goods received and the number of materials used. This method is used to improve the prediction accuracy of material demand, and the analysis results of the predicted number of goods received and the predicted material demand can be more accurate.

Description

Technical Field

[0001] The present application relates to the field of logistics technology, and specifically to a material demand forecasting method, device, computer equipment and storage medium. Background Art

[0002] With the rapid development of social economy, the volume of goods circulation is increasing, and the rapid growth of goods circulation has led to an increasing demand for logistics packaging. In the current logistics work, in order to improve work efficiency, most companies will allocate a fixed number of logistics packaging to staff in advance to meet their daily logistics work material needs, but a small number of companies use artificial intelligence technology to predict material needs.

[0003] However, most of the existing material demand forecasting methods using artificial intelligence technology use a single material or material classification method for model forecasting. The forecasting method is single and does not start from business logic, resulting in the demand forecast results of some materials being far from the actual demand.

[0004] Therefore, the existing material demand forecasting methods have the technical problem of low forecasting accuracy. Summary of the invention

[0005] Based on this, it is necessary to provide a material demand forecasting method, device, computer equipment and storage medium to address the above technical problems, so as to improve the forecasting accuracy of material demand.

[0006] In a first aspect, the present application provides a material demand forecasting method, the method comprising:

[0007] Obtain historical work data of the target user within a preset time period, wherein the historical work data includes work time information, the number of goods received, and the number of materials used;

[0008] Determining the user type of the target user according to the working time information in the historical working data;

[0009] Analyze the historical work data based on the user type of the target user to obtain the predicted receipt volume of goods for the target user;

[0010] The predicted material demand of the target user is obtained according to the predicted cargo receipt quantity, the cargo receipt quantity and the material usage quantity.

[0011] In some embodiments of the present application, the step of determining the user type of the target user according to the working time information in the historical working data includes:

[0012] If the working time information is less than a preset first time threshold, determining that the user type of the target user is a first user type;

[0013] If the working time information is greater than or equal to the first time threshold and less than or equal to a preset second time threshold, determining that the user type of the target user is the second user type;

[0014] If the working time information is greater than the second time threshold, it is determined that the user type of the target user is a third user type, wherein the first time threshold is less than the second time threshold.

[0015] In some embodiments of the present application, the step of analyzing the historical work data based on the user type of the target user to obtain the predicted cargo receipt volume of the target user includes:

[0016] If the user type of the target user is the first user type, determining the target work outlet to which the target user belongs according to the historical work data of the target user, and analyzing the cargo receipt volume corresponding to each user of the target work outlet within a preset time period to obtain the predicted cargo receipt volume of the target user;

[0017] If the user type of the target user is the second user type, acquiring characteristic information of the target user according to the historical work data of the target user, and analyzing the characteristic information to obtain the predicted receipt volume of goods of the target user;

[0018] If the user type of the target user is the third user type, the target work point to which the target user belongs is determined based on the historical work data of the target user, and the number of cargo receipts and the cargo receipt volume corresponding to each user at the target work point within a preset time period are analyzed to obtain the predicted cargo receipt volume of the target user.

[0019] In some embodiments of the present application, if the user type of the target user is the first user type, determining the target work outlet to which the target user belongs according to the historical work data of the target user, and analyzing the cargo receipt volume corresponding to each user of the target work outlet within a preset period of time to obtain the predicted cargo receipt volume of the target user includes:

[0020] If the user type of the target user is the first user type, determining the target work network point to which the target user belongs according to the historical work data of the target user;

[0021] Obtaining the cargo receiving volume corresponding to each user at the target work point within a preset time period, and arranging each cargo receiving volume in ascending order to obtain a cargo receiving volume sequence;

[0022] Determine a preset percentile of the cargo receipt volume in the cargo receipt volume sequence as a first target receipt volume;

[0023] The first target receipt volume is analyzed based on a preset Poisson distribution function to obtain a predicted receipt volume of goods for the target user.

[0024] In some embodiments of the present application, if the user type of the target user is the second user type, the step of acquiring characteristic information of the target user according to the historical work data of the target user, and analyzing the characteristic information to obtain the predicted receipt volume of goods of the target user includes:

[0025] If the user type of the target user is the second user type, acquiring characteristic information of the target user according to the historical work data of the target user, the characteristic information including at least one of the following: basic characteristics, personal mail volume characteristics, branch mail volume characteristics, and district mail volume characteristics;

[0026] Analyze and predict the characteristic information to obtain the predicted number of mails received by the target user;

[0027] If the predicted receipt quantity is less than or equal to the cargo receipt quantity, determining the predicted receipt quantity as the cargo predicted receipt quantity;

[0028] If the predicted receipt quantity is greater than the cargo receipt quantity, the cargo receipt quantity is determined as the cargo predicted receipt quantity.

[0029] In some embodiments of the present application, if the user type of the target user is the third user type, the step of determining the target work outlet to which the target user belongs according to the historical work data of the target user, and analyzing the number of received goods and the number of received goods corresponding to each user at the target work outlet within a preset period of time to obtain the predicted number of received goods of the target user includes:

[0030] If the user type of the target user is the third user type, determining the target work network point to which the target user belongs according to the historical work data of the target user;

[0031] Obtaining the cargo receiving volume corresponding to each user at the target work point within a preset time period, and arranging each cargo receiving volume in ascending order to obtain a cargo receiving volume sequence;

[0032] Determine the cargo receipt volume of the first preset percentile in the cargo receipt volume sequence as the second target receipt volume, and determine the cargo receipt volume of the second preset percentile in the cargo receipt volume sequence as the third target receipt volume;

[0033] The predicted cargo receipt quantity of the target user is obtained according to the second target receipt quantity, the third target receipt quantity and the cargo receipt quantity, wherein the first preset percentile is less than the second preset percentile.

[0034] In some embodiments of the present application, the step of obtaining the predicted cargo receipt quantity of the target user according to the second target receipt quantity, the third target receipt quantity and the cargo receipt quantity includes:

[0035] If the number of received goods is less than the second target number of received goods, acquiring feature information in the historical work data, and analyzing the feature information based on the first classifier to obtain the predicted number of received goods of the target user;

[0036] If the number of received goods is greater than or equal to the second target number of received goods, and less than or equal to the third target number of received goods, then acquiring feature information in the historical work data, and analyzing the feature information based on the second classifier to obtain the predicted number of received goods of the target user;

[0037] If the number of cargo receipts is less than the third target receipt quantity, characteristic information in the historical work data is obtained, and the characteristic information is analyzed based on the third classifier to obtain the predicted cargo receipt quantity of the target user, wherein the first classifier, the second classifier and the third classifier are classifiers trained by different training data.

[0038] In some embodiments of the present application, the step of obtaining the predicted material demand of the target user according to the predicted cargo receipt quantity, the cargo receipt quantity and the material usage quantity includes:

[0039] According to the material usage quantity, the material quantity on hand, the material quantity in transit and the material consumption rate of each material are obtained, wherein the material consumption rate is the quotient of the material usage quantity and the cargo receipt quantity, and the material usage quantity includes the material quantity on hand and the material quantity in transit;

[0040] Obtain the product of the material consumption rate and the predicted receipt quantity of the goods, and obtain the sum of the quantity of the materials on hand and the quantity of the materials in transit;

[0041] The difference between the product value and the sum value is obtained to obtain the predicted material demand of the target user for each material.

[0042] In some embodiments of the present application, after obtaining the material quantity on hand, the material quantity in transit and the material consumption rate of each material according to the material usage quantity, the method further includes:

[0043] If the material consumption rate is greater than or equal to a preset consumption rate threshold, the value of the material consumption rate is determined as the consumption rate threshold;

[0044] If the material consumption rate is less than the consumption rate threshold, it is determined that the value of the material consumption rate remains unchanged.

[0045] In some embodiments of the present application, after obtaining the predicted material demand of the target user according to the predicted cargo receipt quantity, the cargo receipt quantity and the material usage quantity, the method further includes:

[0046] Determining the material type corresponding to the predicted material demand;

[0047] Get the minimum packaging quantity corresponding to the material type;

[0048] Determine the pending demand quantity corresponding to the predicted material demand quantity according to the quotient surplus value between the predicted material demand quantity and the minimum packaging quantity;

[0049] The pending demand quantity is sent to the terminal to receive demand quantity confirmation information fed back by the target user through the terminal.

[0050] In some embodiments of the present application, after sending the pending demand to the terminal, the method further includes:

[0051] If the demand quantity confirmation information fed back by the target user through the terminal is received, the undetermined demand quantity is determined as the to-be-delivered demand quantity, and the total replenishment quantity of each to-be-delivered material is obtained according to the to-be-delivered demand quantity;

[0052] Determine the replenishment frequency of each of the materials to be dispatched according to the total replenishment amount of each of the materials to be dispatched;

[0053] When the replenishment frequency includes at least two replenishment frequencies, the replenishment quantity corresponding to each replenishment frequency is determined according to the demand quantity to be shipped, the replenishment frequency and the minimum packaging quantity.

[0054] In some embodiments of the present application, when the replenishment frequency includes at least two replenishment frequencies, the step of determining the replenishment quantity of each replenishment frequency according to the waiting demand quantity, the replenishment frequency and the minimum packaging quantity includes:

[0055] When the replenishment frequency includes at least two replenishment frequencies, obtaining a quotient between the to-be-shipped demand quantity and the replenishment frequency;

[0056] Based on the minimum packaging quantity, the quotient is rounded to determine the replenishment quantity corresponding to each replenishment frequency.

[0057] In a second aspect, the present application provides a material demand forecasting device, the device comprising:

[0058] A data acquisition module is used to acquire the historical work data of the target user within a preset time period, wherein the historical work data includes work time information, the number of goods received, and the number of materials used;

[0059] A type determination module, used to determine the user type of the target user according to the working time information in the historical working data;

[0060] A data analysis module, configured to analyze the historical work data based on the user type of the target user to obtain the predicted receipt volume of goods of the target user;

[0061] The demand determination module is used to obtain the predicted material demand of the target user according to the predicted cargo receipt quantity, the cargo receipt quantity and the material usage quantity.

[0062] In a third aspect, the present application further provides a server, the server comprising:

[0063] one or more processors;

[0064] Memory; and

[0065] One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to implement the material demand forecasting method.

[0066] In a fourth aspect, the present application also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program is loaded by a processor to execute the steps in the material demand forecasting method.

[0067] The above-mentioned material demand forecasting method, device, computer equipment and storage medium can determine the user type of the target user by obtaining the working time information of the target user within a preset time, and can use different analysis methods to obtain the predicted cargo receipt volume of each type of target user, so that the analysis results of the predicted cargo receipt volume are more accurate and the prediction accuracy of material demand is improved. At the same time, from the perspective of the proportional relationship between the predicted cargo receipt volume and the predicted material demand volume and business logic, the predicted material demand volume of the target user is obtained based on the known predicted cargo receipt volume analysis, which can not only improve the prediction accuracy of material demand, but also improve the work efficiency of material distribution. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0069] Figure 1 It is a schematic diagram of a scenario of a material demand forecasting method in an embodiment of the present application;

[0070] Figure 2 It is a flow chart of the material demand forecasting method in the embodiment of the present application;

[0071] Figure 3 This is a characteristic information diagram of the material demand forecasting method in the embodiment of the present application;

[0072] Figure 4 It is a specific flow chart of the material demand forecasting method in the embodiment of the present application;

[0073] Figure 5 This is a schematic diagram of the application process of the material demand forecasting method in the embodiment of the present application;

[0074] Figure 6 It is a structural schematic diagram of a material demand forecasting device in an embodiment of the present application;

[0075] Figure 7 It is a schematic diagram of the structure of a computer device in an embodiment of the present application. DETAILED DESCRIPTION

[0076] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0077] In the description of the present application, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present application, "plurality" means two or more, unless otherwise clearly and specifically defined.

[0078] In the description of the present application, the word "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in the present application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present application.

[0079] In the embodiment of the present application, the material demand forecasting method is mainly applied to the field of artificial intelligence (AI). Among them, artificial intelligence is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technology of computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can respond in a similar way to human intelligence.

[0080] It should be noted that in the embodiments of the present application, since the material demand forecasting method provided in the present application is executed in a computer device, the processing objects of each computer device exist in the form of data or information. For example, time is actually time information. It can be understood that if size, quantity, location, etc. are mentioned in subsequent embodiments, the corresponding data exist for the computer device to process, and the details will not be repeated here.

[0081] The embodiments of the present application provide a material demand forecasting method, apparatus, computer equipment, and storage medium, which are described in detail below.

[0082] See also Figure 1 , Figure 1A schematic diagram of a scenario of a material demand forecasting method provided in an embodiment of the present application, which material demand forecasting method can be applied to a material demand forecasting system. The material demand forecasting system includes a terminal 100 and a server 200. The terminal 100 can be a device that includes both receiving and transmitting hardware, that is, a device with receiving and transmitting hardware that can perform two-way communication on a two-way communication link. Such a device may include: a cellular or other communication device having a single-line display or a multi-line display or a cellular or other communication device without a multi-line display. The terminal 100 can specifically be a desktop terminal or a mobile terminal, and the terminal 100 can specifically be one of a mobile phone, a tablet computer, a laptop computer, etc. The server 200 can be an independent server, or a server network or a server cluster composed of servers, which includes but is not limited to a computer, a network host, a single network server, a plurality of network server sets, or a cloud server composed of a plurality of servers. The cloud server is composed of a large number of computers or network servers based on cloud computing (Cloud Computing).

[0083] Those skilled in the art will understand that Figure 1 The application environment shown in the figure is only one application scenario of the present application solution and does not constitute a limitation on the application scenario of the present application solution. Other application environments may also include Figure 1 More or less computer equipment as shown in Figure 1 Only one server 200 is shown in the figure. It is understandable that the material demand forecasting system may also include one or more other servers, which are not specifically limited here. Figure 1 As shown, the material demand forecasting system may also include a memory 300 for storing data, such as logistics data, for example, various data of the logistics platform, such as logistics transportation information of logistics outlets such as transfer yards, specifically, express information, delivery vehicle information and logistics outlet information.

[0084] It should be noted that Figure 1 The scenario diagram of the material demand forecasting system shown is only an example. The material demand forecasting system and scenario described in the embodiment of the present invention are intended to more clearly illustrate the technical solution of the embodiment of the present invention, and do not constitute a limitation on the technical solution provided by the embodiment of the present invention. Ordinary technicians in this field can know that with the evolution of the material demand forecasting system and the emergence of new business scenarios, the technical solution provided by the embodiment of the present invention is also applicable to similar technical problems.

[0085] See also Figure 2 The present application embodiment provides a material demand forecasting method, which is mainly applied to the above Figure 1Taking the server 200 in FIG. 1 as an example, the method includes steps S201 to S204, which are specifically as follows:

[0086] S201, obtaining historical work data of a target user within a preset time period, wherein the historical work data includes work time information, the number of goods received, and the number of materials used.

[0087] Among them, the users involved in this application are all logistics staff, including but not limited to the delivery staff responsible for receiving and delivering parcels in logistics companies; the target users are the staff selected to predict the demand for materials. It is understandable that since the daily work of the delivery staff includes receiving (collecting parcels), creating order information, delivering parcels, etc., their daily work will accumulate a lot of work data. Based on these work data, the material demand forecast proposed in this application can be implemented, the quantity of materials required by the target user can be determined in advance, and the corresponding amount of materials can be allocated to the target user in advance, which can improve their logistics work efficiency.

[0088] Among them, the preset duration is a historical period selected for material demand forecasting, and the specific period can be determined according to actual business needs, such as a certain month or year in the past. Based on the above description, since each deliveryman will accumulate a lot of work data in his daily work, for the purpose of predicting future material demand, the historical work data of the target user is selected according to the preset duration, that is, the work data existing in this historical period can be used as the analysis basis for material demand forecasting, so as to achieve the purpose of predicting material demand.

[0089] Among them, the working time information refers to the time when the delivery person joins the company, including but not limited to the number of days, months or years of employment. That is, the embodiment of the present application does not specifically limit the time unit of the working time information.

[0090] The number of received goods refers to the number of goods received by the delivery person within a preset time period, for example, 10 pieces of goods, 50 pieces of goods, and 100 pieces of goods. It should be noted that, since in actual application scenarios, there may be a situation where multiple pieces of goods are sent in combination and share one waybill, this embodiment proposes that the number of received goods is the number of received goods based on the number of waybills, that is, a waybill of 1 represents a number of received goods of 1, regardless of how many pieces of goods are included in the waybill.

[0091] Among them, the materials involved in this application are all materials required for logistics, including but not limited to tapes, adhesive papers, document envelopes and other cargo packaging materials, as well as paper waybills and other cargo label materials; the quantity of materials used refers to the quantity of logistics materials used by the delivery personnel, for example, 1 piece, 3 pieces, 10 pieces, etc. It should be noted that the above materials can also be logistics materials of the same type but with different specifications, for example, a large packaging tape of 46.5cm*38cm*0.05mm and a medium packaging tape of 40.5cm*28cm*0.05mm, which need to be counted in different quantities of materials used.

[0092] Specifically, the terminal 100 involved in the embodiment of the present application can run a specific application software, which can not only allow the delivery personnel to record logistics order information, but also allow the delivery personnel to view the statistics of the collection and use of various materials, including but not limited to the name of the collected (used) materials, the collection (use) material identification, the collection (use) material specifications, the collection (use) material quantity, etc.

[0093] More specifically, part of the target user's work data can be entered in real time through the terminal 100 and stored in the database of the server 200. The entry time may be the creation time of the logistics order, which depends on the collection time of the goods to be sent. The part of the work data that can be entered in real time includes data related to the logistics order, such as recipient information, sender information, recipient address information, etc. In addition to these, other work data may be data obtained by statistics by the server 200 through analyzing the work data entered in real time, such as the number of goods received, the number of materials used, etc. Among them, the number of goods received only requires the server 200 to count the number of waybills completed by the target user, and the number of materials used only requires the server 200 to count the number of different types of materials received (the number of materials obtained by applying to the work unit through the terminal 100), the number in transit (the number of materials confirmed by the delivery staff and in transit waiting to be delivered to the delivery staff) and the number on hand (the number of materials currently owned by the delivery staff); the working time information only requires the server 200 to analyze the target user's joining date and current computer time to determine, for example, the target user's joining date is January 1, 2020, and the current computer time is October 1, 2020, then the server 200 can analyze the target user's working time information to be specifically 274 days, or roughly 9 months.

[0094] S202: Determine the user type of the target user according to the working time information in the historical working data.

[0095] Among them, the user type involved in the embodiment of the present application is a user type determined by the user's working hours. For example, different intervals are divided by a certain critical time. If the working hours of the target user fall into a certain interval, then his user type corresponds to the user type belonging to the interval; for another example, the currently set critical time is 30 days, and the working time interval less than or equal to 30 days corresponds to user type A, and the working time interval greater than 30 days corresponds to user type B. If the working hours of the target user are 25 days, then his user type is A. It should be noted that although the embodiment of the present application points out that the working time information is the number of days after joining the company, it does not exclude that the working time information in other embodiments is the number of working days in a periodic period, such as the number of working days in a week, the number of working days in a month, etc., which is not limited in this embodiment.

[0096] Specifically, after the server 200 obtains the working time information of the target user within the preset time, it can match the working time information with the pre-divided time interval, and then determine the user type of the target user based on the matching result. Among them, the number of time intervals and the time coverage range can be determined according to actual application requirements, and the embodiment of the present application does not make specific limitations. The user type determination step involved in this embodiment will be described in detail below.

[0097] In one embodiment, this step includes: if the working time information is less than a preset first time threshold, determining that the user type of the target user is a first user type; if the working time information is greater than or equal to the first time threshold and less than or equal to a preset second time threshold, determining that the user type of the target user is a second user type; if the working time information is greater than the second time threshold, determining that the user type of the target user is a third user type, wherein the first time threshold is less than the second time threshold.

[0098] Among them, the first duration threshold and the second duration threshold are thresholds set according to actual business needs. For example, the first duration threshold is 30 days and the second duration threshold is 60 days. The specific reason is that: the delivery personnel who have been employed for less than 30 days have almost no historical receipt volume, and their receipt volume is unstable, so it is difficult to predict the receipt volume, so they need to be classified and predicted separately; the delivery personnel who have been employed for 30-60 days have a certain historical receipt behavior, but it is not stable, so they need to be classified into one category for prediction; the delivery personnel who have been employed for more than 60 days have a certain historical receipt behavior, and it is relatively stable, so they can be classified into one category for prediction. It should be noted that although the embodiments of the present application have given examples to illustrate the specific values ​​of the first duration threshold and the second duration threshold, it does not rule out the selection of other numerical values ​​in other business scenarios, and the specific embodiments of the present application are not limited.

[0099] Specifically, the first duration threshold and the second duration threshold can divide the working time into three intervals. Depending on which interval the working time information of the target user falls into, his user type can be determined as the user type corresponding to the interval, and the first user type, the second user type and the third user type are the user types corresponding to the above three intervals respectively.

[0100] For example, the first duration threshold is 30 days and the second duration threshold is 60 days. The user type corresponding to the duration interval of 0-30 days is E, the user type corresponding to the duration interval of 30 days-60 days is F, and the user type corresponding to the duration interval of 60 days-+∞ (positive infinity) days is G. If the working time information of the target user is 45 days, then its user type is F.

[0101] S203: Analyze the historical work data based on the user type of the target user to obtain a predicted cargo receipt volume of the target user.

[0102] Among them, the predicted cargo receipt volume is the cargo receipt volume obtained based on the forecast analysis of historical work data. The purpose of predicting the cargo receipt volume is that the more cargo receipt volume, the greater the demand for materials. Therefore, it is necessary to analyze the proportional relationship between it and the material demand, and finally predict the material demand.

[0103] Specifically, based on the detailed description of the above steps, this application proposes that different types of target users need to be classified into one category and predicted separately. The reason is that different types of target users have different amounts of accumulated work data. If the types are not distinguished and predicted uniformly, it may cause target users with insufficient work data, and their prediction results may differ greatly from actual needs. Therefore, the embodiment of this application will select the corresponding rules to analyze the historical work data of the target user according to the user type analyzed in the previous steps, so as to obtain the predicted receipt volume of goods before obtaining the predicted demand for materials. The historical work data analysis steps involved in this embodiment will be described in detail below.

[0104] In one embodiment, this step includes: if the user type of the target user is the first user type, determining the target work point to which the target user belongs based on the historical work data of the target user, and analyzing the cargo receipt volume of each user corresponding to the target work point within a preset time period to obtain the predicted cargo receipt volume of the target user; if the user type of the target user is the second user type, acquiring the characteristic information of the target user based on the historical work data of the target user, and analyzing the characteristic information to obtain the predicted cargo receipt volume of the target user; if the user type of the target user is the third user type, determining the target work point to which the target user belongs based on the historical work data of the target user, and analyzing the cargo receipt volume and the cargo receipt volume of each user corresponding to the target work point within the preset time period to obtain the predicted cargo receipt volume of the target user.

[0105] Among them, the outlets involved in the embodiments of the present application are all express delivery outlets used in the logistics field. Express delivery outlets refer to stores that receive and send parcels in a certain area. The target work outlet to which the target user belongs is equivalent to the relationship between an employee and a department. The target user's performance data such as the number of parcels received are all included in the total performance of the target work outlet.

[0106] Specifically, the user types involved in the embodiments of the present application include three types: the first user type, the second user type and the third user type. Different analysis schemes are preset for different user types, that is, the analysis scheme corresponding to the first user type is to use the data of 25% of the number of mails received by the outlets and the Poisson distribution function for analysis and prediction, the analysis scheme corresponding to the second user type is to use the LightGBM model (LightGBM is a fast, distributed, high-performance gradient boosting framework based on the decision tree algorithm) to analyze and predict the characteristic information of the target user, and the analysis scheme corresponding to the third user type is to analyze and predict based on the comparison results between the average number of mails received by the target user and the average number of mails received by the outlets. The three analysis schemes involved in this embodiment are described in detail below.

[0107] In one embodiment, if the user type of the target user is the first user type, then according to the historical work data of the target user, the target work point to which the target user belongs is determined, and the cargo receipt volume of each user corresponding to the target work point within a preset time period is analyzed to obtain the predicted cargo receipt volume of the target user. The step includes: if the user type of the target user is the first user type, then according to the historical work data of the target user, the target work point to which the target user belongs is determined; the cargo receipt volume of each user corresponding to the target work point within a preset time period is obtained, and each of the cargo receipt volumes is arranged in ascending order to obtain a cargo receipt volume sequence; the cargo receipt volume of a preset percentile in the cargo receipt volume sequence is determined as the first target receipt volume; and the first target receipt volume is analyzed based on a preset Poisson distribution function to obtain the predicted cargo receipt volume of the target user.

[0108] The cargo receiving volume refers to the cargo receiving volume of each user at the target work point to which the target user belongs within a preset time period, for example, 10 pieces, 20 pieces, or 30 pieces.

[0109] Specifically, after the server 200 analyzes and determines that the user type of the target user is the first user type, it will first determine the target work outlet to which the target user belongs based on the historical work data of the target user in accordance with the first of the three analysis schemes described above. The determination factor of the target work outlet can be the target user's entry division outlet, or the target user's frequent work outlet, or the target user's recent work outlet, which can be determined specifically according to actual business needs. After the server 200 analyzes and determines the target work outlet of the target user, it can further query and obtain the cargo receipt volume of all users corresponding to the target work outlet within a preset time period, and the determination of the target work outlet corresponding to all users depends on the determination method of the target work outlet to which the target user belongs. After the server 200 obtains the cargo receipt volume corresponding to each user of the target work site, in order to obtain the preset percentile of the site receipt volume as a basis for subsequent analysis, it is necessary to sort the cargo receipt volume of each user in ascending order, that is, the sorting order is from small to large for the cargo receipt volume, and obtain the sorted cargo receipt volume sequence, and then determine the cargo receipt volume at the preset percentile in the cargo receipt volume sequence as the first target receipt volume, and then use the first target receipt volume to participate in the preset Poisson distribution function for analysis to obtain the predicted cargo receipt volume of the target user. It should be noted that the preset percentile involved in the embodiment of the present application can be any percentile set according to actual business needs, for example, 25%, 45%, etc.; although the arrangement method involved in the embodiment of the present application is in ascending order, it does not exclude descending order in other embodiments.

[0110] More specifically, the analysis involving the Poisson distribution function is under a stable state. It is assumed that the distribution density of the number of parcels received by the courier (target user) obeys an exponential distribution with a parameter λ, and the material demand is a constant λ. The probability of the occurrence of material demand can be determined by the relevant formula of the Poisson distribution.

[0111] For example, if the number of parcels received by the courier follows a Poisson distribution with parameter λt, then the probability that the number of parcels received by the courier is K within time t is:

[0112] C is a positive integer

[0113] Among them, λt is the amount of goods received within time t, which can be obtained by the preset percentile of the number of goods received at the outlets. P(t) is the probability that the amount of goods received by the delivery person is K. Given the probability P(t), the amount of goods received K can be calculated.

[0114] For example, if the preset percentile is 25%, the average number of parcels received by each delivery person in the outlet is first calculated, and then the top 25% of the parcels received are sorted from small to large, indicating that 25% of all these parcels received are less than this value. For example, the current 11 parcels received are sorted as follows: 1, 3, 4, 6, 7, 9, 12, 15, 16, 18, 20, and the 25% percentile position is (11+1) / 4=3, that is, in this group of parcel quantity sequences, the number arranged in the third position is the 25% percentile, and the corresponding parcel quantity is 4. It can be determined that the first target parcel quantity (λt) is 4 in this example.

[0115] In one embodiment, if the user type of the target user is the second user type, then according to the historical work data of the target user, the characteristic information of the target user is obtained, and the characteristic information is analyzed to obtain the predicted cargo receipt volume of the target user, the step includes: if the user type of the target user is the second user type, then according to the historical work data of the target user, the characteristic information of the target user is obtained, and the characteristic information includes at least one of the following: basic characteristics, personal receipt volume characteristics, branch receipt volume characteristics and district receipt volume characteristics; analyzing and predicting the characteristic information to obtain the predicted receipt volume of the target user; if the predicted receipt volume is less than or equal to the cargo receipt quantity, determining the predicted receipt volume as the predicted cargo receipt quantity; if the predicted receipt volume is greater than the cargo receipt quantity, determining the cargo receipt quantity as the predicted cargo receipt quantity.

[0116] The characteristic information is the directional characteristic information obtained by analyzing the historical work data of the target user, including but not limited to basic characteristics, personal receipt volume characteristics, branch receipt volume characteristics and district receipt volume characteristics. For details of each type of characteristic information, please refer to Figure 3 .

[0117] Specifically, the analysis process of the second analysis scheme corresponding to the second user type involves the LightGBM model. The LightGBM model has been described in the previous embodiment as a fast, distributed, high-performance gradient boosting framework based on the decision tree algorithm, which can be used for sorting, classification, regression and many other machine learning tasks. In this application, the LightGBM model can be used to analyze the feature information to output the predicted receipt volume of the target user, and finally compare the predicted receipt volume with the number of goods received by the target user within a preset historical period, and take the minimum value of the two as the predicted receipt volume of goods.

[0118] In one embodiment, if the user type of the target user is the third user type, the target work outlet to which the target user belongs is determined according to the historical work data of the target user, and the cargo receipt quantity and the cargo receipt quantity of each user corresponding to the target work outlet within a preset time period are analyzed to obtain the predicted cargo receipt quantity of the target user, including: if the user type of the target user is the third user type, the target work outlet to which the target user belongs is determined according to the historical work data of the target user; the cargo receipt quantity of each user corresponding to the target work outlet within a preset time period is obtained, and each of the cargo receipt quantities is arranged in ascending order to obtain a cargo receipt quantity sequence; the cargo receipt quantity of the first preset percentile in the cargo receipt quantity sequence is determined as the second target receipt quantity, and the cargo receipt quantity of the second preset percentile in the cargo receipt quantity sequence is determined as the third target receipt quantity; the predicted cargo receipt quantity of the target user is obtained according to the second target receipt quantity, the third target receipt quantity and the cargo receipt quantity, wherein the first preset percentile is less than the second preset percentile.

[0119] Among them, the first preset percentile and the second preset percentile are percentiles set according to actual business needs, and the first preset percentile is smaller than the second preset percentile. For example, the first preset percentile is 15% and 25%, and the second preset percentile is 45% and 75%.

[0120] Specifically, the analysis process of the third analysis scheme corresponding to the third user type also involves the LightGBM model, and involves three LightGBM models. The three models are trained by different model training data. The type of each model training data depends on the size of the receipt volume of the delivery personnel and the receipt volume of the outlets to which they belong, specifically the difference in the size of the receipt volume of the target user and the receipt volume of the outlets to which the target user belongs at the preset percentile. For example, in this embodiment, the first preset percentile is set to 25%, and the second preset percentile is set to 75%. 25% is less than 75% to meet the value condition. Before comparing the receipt volume values, the server 200 first needs to determine the target work outlet to which the target user belongs, and then obtain the target work outlet corresponding to each user. The target work outlet is arranged. This embodiment illustrates the selection of ascending order to obtain the arranged cargo receipt volume sequence. For the cargo receipt quantity sequence, it is necessary to further select the comparison basis of the subsequent receipt quantity based on the first preset percentile and the second preset percentile, that is, determine the 25% percentile in the cargo receipt quantity sequence as the second target receipt quantity; determine the 75% percentile in the cargo receipt quantity sequence as the third target receipt quantity. The screening principle is the same as the percentile principle described in the above embodiment, and will not be repeated here. Finally, the cargo receipt quantity of the target user in the historical work data is compared with the second target receipt quantity and the third target receipt quantity, and the predicted cargo receipt quantity of the target user is finally determined based on the comparison results.

[0121] For example, the current cargo receipt quantity sequence is: 1, 3, 4, 6, 7, 9, 12, 15, 16, 18, 20, the 25% quantile position is (11+1) / 4=3, and the corresponding second target receipt quantity is 4; the 75% quantile position is (11+1) / 4*3=9, and the corresponding third target receipt quantity is 16. If the target user has 12 cargo receipts in the historical work data, the relationship between the cargo receipt quantity 12, the second target receipt quantity 4, and the third target receipt quantity 16 can be analyzed to determine the target user's predicted cargo receipt quantity.

[0122] In one embodiment, the step of obtaining the predicted cargo receipt volume of the target user based on the second target receipt volume, the third target receipt volume and the cargo receipt volume includes: if the cargo receipt volume is less than the second target receipt volume, obtaining feature information in the historical work data, and analyzing the feature information based on the first classifier to obtain the predicted cargo receipt volume of the target user; if the cargo receipt volume is greater than or equal to the second target receipt volume and less than or equal to the third target receipt volume, obtaining feature information in the historical work data, and analyzing the feature information based on the second classifier to obtain the predicted cargo receipt volume of the target user; if the cargo receipt volume is less than the third target receipt volume, obtaining feature information in the historical work data, and analyzing the feature information based on the third classifier to obtain the predicted cargo receipt volume of the target user, wherein the first classifier, the second classifier and the third classifier are classifiers trained by different training data.

[0123] Among them, the first classifier, the second classifier and the third classifier are trained using different model training data to obtain the LightGBM model, and the model training data comes from the feature information in the user's (delivery staff) historical work data.

[0124] Specifically, when the number of received goods of the target user is less than the second target received goods, the characteristic information in the historical work data can be obtained and input into the trained first classifier for analysis, so as to obtain the output result of the first classifier as the predicted received goods of the target user, and the model training data of the first classifier is also the characteristic information of the user whose number of received goods is less than the second target received goods. When the number of received goods of the target user is greater than or equal to the second target received goods, and less than or equal to the third target received goods, the characteristic information in the historical work data can be obtained and input into the trained second classifier for analysis, so as to obtain the output result of the second classifier as the predicted received goods of the target user, and the model training data of the second classifier is also the characteristic information of the user whose number of received goods is between the second target received goods and the third target received goods. When the number of received goods of the target user is less than the third target received goods, the characteristic information in the historical work data can be obtained and input into the trained third classifier for analysis, so as to obtain the output result of the third classifier as the predicted received goods of the target user, and the model training data of the third classifier is also the characteristic information of the user whose number of received goods is less than the third target received goods.

[0125] For example, the second target quantity of received parcels is 4 and the third target quantity of received parcels is 16. If the target user's quantity of received parcels is 3, the quantity of received parcels is less than the second target quantity of received parcels, and the server 200 can obtain the target user's characteristic information and input it into the first classifier for analysis; if the target user's quantity of received parcels is 5, the quantity of received parcels is greater than the second target quantity of received parcels and less than the third target quantity of received parcels, and the server 200 can obtain the target user's characteristic information and input it into the second classifier for analysis; if the target user's quantity of received parcels is 20, the quantity of received parcels is greater than the third target quantity of received parcels, and the server 200 can obtain the target user's characteristic information and input it into the third classifier for analysis.

[0126] S204: Obtain the predicted material demand of the target user according to the predicted cargo receipt quantity, the cargo receipt quantity and the material usage quantity.

[0127] The predicted material demand refers to the demand quantity obtained by the current demand forecast for logistics materials, for example, 10, 20, 50, etc.

[0128] Specifically, the parameters required for calculating the predicted demand for materials can be obtained based on the quantity of materials used, such as the quantity of materials on hand and the quantity of materials in transit included in the quantity of materials used. The parameters required for calculating the predicted demand for materials can also be obtained based on the quantity of materials used and the quantity of goods received, such as the quotient of the quantity of materials used and the quantity of goods received, that is, the material consumption rate. Finally, the predicted demand for materials of the target user can be analyzed and obtained based on the material consumption rate, the predicted quantity of goods received, the quantity of materials on hand, and the quantity of materials in transit. The steps for obtaining the predicted demand for materials involved in this application will be described in detail below.

[0129] In one embodiment, this step includes: according to the material usage quantity, obtaining the material quantity on hand, the material quantity in transit and the material consumption rate of each material, the material consumption rate is the quotient of the material usage quantity and the cargo receipt quantity, and the material usage quantity includes the material quantity on hand and the material quantity in transit; obtaining the product value between the material consumption rate and the predicted cargo receipt quantity, and obtaining the sum value between the material quantity on hand and the material quantity in transit; obtaining the difference between the product value and the sum value, and obtaining the predicted material demand of the target user for each material.

[0130] The quantity of materials on hand refers to the number of materials that the target user (delivery person) currently has, for example, 10, 20, 50, etc.

[0131] The quantity of materials in transit refers to the quantity of materials that the target user (delivery person) has confirmed they need and are in transit, for example, 10, 20, 50, etc.

[0132] Specifically, material consumption rate = material usage quantity ÷ cargo receipt quantity, and both material usage quantity and cargo receipt quantity are data in historical work data; material on hand quantity + material in transit quantity + material collection quantity = material usage quantity; therefore, after server 200 obtains the material consumption rate, material on hand quantity and material in transit quantity, it can combine the predicted cargo receipt quantity to analyze and obtain the predicted material demand for the materials, wherein the predicted material demand = material consumption rate * predicted cargo receipt quantity - material on hand quantity - material in transit quantity.

[0133] In one embodiment, after obtaining the quantity of materials on hand, the quantity of materials in transit and the material consumption rate of each material according to the quantity of materials used, the method further includes: if the material consumption rate is greater than or equal to a preset consumption rate threshold, determining the value of the material consumption rate as the consumption rate threshold; if the material consumption rate is less than the consumption rate threshold, determining that the value of the material consumption rate remains unchanged.

[0134] The consumption rate threshold is a preset critical value used to determine whether material consumption is abnormal, for example, 80%.

[0135] Specifically, after the server 200 calculates the material consumption rate, it is necessary to analyze the material consumption rate to determine whether the current material consumption is abnormal. If so, it is necessary to perform correction processing so as to calculate the predicted material demand using a reasonable material consumption rate. Therefore, using a preset consumption rate threshold to analyze whether the material consumption rate is abnormal is a more reasonable abnormal analysis method. The embodiment of the present application proposes that if the calculated material consumption rate is greater than or equal to the consumption rate threshold, the value of the material consumption rate is determined as the consumption rate threshold. If the material consumption rate is less than the consumption rate threshold, it can be considered that the material consumption rate is normal and does not need to be corrected, that is, the value of the material consumption rate is determined to be unchanged. However, it should be noted that for logistics materials in the logistics field, especially packaging tape, scotch tape, conventional waybills, document envelopes and other logistics materials such as document envelopes and packaging tapes, if the material consumption rate is less than 100%, not only does it need to be corrected, but it will also output abnormalities. However, the data of scotch tape and conventional waybills are generally stable, with few abnormal conditions, and no abnormalities will be output.

[0136] In one embodiment, after obtaining the predicted material demand of the target user based on the predicted cargo receipt quantity, the cargo receipt quantity and the material usage quantity, the method further includes: determining the material type corresponding to the predicted material demand; obtaining the minimum packaging quantity corresponding to the material type; determining the pending demand corresponding to the predicted material demand based on the quotient between the predicted material demand and the minimum packaging quantity; and sending the pending demand to the terminal to receive the demand confirmation information fed back by the target user through the terminal.

[0137] Among them, packaging quantity refers to the number or weight of products placed in the outer packaging under a certain standard; minimum packaging quantity refers to the minimum number of materials that can be placed in the outer packaging, for example, 5, 15, etc.

[0138] Specifically, since a single material is usually packaged as a whole according to a certain quantity, whether it is the production or sale of materials, there may be a situation where an outer package contains multiple independent small packages, and the number of items contained in this package is the minimum packaging quantity. Therefore, after the server 200 analyzes and obtains the target user's predicted material demand, it is also necessary to consider the minimum packaging quantity of the corresponding material type. Based on the predicted material demand and the minimum packaging quantity of the corresponding material type, the server 200 analyzes and obtains the valid pending demand corresponding to the predicted material demand, so as to feedback the pending demand to the target user (delivery staff) through the terminal 100, and then obtain the confirmation information submitted by the target user through the terminal 100.

[0139] More specifically, according to the quotient between the predicted material demand and the minimum package quantity, the step of determining the pending demand corresponding to the predicted material demand is actually to divide the predicted demand by the minimum package quantity. If the remainder is greater than 50% of the minimum package quantity, the remainder is increased to make up 1 minimum package quantity; if the remainder is less than 50% of the minimum package quantity, the remainder is subtracted. For example, if the predicted material demand of a certain material is 25 and the minimum package quantity is 15, the remainder of the two is 10>(15*50%=7.5), then the remainder is increased by 5 to obtain a minimum package quantity (15), and the pending demand is 30; if the predicted material demand of a certain material is 20 and the minimum package quantity is 15, then the remainder of the two is 5<(15*50%=7.5), then the remainder 5 is subtracted, and the pending demand is 15. The pending demand obtained by parsing the server 200 can be fed back to the target user through the terminal 100 for viewing. After the target user views the pending demand for each material and submits confirmation information, the server 200 can notify the staff of the logistics company to allocate the corresponding materials to the target user to solve the target user's demand for such materials.

[0140] In one embodiment, after sending the pending demand to the terminal, the method further includes: if demand confirmation information fed back by the target user through the terminal is received, determining the pending demand as the pending demand, and obtaining the total replenishment quantity of each pending material based on the pending demand; determining the replenishment frequency of each pending material based on the total replenishment quantity of each pending material; when the replenishment frequency includes at least two replenishment frequencies, determining the replenishment quantity corresponding to each replenishment frequency based on the pending demand, the replenishment frequency and the minimum packaging quantity.

[0141] Among them, the demand quantity to be shipped refers to the quantity of materials determined by the logistics company to be shipped to the delivery personnel, for example, 10, 20, 30, etc.

[0142] The total replenishment quantity refers to the total replenishment weight of materials to be shipped, for example, 10KG, 20KG, etc.

[0143] Among them, replenishment frequency refers to the replenishment frequency of materials to be shipped, for example, 4 times a month, 2 times a week, etc.

[0144] Specifically, after receiving the demand confirmation information fed back by the target user through the terminal 100, the server 200 can determine the pending demand as the waiting demand, and obtain the total replenishment amount of each waiting material based on the individual weight of the material corresponding to the waiting demand and the waiting demand, and then use the replenishment weight to determine the replenishment frequency of the waiting material. For example, if the total replenishment amount is greater than 0KG and less than or equal to 30KG, the replenishment frequency is determined to be 1 time; if the total replenishment amount is greater than 30KG and less than or equal to 50KG, the replenishment frequency is determined to be 2 times; if the total replenishment amount is greater than 50KG and less than or equal to 80KG, the replenishment frequency is determined to be 3 times; if the total replenishment amount is greater than 80KG, the replenishment frequency is determined to be 4 times.

[0145] In one embodiment, when the replenishment frequency includes at least two replenishment frequencies, the step of determining the replenishment quantity of each replenishment frequency according to the waiting demand quantity, the replenishment frequency and the minimum packaging quantity includes: when the replenishment frequency includes at least two replenishment frequencies, obtaining the quotient between the waiting demand quantity and the replenishment frequency; based on the minimum packaging quantity, rounding the quotient to determine the replenishment quantity corresponding to each replenishment frequency.

[0146] Specifically, if the replenishment frequency analyzed by the server 200 is 1, all the materials to be shipped can be delivered to the target user at one time. However, if the replenishment frequency exceeds 1, the replenishment quantity of each replenishment needs to be further analyzed. Therefore, the embodiment of the present application proposes that when the replenishment frequency includes at least two replenishment frequencies, the quotient between the demand quantity to be shipped and the replenishment frequency can be obtained, and then the quotient can be rounded based on the minimum packaging quantity to determine the replenishment quantity corresponding to each replenishment frequency.

[0147] For example, if the demand for a certain material is 45, the replenishment frequency is 4 times, and the minimum package quantity is 5, then 45÷4=11 with a remainder of 1, which is rounded to an integer of 10, and the replenishment quantity for the second, third, and fourth deliveries is determined to be 10, and the replenishment quantity for the first delivery is 15. For another example, if the demand for a certain material is 110, the replenishment frequency is 4 times, and the minimum package quantity is 10, then 110÷4=27 with a remainder of 2, which is rounded to an integer of 30, and the replenishment quantity for the second, third, and fourth deliveries is determined to be 30. The last delivery can be calculated by subtracting the quantity of the previous deliveries from the total weight.

[0148] The above embodiment provides a material demand forecasting method, which determines the user type of the target user by obtaining the working time information of the target user within a preset time period, and can use different analysis methods to obtain the predicted cargo receipt volume of each type of target user, so that the analysis result of the predicted cargo receipt volume is more accurate and the prediction accuracy of material demand is improved. At the same time, from the perspective of the proportional relationship between the predicted cargo receipt volume and the predicted material demand volume and business logic, the predicted material demand volume of the target user is obtained based on the known predicted cargo receipt volume analysis, which can not only improve the prediction accuracy of material demand, but also improve the work efficiency of material distribution.

[0149] It should be understood that Figure 2 At least part of the steps may include multiple steps or multiple stages. These steps or stages are not necessarily performed at the same time, but can be performed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed in turn or alternately with other steps or at least part of the steps or stages in other steps.

[0150] In order to enable those skilled in the art to fully understand the material demand forecasting method proposed in this application, this application also provides an application scenario, which applies the above-mentioned material demand forecasting method. Specifically, the application of the material demand forecasting method in this application scenario will be combined with Figure 4 and Figure 5 The following instructions are given:

[0151] like Figure 4 As shown in the figure, before forecasting material demand, we first need to obtain the historical work data of each target user as the data source for analysis and forecasting, which includes work time information, the number of goods received, and the number of materials used. After obtaining the data source required for analysis and forecasting, we can classify users according to the work time information of each target user so that the analysis results are accurate and effective. There are three categories: the number of days in the job is less than 30 days, the number of days in the job is between 30 and 60 days, and the number of days in the job is more than 60 days.

[0152] For the first type of user (with less than 30 days of employment), the analysis method uses the 25% quantile and Poisson distribution function of the number of parcels received at the outlets. The detailed analysis steps have been fully described in the above embodiments and will not be repeated here. For the second type of user (with between 30 and 60 days of employment), the analysis method uses parameter information such as models and average values. The detailed analysis steps have been fully described in the above embodiments and will not be repeated here. For the third type of user (with more than 60 days of employment), the analysis method uses the ratio of the number of parcels received by the user to the number of parcels received at the outlets. The detailed analysis steps have been fully described in the above embodiments and will not be repeated here. It should be noted that for the analysis method of the third type of user, before inputting the data into the model, it is necessary to further classify the users according to the ratio of the number of parcels received by the users to the number of parcels received at the outlets, that is, including category 1, category 2, and category 3, and then input the historical work data of users of different categories into different models to obtain the predicted number of parcels received by each model.

[0153] Finally, after obtaining the predicted cargo receipt volume, it is necessary to analyze the predicted cargo receipt volume to obtain the predicted material demand, so that after the target user confirms the demand through the terminal, the corresponding demand of each type of material can be delivered to the target user to realize the material replenishment for each target user.

[0154] In addition, it should be noted that after obtaining the predicted material demand, the long short-term memory model (LSTM) can be used for another prediction, and then the abnormal values ​​predicted by the LightGBM model are compared one by one with the data predicted by the LSTM model, and the value closest to the number of goods received is taken as the valid value to achieve the final material replenishment.

[0155] The above-mentioned material demand forecasting method can not only improve the forecasting accuracy of material demand, but also improve the work efficiency of material allocation.

[0156] In order to better implement the material demand forecasting method in the embodiment of the present application, based on the material demand forecasting method, the embodiment of the present application also provides a material demand forecasting device 600, such as Figure 6 As shown, the material demand forecasting device 600 includes:

[0157] The data acquisition module 602 is used to acquire the historical work data of the target user within a preset time period, wherein the historical work data includes work time information, the number of goods received, and the number of materials used;

[0158] A type determination module 604, configured to determine the user type of the target user according to the working time information in the historical working data;

[0159] A data analysis module 606 is used to analyze the historical work data based on the user type of the target user to obtain the predicted receipt volume of goods of the target user;

[0160] The demand determination module 608 is used to obtain the predicted material demand of the target user according to the predicted cargo receipt quantity, the cargo receipt quantity and the material usage quantity.

[0161] In one embodiment, the type determination module 604 is also used to determine that the user type of the target user is a first user type if the working time information is less than a preset first time threshold; if the working time information is greater than or equal to the first time threshold and less than or equal to a preset second time threshold, determine that the user type of the target user is a second user type; if the working time information is greater than the second time threshold, determine that the user type of the target user is a third user type, wherein the first time threshold is less than the second time threshold.

[0162] In one embodiment, the data analysis module 606 is also used to determine the target work point to which the target user belongs based on the historical work data of the target user if the user type of the target user is the first user type, and analyze the cargo receipt volume of each user corresponding to the target work point within a preset time period to obtain the predicted cargo receipt volume of the target user; if the user type of the target user is the second user type, obtain the characteristic information of the target user based on the historical work data of the target user, and analyze the characteristic information to obtain the predicted cargo receipt volume of the target user; if the user type of the target user is the third user type, determine the target work point to which the target user belongs based on the historical work data of the target user, and analyze the cargo receipt volume and the cargo receipt volume of each user corresponding to the target work point within the preset time period to obtain the predicted cargo receipt volume of the target user.

[0163] In one embodiment, the data analysis module 606 is also used to determine the target work point to which the target user belongs based on the historical work data of the target user if the user type of the target user is the first user type; obtain the cargo receipt volume of each user corresponding to the target work point within a preset time period, and arrange the cargo receipt volumes in ascending order to obtain a cargo receipt volume sequence; determine the cargo receipt volume of a preset percentile in the cargo receipt volume sequence as the first target receipt volume; analyze the first target receipt volume based on a preset Poisson distribution function to obtain the predicted cargo receipt volume of the target user.

[0164] In one embodiment, the data analysis module 606 is also used to obtain characteristic information of the target user based on the historical work data of the target user if the user type of the target user is the second user type, the characteristic information including at least one of the following: basic characteristics, personal receipt volume characteristics, branch receipt volume characteristics, and district receipt volume characteristics; analyze and predict the characteristic information to obtain the predicted receipt volume of the target user; if the predicted receipt volume is less than or equal to the cargo receipt quantity, determine the predicted receipt volume as the cargo predicted receipt volume; if the predicted receipt volume is greater than the cargo receipt quantity, determine the cargo receipt quantity as the cargo predicted receipt volume.

[0165] In one embodiment, the data analysis module 606 is also used to determine the target work point to which the target user belongs based on the historical work data of the target user if the user type of the target user is the third user type; obtain the cargo receipt volume of each user corresponding to the target work point within a preset time period, and arrange the cargo receipt volumes in ascending order to obtain a cargo receipt volume sequence; determine the cargo receipt volume of the first preset percentile in the cargo receipt volume sequence as the second target receipt volume, and determine the cargo receipt volume of the second preset percentile in the cargo receipt volume sequence as the third target receipt volume; obtain the predicted cargo receipt volume of the target user based on the second target receipt volume, the third target receipt volume and the cargo receipt volume, wherein the first preset percentile is less than the second preset percentile.

[0166] In one embodiment, the data analysis module 606 is also used to obtain feature information in the historical work data if the number of received goods is less than the second target number of received goods, and analyze the feature information based on the first classifier to obtain the predicted number of received goods for the target user; if the number of received goods is greater than or equal to the second target number of received goods, and less than or equal to the third target number of received goods, obtain feature information in the historical work data, and analyze the feature information based on the second classifier to obtain the predicted number of received goods for the target user; if the number of received goods is less than the third target number of received goods, obtain feature information in the historical work data, and analyze the feature information based on the third classifier to obtain the predicted number of received goods for the target user, wherein the first classifier, the second classifier and the third classifier are classifiers trained using different training data.

[0167] In one embodiment, the demand determination module 608 is also used to obtain the quantity of materials on hand, the quantity of materials in transit and the material consumption rate of each material based on the quantity of materials used, the material consumption rate being the quotient of the quantity of materials used and the quantity of goods received, the quantity of materials used including the quantity of materials on hand and the quantity of materials in transit; obtain the product value between the material consumption rate and the predicted quantity of goods received, and obtain the sum value between the quantity of materials on hand and the quantity of materials in transit; obtain the difference between the product value and the sum value to obtain the predicted material demand of the target user for each material.

[0168] In one embodiment, the demand determination module 608 is also used to determine the value of the material consumption rate as the consumption rate threshold if the material consumption rate is greater than or equal to a preset consumption rate threshold; if the material consumption rate is less than the consumption rate threshold, determine that the value of the material consumption rate remains unchanged.

[0169] In one embodiment, the material demand forecasting device 600 also includes a packaging quantity analysis module, which is used to determine the material type corresponding to the predicted material demand; obtain the minimum packaging quantity corresponding to the material type; determine the pending demand corresponding to the predicted material demand based on the quotient between the predicted material demand and the minimum packaging quantity; and send the pending demand to the terminal to receive the demand confirmation information fed back by the target user through the terminal.

[0170] In one embodiment, the packaging quantity analysis module 610 is also used to determine the pending demand as the to-be-sent demand if it receives demand confirmation information fed back by the target user through the terminal, and obtain the total replenishment quantity of each to-be-sent material based on the to-be-sent demand; determine the replenishment frequency of each to-be-sent material based on the total replenishment quantity of each to-be-sent material; when the replenishment frequency includes at least two replenishment frequencies, determine the replenishment quantity corresponding to each replenishment frequency based on the to-be-sent demand, the replenishment frequency and the minimum packaging quantity.

[0171] In one embodiment, the packaging quantity analysis module 610 is also used to obtain the quotient between the pending demand quantity and the replenishment frequency when the replenishment frequency includes at least two replenishment frequencies; based on the minimum packaging quantity, the quotient is rounded to determine the replenishment quantity corresponding to each replenishment frequency.

[0172] In the above embodiment, not only the prediction accuracy of material demand can be improved, but also the work efficiency of material allocation can be improved.

[0173] In some embodiments of the present application, the material demand forecasting device 600 can be implemented in the form of a computer program. The computer program can be implemented in Figure 7The computer device shown in the figure is run on the computer device. The memory of the computer device can store various program modules constituting the material demand forecasting device 600, for example, Figure 6 The data acquisition module 602, type determination module 604, data analysis module 606 and demand determination module 608 are shown. The computer program composed of various program modules enables the processor to execute the steps of the material demand forecasting method of each embodiment of the present application described in this specification.

[0174] For example, Figure 7 The computer device shown can be Figure 6 The data acquisition module 602 in the material demand forecasting device 600 shown executes step S201. The computer device can execute step S202 through the type determination module 604. The computer device can execute step S203 through the data analysis module 606. The computer device can execute step S204 through the demand determination module 608. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external computer device through a network connection. When the computer program is executed by the processor, a material demand forecasting method is implemented.

[0175] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0176] In some embodiments of the present application, a computer device is provided, comprising one or more processors; a memory; and one or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor in the steps of the material demand forecasting method. The steps of the material demand forecasting method here may be the steps of the material demand forecasting method in each of the above embodiments.

[0177] In some embodiments of the present application, a computer-readable storage medium is provided, which stores a computer program, and the computer program is loaded by a processor, so that the processor executes the steps of the material demand forecasting method described above. The steps of the material demand forecasting method here can be the steps of the material demand forecasting method in each of the above embodiments.

[0178] The above is a detailed introduction to a material demand forecasting method, device, computer equipment and storage medium provided in the embodiments of the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for technical personnel in this field, according to the ideas of the present invention, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A material demand forecasting method, characterized in that: The method comprises: Obtain historical work data of the target user within a preset time period, wherein the historical work data includes work time information, the number of goods received, and the number of materials used; Determining the user type of the target user according to the working time information in the historical working data; Analyze the historical work data based on the user type of the target user to obtain the predicted receipt volume of goods for the target user; Obtaining the predicted material demand of the target user according to the predicted cargo receipt quantity, the cargo receipt quantity and the material usage quantity; The step of determining the user type of the target user according to the working time information in the historical working data includes: If the working time information is less than a preset first time threshold, determining that the user type of the target user is a first user type; If the working time information is greater than or equal to the first time threshold and less than or equal to a preset second time threshold, determining that the user type of the target user is the second user type; If the working time information is greater than the second time threshold, determining that the user type of the target user is a third user type, wherein the first time threshold is less than the second time threshold; The step of analyzing the historical work data based on the user type of the target user to obtain the predicted receipt volume of goods for the target user includes: If the user type of the target user is the first user type, determining the target work outlet to which the target user belongs according to the historical work data of the target user, and analyzing the cargo receipt volume corresponding to each user of the target work outlet within a preset time period to obtain the predicted cargo receipt volume of the target user; If the user type of the target user is the second user type, acquiring characteristic information of the target user according to the historical work data of the target user, and analyzing the characteristic information to obtain the predicted receipt volume of goods of the target user; If the user type of the target user is the third user type, the target work point to which the target user belongs is determined based on the historical work data of the target user, and the number of cargo receipts and the cargo receipt volume corresponding to each user at the target work point within a preset time period are analyzed to obtain the predicted cargo receipt volume of the target user.

2. The material demand forecasting method according to claim 1, characterized in that: If the user type of the target user is the first user type, the step of determining the target work outlet to which the target user belongs according to the historical work data of the target user, and analyzing the cargo receipt volume corresponding to each user of the target work outlet within a preset period of time to obtain the predicted cargo receipt volume of the target user includes: If the user type of the target user is the first user type, determining the target work network point to which the target user belongs according to the historical work data of the target user; Obtaining the cargo receiving volume corresponding to each user at the target work point within a preset time period, and arranging each cargo receiving volume in ascending order to obtain a cargo receiving volume sequence; Determine a preset percentile of the cargo receipt volume in the cargo receipt volume sequence as a first target receipt volume; The first target receipt volume is analyzed based on a preset Poisson distribution function to obtain a predicted receipt volume of goods for the target user.

3. The material demand forecasting method according to claim 1, characterized in that: If the user type of the target user is the second user type, the step of acquiring characteristic information of the target user according to the historical work data of the target user, and analyzing the characteristic information to obtain the predicted receipt volume of goods of the target user includes: If the user type of the target user is the second user type, acquiring characteristic information of the target user according to the historical work data of the target user, the characteristic information including at least one of the following: basic characteristics, personal mail volume characteristics, branch mail volume characteristics, and district mail volume characteristics; Analyze and predict the characteristic information to obtain the predicted number of mails received by the target user; If the predicted number of received items is less than or equal to the number of received items of goods, determining the predicted number of received items as the predicted number of received items of goods; If the predicted receipt quantity is greater than the cargo receipt quantity, the cargo receipt quantity is determined as the cargo predicted receipt quantity.

4. The material demand forecasting method according to claim 1, characterized in that: If the user type of the target user is the third user type, the step of determining the target work outlet to which the target user belongs according to the historical work data of the target user, and analyzing the cargo receipt quantity and the cargo receipt quantity of each user corresponding to the target work outlet within a preset time period to obtain the predicted cargo receipt quantity of the target user includes: If the user type of the target user is the third user type, determining the target work network point to which the target user belongs according to the historical work data of the target user; Obtaining the cargo receiving volume corresponding to each user at the target work point within a preset time period, and arranging each cargo receiving volume in ascending order to obtain a cargo receiving volume sequence; Determine the cargo receipt volume of the first preset percentile in the cargo receipt volume sequence as the second target receipt volume, and determine the cargo receipt volume of the second preset percentile in the cargo receipt volume sequence as the third target receipt volume; The predicted cargo receipt quantity of the target user is obtained according to the second target receipt quantity, the third target receipt quantity and the cargo receipt quantity, wherein the first preset percentile is less than the second preset percentile.

5. The material demand forecasting method according to claim 4, characterized in that: The step of obtaining the predicted cargo receiving quantity of the target user according to the second target receiving quantity, the third target receiving quantity and the cargo receiving quantity comprises: If the number of received goods is less than the second target number of received goods, acquiring feature information in the historical work data, and analyzing the feature information based on the first classifier to obtain the predicted number of received goods of the target user; If the number of received goods is greater than or equal to the second target number of received goods, and less than or equal to the third target number of received goods, then acquiring feature information in the historical work data, and analyzing the feature information based on the second classifier to obtain the predicted number of received goods of the target user; If the number of cargo receipts is less than the third target receipt quantity, characteristic information in the historical work data is obtained, and the characteristic information is analyzed based on the third classifier to obtain the predicted cargo receipt quantity of the target user, wherein the first classifier, the second classifier and the third classifier are classifiers trained by different training data.

6. The material demand forecasting method according to claim 1, characterized in that: The step of obtaining the predicted material demand of the target user according to the predicted cargo receipt quantity, the cargo receipt quantity and the material usage quantity comprises: According to the material usage quantity, the material quantity on hand, the material quantity in transit and the material consumption rate of each material are obtained, wherein the material consumption rate is the quotient of the material usage quantity and the cargo receipt quantity, and the material usage quantity includes the material quantity on hand and the material quantity in transit; Obtain the product of the material consumption rate and the predicted cargo receipt quantity, and obtain the sum of the material quantity on hand and the material quantity in transit; The difference between the product value and the sum value is obtained to obtain the predicted material demand of the target user for each material.

7. The material demand forecasting method according to claim 6, characterized in that: After obtaining the material quantity on hand, the material quantity in transit and the material consumption rate of each material according to the material usage quantity, the method further includes: If the material consumption rate is greater than or equal to a preset consumption rate threshold, the value of the material consumption rate is determined as the consumption rate threshold; If the material consumption rate is less than the consumption rate threshold, it is determined that the value of the material consumption rate remains unchanged.

8. The material demand forecasting method according to claim 1, characterized in that: After obtaining the predicted material demand of the target user according to the predicted cargo receipt quantity, the cargo receipt quantity and the material usage quantity, the method further includes: Determining the material type corresponding to the predicted material demand; Get the minimum packaging quantity corresponding to the material type; Determine the pending demand quantity corresponding to the predicted material demand quantity according to the quotient surplus value between the predicted material demand quantity and the minimum packaging quantity; The pending demand quantity is sent to the terminal to receive demand quantity confirmation information fed back by the target user through the terminal.

9. The material demand forecasting method according to claim 8, characterized in that: After sending the pending demand to the terminal, the method further includes: If the demand quantity confirmation information fed back by the target user through the terminal is received, the undetermined demand quantity is determined as the to-be-delivered demand quantity, and the total replenishment quantity of each to-be-delivered material is obtained according to the to-be-delivered demand quantity; Determine the replenishment frequency of each of the materials to be dispatched according to the total replenishment amount of each of the materials to be dispatched; When the replenishment frequency includes at least two replenishment frequencies, the replenishment quantity corresponding to each replenishment frequency is determined according to the demand quantity to be shipped, the replenishment frequency and the minimum packaging quantity.

10. The material demand forecasting method according to claim 9, characterized in that: When the replenishment frequency includes at least two replenishment frequencies, the step of determining the replenishment quantity of each replenishment frequency according to the demand quantity to be shipped, the replenishment frequency and the minimum packaging quantity includes: When the replenishment frequency includes at least two replenishment frequencies, obtaining a quotient between the to-be-shipped demand quantity and the replenishment frequency; Based on the minimum packaging quantity, the quotient is rounded to determine the replenishment quantity corresponding to each replenishment frequency.

11. A material demand forecasting device, characterized in that: The device comprises: A data acquisition module is used to acquire the historical work data of the target user within a preset time period, wherein the historical work data includes work time information, the number of goods received, and the number of materials used; A type determination module, used to determine the user type of the target user according to the working time information in the historical working data; A data analysis module, configured to analyze the historical work data based on the user type of the target user to obtain the predicted receipt volume of goods of the target user; A demand determination module, used to obtain the predicted material demand of the target user according to the predicted cargo receipt quantity, the cargo receipt quantity and the material usage quantity; The step of determining the user type of the target user according to the working time information in the historical working data includes: If the working time information is less than a preset first time threshold, determining that the user type of the target user is a first user type; If the working time information is greater than or equal to the first time threshold and less than or equal to a preset second time threshold, determining that the user type of the target user is the second user type; If the working time information is greater than the second time threshold, determining that the user type of the target user is a third user type, wherein the first time threshold is less than the second time threshold; The step of analyzing the historical work data based on the user type of the target user to obtain the predicted receipt volume of goods for the target user includes: If the user type of the target user is the first user type, determining the target work outlet to which the target user belongs according to the historical work data of the target user, and analyzing the cargo receipt volume corresponding to each user of the target work outlet within a preset time period to obtain the predicted cargo receipt volume of the target user; If the user type of the target user is the second user type, acquiring characteristic information of the target user according to the historical work data of the target user, and analyzing the characteristic information to obtain the predicted receipt volume of goods of the target user; If the user type of the target user is the third user type, the target work point to which the target user belongs is determined based on the historical work data of the target user, and the number of cargo receipts and the cargo receipt volume corresponding to each user at the target work point within a preset time period are analyzed to obtain the predicted cargo receipt volume of the target user.

12. A computer device, characterized in that: The computer device comprises: one or more processors; Memory; and One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to implement the material demand forecasting method according to any one of claims 1 to 10.

13. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and the computer program is loaded by a processor to execute the steps in the material demand forecasting method according to any one of claims 1 to 10.

Images