CN114202130A

CN114202130A - Flow transfer amount prediction multitask model generation method, scheduling method, device and equipment

Info

Publication number: CN114202130A
Application number: CN202210127957.7A
Authority: CN
Inventors: 刘葳; 庄晓天; 王忠帅
Original assignee: Beijing Jingdong Zhenshi Information Technology Co Ltd
Current assignee: Beijing Jingdong Zhenshi Information Technology Co Ltd
Priority date: 2022-02-11
Filing date: 2022-02-11
Publication date: 2022-03-18

Abstract

The embodiment of the disclosure discloses a method for generating a traffic prediction multitask model, a method for scheduling, a device and equipment. One embodiment of the method comprises: acquiring a historical order information set and a historical value reduction information set of a target article in a target historical time period; based on each order date and historical value reduction information set included in the historical order information set, performing characteristic processing on the historical order information set and the historical value reduction information set to obtain a processed historical order information set serving as a sample historical order information set; and generating a flow amount prediction multitask model according to the preset loss function and each value reduction flow amount characteristic, non-value reduction flow amount characteristic and value reduction characteristic which are included in the sample historical order information set. The method and the device improve the accuracy of the traffic prediction result and reduce the waste of transportation resources.

Description

Flow transfer amount prediction multitask model generation method, scheduling method, device and equipment

Technical Field

The embodiment of the disclosure relates to the technical field of computers, in particular to a method, a device and equipment for generating a traffic prediction multitask model.

Background

The value reduction is to reduce the value paid by the user when the articles are circulated so as to achieve a preset target (for example, to improve the circulation amount of the articles). At present, before the value of the goods is reduced, a generally adopted method for predicting the flow rate is as follows: the baseline predicted amount of the runoff corresponding to the point in time of the value reduction is replaced with the amount of the runoff for which the value of the predicted payment has been subjected to the value reduction.

However, when the amount of the flow is predicted in the above manner, there are often the following technical problems: only predicting the value-subtracted traffic of the paid value cannot simultaneously acquire the value-subtracted traffic during the value subtraction and the non-value-subtracted traffic affected by the value subtraction, and the influence on the traffic at the previous and subsequent time periods during the value subtraction cannot be considered, so that the accuracy of the traffic prediction result is low, the number of times that the quantity of articles scheduled according to the prediction result cannot meet the demand is large, the number of times that articles need to be scheduled is large, and the waste of transportation resources is caused.

Disclosure of Invention

This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description. This summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.

Some embodiments of the present disclosure propose a method of generating a traffic prediction multitask model, a scheduling method, an apparatus, an electronic device and a computer-readable medium to solve one or more of the technical problems mentioned in the background section above.

In a first aspect, some embodiments of the present disclosure provide a method for generating a traffic prediction multitask model, the method including: acquiring a historical order information set and a historical value reduction information set of a target article in a target historical time period; performing feature processing on the historical order information set and the historical value reduction information set based on each order date and the historical value reduction information set which are included in the historical order information set, and obtaining a processed historical order information set as a sample historical order information set, wherein the sample historical order information in the sample historical order information set comprises a value reduction traffic characteristic, a non-value reduction traffic characteristic and a value reduction characteristic; and generating a flow amount prediction multitask model according to the preset loss function and each value reduction flow amount characteristic, non-value reduction flow amount characteristic and value reduction characteristic included in the sample historical order information set.

Optionally, the performing feature processing on the historical order information set and the historical value reduction information set includes: and in response to the historical order information meeting the abnormal conditions of the flow rate existing in the historical order information set, removing the historical order information meeting the abnormal conditions of the flow rate from the historical order information set.

Optionally, the performing feature processing on the historical order information set and the historical value reduction information set further includes: responding to the historical order information meeting the reflow volume backfilling condition in the historical order information set, and performing reflow volume backfilling processing on the historical order information meeting the reflow volume backfilling condition.

Optionally, the performing feature processing on the historical order information set and the historical value reduction information set further includes: for each historical value reduction information in the historical value reduction information set, responding to that the value reduction attribute value included in the historical value reduction information is a non-numerical value type, and performing numerical value conversion processing on the value reduction attribute value to obtain a converted value reduction attribute value serving as a value reduction power value; and determining each determined value reduction force value as a value reduction characteristic of sample historical order information corresponding to historical value reduction information corresponding to the value reduction force value.

Optionally, the performing feature processing on the historical order information set and the historical value reduction information set further includes: determining the type of the value reduction time period of each historical value reduction information according to the value reduction time period included in each historical value reduction information in the historical value reduction information set; and determining each determined value reduction period type as a value reduction characteristic of the sample historical order information corresponding to the historical value reduction information corresponding to the value reduction period type.

In a second aspect, some embodiments of the present disclosure provide a traffic prediction multitask model generation device, including: an acquisition unit configured to acquire a historical order information set and a historical value reduction information set of a target item in a target historical period; the characteristic processing unit is configured to perform characteristic processing on the historical order information set and the historical value reduction information set based on each order date and the historical value reduction information set which are included in the historical order information set, and obtain a processed historical order information set as a sample historical order information set, wherein the sample historical order information in the sample historical order information set includes a value reduction traffic characteristic, a non-value reduction traffic characteristic and a value reduction characteristic; and the generating unit is configured to generate a traffic prediction multitask model according to a preset loss function and the value reduction traffic characteristic, the non-value reduction traffic characteristic and the value reduction characteristic which are included in the sample historical order information set.

Optionally, the feature processing unit of the traffic prediction multitask model generating device includes: and the rejecting unit is configured to reject the historical order information meeting the abnormal conditions of the flow rate from the historical order information set in response to the historical order information meeting the abnormal conditions of the flow rate existing in the historical order information set.

Optionally, the feature processing unit of the traffic prediction multitask model generating device further includes: and the backfilling unit is configured to respond to the historical order information meeting the reflow backfilling condition in the historical order information set, and perform reflow backfilling processing on the historical order information meeting the reflow backfilling condition.

Optionally, the feature processing unit of the traffic prediction multitask model generating device further includes: a numerical value conversion unit and a first determination unit. The value conversion unit is configured to, for each historical value reduction information in the historical value reduction information set, in response to a non-numerical value of the value reduction attribute value included in the historical value reduction information, perform a value conversion process on the value reduction attribute value to obtain a converted value reduction attribute value as a value reduction power value. The first determination unit is configured to determine each determined value reduction force value as a value reduction characteristic of the sample historical order information corresponding to the historical value reduction information corresponding to the value reduction force value.

Optionally, the feature processing unit of the traffic prediction multitask model generating device further includes: a second determination unit and a third determination unit. Wherein the second determination unit is configured to determine a value reduction period type of each historical value reduction information according to a value reduction period included in each historical value reduction information in the historical value reduction information set. The second determination unit is configured to determine each determined value reduction period type as a value reduction characteristic of the sample historical order information corresponding to the historical value reduction information corresponding to the value reduction period type.

In a third aspect, some embodiments of the present disclosure provide a scheduling method, including: in response to receiving the target value reduction information for the target item, performing the steps of: inputting target value reduction information into a traffic prediction multitask model to obtain a predicted value reduction traffic sequence and a predicted non-value reduction traffic sequence of the target object in a value reduction period included in the target value reduction information, wherein the traffic prediction multitask model is generated by adopting the method described in any one implementation manner of the first aspect; generating a value reduction simulation result based on the value reduction target information included in the predicted value reduction transfer amount sequence, the predicted non-value reduction transfer amount sequence and the target value reduction information; and executing the article scheduling operation in response to the value reduction simulation result meeting the value reduction target condition corresponding to the value reduction target information.

Optionally, the foregoing steps further include: and responding to the situation that the value reduction simulation result does not meet the value reduction target condition corresponding to the value reduction target information, and sending the abnormal prompt information corresponding to the target value reduction information to the associated display equipment.

Optionally, the foregoing steps further include: and in response to the value reduction simulation result not meeting the value reduction target condition corresponding to the value reduction target information and receiving modified target value reduction information corresponding to the target value reduction information, taking the modified target value reduction information as target value reduction information, and executing the steps again.

In a fourth aspect, some embodiments of the present disclosure provide a scheduling apparatus, the apparatus comprising: an execution unit configured to, in response to receiving the target value reduction information of the target item, execute the steps of: inputting target value reduction information into a traffic prediction multitask model to obtain a predicted value reduction traffic sequence and a predicted non-value reduction traffic sequence of the target object in a value reduction period included in the target value reduction information, wherein the traffic prediction multitask model is generated by adopting the method described in any one implementation manner of the first aspect; generating a value reduction simulation result based on the value reduction target information included in the predicted value reduction transfer amount sequence, the predicted non-value reduction transfer amount sequence and the target value reduction information; and executing the article scheduling operation in response to the value reduction simulation result meeting the value reduction target condition corresponding to the value reduction target information.

In a fifth aspect, some embodiments of the present disclosure provide an electronic device, comprising: one or more processors; a storage device having one or more programs stored thereon, which when executed by one or more processors, cause the one or more processors to implement the method described in any of the implementations of the first or third aspects.

In a sixth aspect, some embodiments of the present disclosure provide a computer readable medium having a computer program stored thereon, where the program when executed by a processor implements the method described in any of the implementations of the first or third aspect.

The above embodiments of the present disclosure have the following advantages: the traffic prediction multitask model obtained by the traffic prediction multitask model generating method of some embodiments of the disclosure can reduce waste of transportation resources. Specifically, the reasons for the waste of transportation resources are: only predicting the value-subtracted traffic of the paid value cannot simultaneously acquire the value-subtracted traffic during the value subtraction and the non-value-subtracted traffic affected by the value subtraction, and the influence on the traffic at the previous and subsequent time periods during the value subtraction cannot be considered, so that the accuracy of the traffic prediction result is low, the number of times that the quantity of articles scheduled according to the prediction result cannot meet the demand is large, the number of times that articles need to be scheduled is large, and the waste of transportation resources is caused. Based on this, the method for generating a traffic prediction multitask model according to some embodiments of the present disclosure first obtains a historical order information set and a historical value reduction information set of a target item in a target historical time period. Thus, the acquired historical order information set and the historical value reduction information set can be used as source data for generating the traffic prediction multitask model. Then, based on each order date and the historical value reduction information set included in the historical order information set, performing feature processing on the historical order information set and the historical value reduction information set to obtain a processed historical order information set serving as a sample historical order information set. The sample historical order information in the sample historical order information set comprises a value reduction traffic characteristic, a non-value reduction traffic characteristic and a value reduction characteristic. Therefore, the acquired source data can be subjected to feature processing, so that the processed source data can be used as sample data for generating the traffic flow prediction multitask model. And finally, generating a flow prediction multitask model according to a preset loss function and each value reduction flow characteristic, non-value reduction flow characteristic and value reduction characteristic included in the sample historical order information set. Thus, the preset loss function may enhance the attention of the traffic prediction multitask model to non-value-reducing traffic of the target item in the time periods before and after the value-reducing period. Also, because the generated traffic prediction multitask model is a multitask model, it is possible to simultaneously predict a value-reducing traffic during a value reduction period and a non-value-reducing traffic affected by the value reduction. And the accuracy of the traffic prediction result is improved, and the times that the quantity of the articles scheduled according to the prediction result cannot meet the demand are reduced. Therefore, the times of article scheduling are reduced, and the waste of transportation resources is reduced.

Drawings

The above and other features, advantages and aspects of various embodiments of the present disclosure will become more apparent by referring to the following detailed description when taken in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numbers refer to the same or similar elements. It should be understood that the drawings are schematic and that elements and elements are not necessarily drawn to scale.

FIG. 1 is a schematic diagram of one application scenario of a method of traffic prediction multitask model generation according to some embodiments of the present disclosure;

FIG. 2 is a schematic diagram of one application scenario of a scheduling method according to some embodiments of the present disclosure;

FIG. 3 is a flow diagram of some embodiments of a method of generating a traffic prediction multitask model according to the present disclosure;

fig. 4 is a flow diagram of some embodiments of a scheduling method according to the present disclosure;

FIG. 5 is a block diagram of some embodiments of a traffic prediction multitask model generating device according to the present disclosure;

fig. 6 is a schematic structural diagram of some embodiments of a scheduling apparatus according to the present disclosure;

FIG. 7 is a schematic structural diagram of an electronic device suitable for use in implementing some embodiments of the present disclosure.

Detailed Description

Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the present disclosure are shown in the drawings, it is to be understood that the disclosure may be embodied in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the disclosure are for illustration purposes only and are not intended to limit the scope of the disclosure.

It should be noted that, for convenience of description, only the portions related to the related invention are shown in the drawings. The embodiments and features of the embodiments in the present disclosure may be combined with each other without conflict.

It should be noted that the terms "first", "second", and the like in the present disclosure are only used for distinguishing different devices, modules or units, and are not used for limiting the order or interdependence relationship of the functions performed by the devices, modules or units.

It is noted that references to "a", "an", and "the" modifications in this disclosure are intended to be illustrative rather than limiting, and that those skilled in the art will recognize that "one or more" may be used unless the context clearly dictates otherwise.

The names of messages or information exchanged between devices in the embodiments of the present disclosure are for illustrative purposes only, and are not intended to limit the scope of the messages or information.

The present disclosure will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.

Fig. 1 is a schematic diagram of an application scenario of a method of generating a traffic prediction multitask model according to some embodiments of the present disclosure.

In the application scenario of fig. 1, first, the computing device 101 may obtain a historical order information set 102 and a historical value reduction information set 103 for the target item over a target historical period. Then, the computing device 101 may perform feature processing on the historical order information set 102 and the historical value reduction information set 103 based on each order date included in the historical order information set 102 and the historical value reduction information set 103, and obtain a processed historical order information set as a sample historical order information set 104. The sample historical order information in the sample historical order information set 104 includes a value reduction traffic characteristic, a non-value reduction traffic characteristic, and a value reduction characteristic. Finally, the computing device 101 may generate the traffic prediction multitask model 106 according to the preset loss function 105 and the value reduction characteristics, the non-value reduction traffic characteristics and the value reduction characteristics included in the sample historical order information set 104.

The computing device 101 may be hardware or software. When the computing device is hardware, it may be implemented as a distributed cluster composed of multiple servers or terminal devices, or may be implemented as a single server or a single terminal device. When the computing device is embodied as software, it may be installed in the hardware devices enumerated above. It may be implemented, for example, as multiple software or software modules to provide distributed services, or as a single software or software module. And is not particularly limited herein.

It should be understood that the number of computing devices in FIG. 1 is merely illustrative. There may be any number of computing devices, as implementation needs dictate.

Fig. 2 is a schematic diagram of one application scenario of a scheduling method according to some embodiments of the present disclosure.

In the application scenario of fig. 2, the computing device 201 may, in response to receiving the target value reduction information 202 for the target item, perform the following steps: first, the target value reduction information 202 is input into the traffic prediction multitask model 203, and a predicted value reduction traffic sequence 204 and a predicted non-value reduction traffic sequence 205 of the target item within the value reduction period 2021 included in the target value reduction information 202 are obtained. The traffic prediction multitask model 203 may be the traffic prediction multitask model 106 in fig. 1. Second, a value reduction simulation result 206 is generated based on the predicted value reduction traffic 204, the predicted non-value reduction traffic 205, and the value reduction target information 2022 included in the target value reduction information 202. Third, in response to the value reduction simulation result 206 satisfying the value reduction target condition 207 corresponding to the value reduction target information 2022, an article scheduling operation is performed. For example, the computing device 201 may control the associated dispatching vehicle 208 to perform an item dispatching operation based on the predicted value abatement slipstream sequence 204 and the predicted non-value abatement slipstream sequence 205 for the target item described above.

The computing device 201 may be hardware or software. When the computing device is hardware, it may be implemented as a distributed cluster composed of multiple servers or terminal devices, or may be implemented as a single server or a single terminal device. When the computing device is embodied as software, it may be installed in the hardware devices enumerated above. It may be implemented, for example, as multiple software or software modules to provide distributed services, or as a single software or software module. And is not particularly limited herein.

It should be understood that computing device 101 in fig. 1 and computing device 201 in fig. 2 may be the same computing device or may be different computing devices. The number of computing devices in fig. 2 is merely illustrative. There may be any number of computing devices, as implementation needs dictate.

With continued reference to fig. 3, a flow 300 of some embodiments of a method of generating a traffic prediction multitask model according to the present disclosure is shown. The method for generating the traffic prediction multitask model comprises the following steps of:

step 301, acquiring a historical order information set and a historical value reduction information set of a target article in a target historical time period.

In some embodiments, an executing entity (e.g., the computing device shown in fig. 1 or fig. 2) of the traffic prediction multitask model generating method may obtain the historical order information set and the historical value reduction information set of the target item in the target historical period from the terminal through a wired connection manner or a wireless connection manner. The target object may be a previously selected object. For example, the target item may be any item in a fast-moving article. The target historical period may be a preset historical period including each value reduction time point (promotion time) and each non-value reduction time point (non-promotion time). The historical order information in the historical order information set may be order information corresponding to the target item submitted by the user terminal within the target historical time period. The historical order information may include, but is not limited to: order date, flow volume, value reduction identification. The order date may be a date when the user terminal submits order information corresponding to the target item. The circulation amount may be a circulation amount (sales amount) of the target item. The value reduction identification can represent whether the value of the target item paid by the user is reduced or not. For example, the value reduction identification may be a "value reduction" (promotion). The value reduction identification may also be a "non-value reduction" (non-promotion). The historical value reduction information in the historical value reduction information set may be information for reducing the value of the target item in the historical period. The historical value reduction information may include, but is not limited to: a value reduction type, a value reduction attribute value, a value reduction period, and value reduction target information. The value reducing type may be a type that reduces the value of the target item. For example, the type of value reduction described above may be "discounted". The type of value reduction described above may be "full reduction". The type of value reduction described above may be "bonus". The value-reduction attribute value may be a degree of reduction in the value of the target item. For example, the value-diminishing attribute value may be "0.8", which characterizes a value-eighting fold on the target item. The value-diminishing attribute value may be "full 200 minus 20". The value reducing period may be a period of time in which the value of the target item is reduced. It is understood that the value-diminishing period is in the target history period. The value reduction target information may be information related to a target to be achieved for value reduction of the target item. For example, the value reduction target information may be, but is not limited to: cleaning up inventory, rushing to flow volume (e.g., rushing to sell volume), maximizing value capture (e.g., maximizing revenue). In practice, the executing entity may obtain, from the terminal, each historical order information of the order date within the target historical period, to obtain a historical order information set, and obtain, from the terminal, each historical value reduction information of the value reduction period within the target historical period, to obtain a historical value reduction information set.

It should be noted that the wireless connection means may include, but is not limited to, a 3G/4G connection, a WiFi connection, a bluetooth connection, a WiMAX connection, a Zigbee connection, a uwb (ultra wideband) connection, and other wireless connection means now known or developed in the future.

Step 302, based on each order date and historical value reduction information set included in the historical order information set, performing feature processing on the historical order information set and the historical value reduction information set to obtain a processed historical order information set as a sample historical order information set.

In some embodiments, the execution subject may perform feature processing on the historical order information set and the historical value reduction information set based on each order date included in the historical order information set and the historical value reduction information set, and obtain a processed historical order information set as a sample historical order information set. The sample historical order information in the sample historical order information set comprises a value reduction traffic characteristic, a non-value reduction traffic characteristic and a value reduction characteristic. In practice, first, the execution subject may combine the historical order information sets corresponding to the same order date to obtain a historical order information set. Then, for each historical order information group in the historical order information group set, the executing body may perform the following steps:

in a first step, the sum of the amounts of the flows included in the historical order information identifying each value reduction included in the set of historical order information as a "value reduction" (e.g., a promotion) is determined as a value reduction flow.

Second, the sum of the amounts of the streamings included in the historical order information identifying each value reduction included in the historical order information set as a "non-value reduction" (e.g., non-promotion) is determined as a non-value reduction streamlining amount.

And thirdly, in response to the fact that the order date corresponding to the historical order information set is determined to be in the value reduction period included by the historical value reduction information in the historical value reduction information set, combining the value reduction type, the value reduction attribute value and the value reduction period included by the historical value reduction information, and the value reduction traffic volume, the non-value reduction traffic volume, the order date and the value reduction day representing the 'value reduction day' (for example, the value reduction day may be a holiday) into the sample historical order information. The value reduction day identifier representing the "value reduction day" may be 1.

And fourthly, in response to the fact that the order date corresponding to the historical order information group is not in the value reduction period included in the historical value reduction information set, combining the value reduction type with the attribute value being null value, the value reduction attribute value and the value reduction period, the non-value reduction traffic, the order date and the value reduction day mark representing the 'non-value reduction day' into sample historical order information. The value reduction day designation above that characterizes the "non-value reduction day" may be 0.

The value reduction traffic included in each sample historical order information in the obtained sample historical order information set is a value reduction traffic characteristic, the non-value reduction traffic included in the sample historical order information is a non-value reduction traffic characteristic, and the value reduction type, the value reduction attribute value and the value reduction time period included in the sample historical order information are value reduction characteristics.

Optionally, the execution main body may remove, from the historical order information set, historical order information that satisfies the abnormal conditions of the amount of circulation in response to the historical order information that satisfies the abnormal conditions of the amount of circulation existing in the historical order information set. The abnormal condition of the amount of circulation may be "the amount of circulation included in the historical order information is greater than or equal to a preset threshold". Here, the specific setting of the preset threshold is not limited. Therefore, historical order information with abnormal traffic volume can be eliminated, and the influence of the historical order information with abnormal traffic volume on the generated model is reduced.

Optionally, the execution subject may perform a drift backfill process on the historical order information satisfying the drift backfill condition in response to the historical order information satisfying the drift backfill condition existing in the historical order information set. The above-mentioned drift volume backfill condition may be "the stock of the target item is less than or equal to the target threshold value within the order date included in the historical order information". The target threshold may be a product of a sum of respective traffic amounts of the target item and a preset coefficient in a preset historical time period before the order date. For example, the preset historical time period may be a time period from a date 28 days before the order date to the order date. The preset coefficient may be a preset coefficient. For example, the preset coefficient may be 0.5. In practice, the execution subject may replace the circulation amount included in the historical order information with an average value of the circulation amounts of the target item in a preset historical time period before the order date. Therefore, the sales volume backfill processing can be carried out on the historical order information corresponding to the circulation volume influenced by the insufficient inventory, and the influence of the historical order information influenced by the insufficient inventory on the generated model is reduced.

Alternatively, first, for each historical value reduction information in the historical value reduction information set, the execution subject may perform a numerical conversion process on the value reduction attribute value in response to the value reduction attribute value included in the historical value reduction information being a non-numerical value, to obtain a converted value reduction attribute value as the value reduction capability value. In practice, the execution subject may determine, as the value reduction capability value of the historical value reduction information, a ratio of a reduction value included in the value reduction attribute value included in the historical value reduction information to a full limit value in response to the type of the value reduction included in the historical value reduction information being "full reduction". For example, the value-diminishing attribute value may be "200 full minus 20, the minus value" 20 ", and the full value" 200 ". In practice, the execution subject may further determine, as the value reduction capability value of the historical value reduction information, a ratio of a bonus value included in the value reduction attribute value included in the historical value reduction information to a full value in response to the type of value reduction included in the historical value reduction information being "full bonus". For example, the value-diminishing attribute value may be "20 items are gifted at 200 dollars," 20 value is gifted, "200 value is full. Then, the execution subject may determine each determined value reduction degree value as a value reduction characteristic of the sample historical order information corresponding to the historical value reduction information corresponding to the value reduction degree value. Thus, a non-numeric-type value-reducing attribute value can be converted into a continuous numeric-type value-reducing attribute value.

Alternatively, first, the execution subject may determine a value reduction period type of each historical value reduction information according to a value reduction period included in each historical value reduction information in the historical value reduction information set. In practice, first, the execution subject may sort the value reduction periods included in the historical value reduction information in an ascending order to obtain a value reduction period sequence. Then, the value reduction period sequences may be divided into value reduction period group sequences according to a preset percentile array, such that each value reduction period group belongs to one value reduction period type. For example, the preset percentile array may be: 40%, 70% and 100%. The execution subject may determine the first 40% of the value reduction periods in the sequence of value reduction periods as a short-term class of value reduction period types. The execution subject may determine 40% -70% of the value abatement periods in the sequence of value abatement periods as a medium class of value abatement period types. The execution subject may determine a value reduction period after 70% of the sequence of value reduction periods as a long-term class of value reduction period types. And finally, determining each determined value reduction period type as the value reduction characteristic of the sample historical order information corresponding to the historical value reduction information corresponding to the value reduction period type. Thus, each value reduction period may be divided into different value reduction period types.

And 303, generating a flow prediction multitask model according to a preset loss function and each value reduction flow characteristic, non-value reduction flow characteristic and value reduction characteristic included in the sample historical order information set.

In some embodiments, the execution subject may generate the traffic prediction multitask model according to a preset loss function and each of the value-cut traffic characteristics, the non-value-cut traffic characteristics and the value-cut characteristics included in the sample historical order information set. Wherein the predetermined loss function may be L_MTL＝w₁×L₁×is_promo+w₂×L₂. Wherein L is_MTLRepresenting a preset loss function. w is a₁A weight parameter representing a value reduction traffic prediction task. L is₁Representing a loss value for the value reduction traffic prediction task. is _ promo represents the value reduction day flag, and is 0 or 1. w is a₂A weight parameter representing a non-value-reducing traffic prediction task. L is₂Representing the loss value of the non-value-reducing traffic prediction task. In practice, the execution subject may input the sample historical order information set to a multitask learning model using the preset loss function as a loss function, and train to obtain a traffic prediction multitask model. For example, the multi-task learning model may be a model implemented based on LSTM (long-short term memory model). The flow prediction multitask model is used for generating a predicted value reduction flow sequence and a predicted non-value reduction flow sequence of the target object according to the value reduction information.

With further reference to fig. 4, a flow 400 of further embodiments of a scheduling method is illustrated. The process 400 of the scheduling method includes the following steps:

step 401, in response to receiving the target value reduction information of the target item, performing the following steps:

step 4011, importing the target value reduction information into a traffic prediction multitask model, and obtaining a predicted value reduction traffic sequence and a predicted non-value reduction traffic sequence of the target object within a value reduction period included in the target value reduction information.

In some embodiments, an executing entity (e.g., a computing device shown in fig. 1 or fig. 2) of the scheduling method may stream the target value reduction information into the traffic prediction multitask model to obtain the predicted value reduction traffic and the predicted non-value reduction traffic of the target item. The above-mentioned traffic prediction multitask model is generated by the method in the embodiments corresponding to fig. 3. The target value reduction information may be currently received value reduction information. The predicted value abatement traffic in the predicted value abatement traffic sequence may be traffic subjected to value abatement at a future time point of the target item, which is predicted by the traffic prediction multitask model. The predicted non-value-reducing traffic in the sequence of predicted non-value-reducing traffic may be traffic that is predicted by the traffic prediction multitask model and that is not subjected to value reduction at a future time point by the target object. The future point in time is within the value reduction period. Therefore, the value reduction traffic and the non-value reduction traffic within the value reduction period can be predicted according to the current value reduction information and the previously generated traffic prediction multitask model.

Step 4012, generating a value reduction simulation result based on the value reduction target information included in the predicted value reduction traffic sequence, the predicted non-value reduction traffic sequence, and the target value reduction information.

In some embodiments, the execution subject may generate the value reduction simulation result based on the value reduction target information included in the predicted value reduction traffic sequence, the predicted non-value reduction traffic sequence, and the target value reduction information. The value reduction target information may be, but is not limited to: cleaning the stock, flushing flow and transferring quantity, and obtaining the maximum value. In practice, the executing body may determine the target future point in time as a value reduction simulation result in response to the value reduction target information being a clean stock and the sum of each predicted value reduction transfer and each predicted non-value reduction transfer before the target future point in time in the predicted value reduction transfer sequence and the predicted non-value reduction transfer sequence being equal to or greater than the stock amount of the target item. The target future point in time may be any future point in time within the value reduction period. The execution subject may further determine, as the value reduction simulation result, a ratio of each of the predicted value reduction transitions to the sum of each of the predicted non-value reduction transitions and the stock amount of the target item in response to the value reduction target information being clear stock and the sum of each of the predicted value reduction transitions in the predicted value reduction transition series and each of the predicted non-value reduction transitions in the predicted non-value reduction transition series being smaller than the stock amount of the target item.

The execution main body may further determine, as the value reduction simulation result, a sum of each predicted value-reduction transfer amount in the predicted value-reduction transfer amount sequence and each predicted non-value-reduction transfer amount in the predicted non-value-reduction transfer amount sequence, in response to the value reduction target information being the rush transfer amount.

The execution subject may further determine, as the value reduction simulation result, a sum of an acquired value (e.g., revenue) corresponding to each predicted value reduction transfer in the sequence of predicted value reduction transfers and an acquired value corresponding to each predicted non-value reduction transfer in the sequence of predicted non-value reduction transfers in response to the value reduction target information being the maximum value acquisition. Thus, a corresponding value reduction simulation result can be generated based on the value reduction target information set in advance.

And 4013, in response to the result of the value reduction simulation meeting the value reduction target condition corresponding to the value reduction target information, executing an article scheduling operation.

In some embodiments, the executing agent may execute the item scheduling operation in response to the value reduction simulation result satisfying a value reduction target condition corresponding to the value reduction target information. The value reduction target condition may be a preset condition corresponding to the value reduction target information. For example, when the value reduction target information is a clean stock, the value reduction target condition may be "the value reduction simulation result is a time point type". When the value reduction target information is the rush current transfer amount, the value reduction target condition may be that "the value reduction simulation result is equal to or greater than a preset rush current transfer amount". When the value reduction target information is obtained for the maximum value, the value reduction target condition may be "the value reduction simulation result is equal to or greater than the preset obtaining value". In practice, the executing entity may control the associated dispatching vehicle to execute the item dispatching operation according to the predicted value-reducing traffic and the predicted non-value-reducing traffic of the target item. The dispatching vehicle may be a vehicle for dispatching an item. For example, the dispatching vehicle may be an unmanned vehicle. Specifically, first, the sum of each predicted value-subtracted flow included in the above-described predicted value-subtracted flow sequence and each predicted non-value-subtracted flow included in the above-described predicted non-value-subtracted flow sequence may be determined as a predicted flow. Then, a difference between the predicted diversion amount and the stock amount of the target item may be determined as a modulation amount. Finally, the dispatching vehicle can be controlled to dispatch the dispatching quantity of the target object. Thus, item scheduling can be performed when the predicted value-reducing and non-value-reducing traffic within the value-reducing period satisfies a condition.

Optionally, the executing body may further send, in response to that the value reduction simulation result does not satisfy the value reduction target condition corresponding to the value reduction target information, the abnormality prompting information corresponding to the target value reduction information to the associated display device. The abnormal prompt information may be information that prompts that the value reduction target information cannot satisfy the value reduction target condition corresponding to the value reduction target information. For example, the abnormality presentation information may be "the value reduction information is reset to the value reduction information when the set target cannot be reached". Thus, when the target value reduction information does not satisfy the value reduction target condition corresponding to the value reduction target information, relevant persons (for example, merchants) can be prompted to reset the value reduction information.

Optionally, the executing body may further execute the step again by taking the modified target value reduction information as the target value reduction information in response to that the value reduction simulation result does not satisfy the value reduction target condition corresponding to the value reduction target information and that the modified target value reduction information corresponding to the target value reduction information is received. The modified target value reduction information may be the value reduction information resubmitted by the terminal (e.g., the merchant terminal). Thus, the above steps may be performed again after the relevant person (e.g., the merchant) resets the value reduction information.

The above embodiments of the present disclosure have the following advantages: by the scheduling method of some embodiments of the present disclosure, waste of transportation resources may be reduced. Specifically, the reasons for the waste of transportation resources are: the conventional traffic prediction model only predicts the traffic of paid value reduced by value, cannot simultaneously acquire the value reduced traffic and non-value reduced traffic influenced by the value reduction during the period of value reduction, cannot consider the influence on the traffic in the period before and after the period of value reduction, and causes lower accuracy of a traffic prediction result, so that more times that the quantity of articles scheduled according to the prediction result cannot meet the demand are caused, and waste of transportation resources is caused. Based on this, the scheduling method of some embodiments of the present disclosure predicts a predicted value-subtracted traffic sequence and a predicted non-value-subtracted traffic sequence of the target item within the value-subtracted period included in the target value-subtracted information through the traffic prediction multitask model generated by those embodiments corresponding to fig. 3. Then, a value reduction simulation result is generated based on the value reduction target information included in the predicted value reduction traffic sequence, the predicted non-value reduction traffic sequence, and the target value reduction information. Thus, a corresponding value reduction simulation result can be generated based on the value reduction target information set in advance. And then, responding to the value reduction simulation result meeting the value reduction target condition corresponding to the value reduction target information, and executing the article scheduling operation. Thus, item scheduling can be performed when the predicted value-reducing and non-value-reducing traffic within the value-reducing period satisfies a condition. And further, the simulation of the value reduction target is realized according to the set value reduction information. Also, because the generated traffic prediction multitask model is a multitask model, it is possible to simultaneously predict a value-reducing traffic during a value reduction period and a non-value-reducing traffic affected by the value reduction. And the accuracy of the traffic prediction result is improved, and the times that the quantity of the articles scheduled according to the prediction result cannot meet the demand are reduced. Thereby reducing the waste of transportation resources.

With further reference to fig. 5, as an implementation of the methods illustrated in the above figures, the present disclosure provides some embodiments of a traffic prediction multitask model generating device, which correspond to those of the method embodiments illustrated in fig. 3, and which may be applied in various electronic devices in particular.

As shown in fig. 5, the traffic prediction multitask model generating device 500 of some embodiments includes: an acquisition unit 501, a feature processing unit 502, and a generation unit 503. Wherein the obtaining unit 501 is configured to obtain a historical order information set and a historical value reduction information set of the target item in a target historical period; the feature processing unit 501 is configured to perform feature processing on the historical order information set and the historical value reduction information set based on each order date included in the historical order information set and the historical value reduction information set, and obtain a processed historical order information set as a sample historical order information set, where sample historical order information in the sample historical order information set includes a value reduction traffic characteristic, a non-value reduction traffic characteristic, and a value reduction characteristic; the generating unit 503 is configured to generate a traffic prediction multitask model according to a preset loss function and each value reduction traffic characteristic, non-value reduction traffic characteristic and value reduction characteristic included in the sample historical order information set.

In an optional implementation manner of some embodiments, the feature processing unit 502 of the traffic prediction multitask model generating device 500 may include: and a removing unit (not shown in the figure) configured to remove the historical order information meeting the abnormal conditions of the flow rate from the historical order information set in response to the historical order information meeting the abnormal conditions of the flow rate existing in the historical order information set.

In an optional implementation manner of some embodiments, the feature processing unit 502 of the traffic prediction multitask model generating device 500 may further include: and a backfill unit (not shown in the figure) configured to perform the flow back filling processing on the historical order information meeting the flow back filling condition in response to the historical order information meeting the flow back filling condition existing in the historical order information set.

In an optional implementation manner of some embodiments, the feature processing unit 502 of the traffic prediction multitask model generating device 500 may further include: a numerical value conversion unit and a first determination unit (not shown in the figure). The value conversion unit is configured to, for each historical value reduction information in the historical value reduction information set, in response to a non-numerical value of the value reduction attribute value included in the historical value reduction information, perform a value conversion process on the value reduction attribute value to obtain a converted value reduction attribute value as a value reduction power value. The first determination unit is configured to determine each determined value reduction force value as a value reduction characteristic of the sample historical order information corresponding to the historical value reduction information corresponding to the value reduction force value.

In an optional implementation manner of some embodiments, the feature processing unit 502 of the traffic prediction multitask model generating device 500 may further include: a second determination unit and a third determination unit (not shown in the figure). Wherein the second determination unit is configured to determine a value reduction period type of each historical value reduction information according to a value reduction period included in each historical value reduction information in the historical value reduction information set. The second determination unit is configured to determine each determined value reduction period type as a value reduction characteristic of the sample historical order information corresponding to the historical value reduction information corresponding to the value reduction period type.

It will be understood that the elements described in the apparatus 500 correspond to various steps in the method described with reference to fig. 3. Thus, the operations, features and resulting advantages described above with respect to the method are also applicable to the apparatus 500 and the units included therein, and are not described herein again.

With further reference to fig. 6, as an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of a scheduling apparatus, which correspond to those shown in fig. 4, and which may be applied in various electronic devices.

As shown in fig. 6, the scheduling apparatus 600 of some embodiments includes: an execution unit 601 configured to, in response to receiving the target value reduction information of the target item, perform the following steps: inputting target value reduction information into a flow prediction multitask model to obtain a prediction value reduction flow sequence and a prediction non-value reduction flow sequence of the target object in a value reduction period included in the target value reduction information; generating a value reduction simulation result based on the value reduction target information included in the predicted value reduction transfer amount sequence, the predicted non-value reduction transfer amount sequence and the target value reduction information; and executing the article scheduling operation in response to the value reduction simulation result meeting the value reduction target condition corresponding to the value reduction target information.

In an optional implementation manner of some embodiments, the foregoing step may further include: and responding to the situation that the value reduction simulation result does not meet the value reduction target condition corresponding to the value reduction target information, and sending the abnormal prompt information corresponding to the target value reduction information to the associated display equipment.

In an optional implementation manner of some embodiments, the foregoing step may further include: and in response to the value reduction simulation result not meeting the value reduction target condition corresponding to the value reduction target information and receiving modified target value reduction information corresponding to the target value reduction information, taking the modified target value reduction information as target value reduction information, and executing the steps again.

It will be understood that the elements described in the apparatus 600 correspond to various steps in the method described with reference to fig. 4. Thus, the operations, features and resulting advantages described above with respect to the method are also applicable to the apparatus 600 and the units included therein, and are not described herein again.

Referring now to FIG. 7, shown is a schematic block diagram of an electronic device (e.g., the computing device of FIG. 1 or FIG. 2) 700 suitable for use in implementing some embodiments of the present disclosure. The electronic device shown in fig. 7 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present disclosure.

As shown in fig. 7, electronic device 700 may include a processing means (e.g., central processing unit, graphics processor, etc.) 701 that may perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM)702 or a program loaded from storage 708 into a Random Access Memory (RAM) 703. In the RAM703, various programs and data necessary for the operation of the electronic apparatus 700 are also stored. The processing device 701, the ROM 702, and the RAM703 are connected to each other by a bus 704. An input/output (I/O) interface 705 is also connected to bus 704.

Generally, the following devices may be connected to the I/O interface 705: input devices 706 including, for example, a touch screen, touch pad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; an output device 707 including, for example, a Liquid Crystal Display (LCD), a speaker, a vibrator, and the like; storage 708 including, for example, magnetic tape, hard disk, etc.; and a communication device 709. The communication means 709 may allow the electronic device 700 to communicate wirelessly or by wire with other devices to exchange data. While fig. 7 illustrates an electronic device 700 having various means, it is to be understood that not all illustrated means are required to be implemented or provided. More or fewer devices may alternatively be implemented or provided. Each block shown in fig. 7 may represent one device or may represent multiple devices as desired.

In particular, according to some embodiments of the present disclosure, the processes described above with reference to the flow diagrams may be implemented as computer software programs. For example, some embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In some such embodiments, the computer program may be downloaded and installed from a network via communications means 709, or may be installed from storage 708, or may be installed from ROM 702. The computer program, when executed by the processing device 701, performs the above-described functions defined in the methods of some embodiments of the present disclosure.

It should be noted that the computer readable medium described in some embodiments of the present disclosure may be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In some embodiments of the disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In some embodiments of the present disclosure, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, optical cables, RF (radio frequency), etc., or any suitable combination of the foregoing.

In some embodiments, the clients, servers may communicate using any currently known or future developed network Protocol, such as HTTP (HyperText Transfer Protocol), and may interconnect with any form or medium of digital data communication (e.g., a communications network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), the Internet (e.g., the Internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future developed network.

The computer readable medium may be embodied in the electronic device; or may exist separately without being assembled into the electronic device. The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to: acquiring a historical order information set and a historical value reduction information set of a target article in a target historical time period; performing feature processing on the historical order information set and the historical value reduction information set based on each order date and the historical value reduction information set which are included in the historical order information set, and obtaining a processed historical order information set as a sample historical order information set, wherein the sample historical order information in the sample historical order information set comprises a value reduction traffic characteristic, a non-value reduction traffic characteristic and a value reduction characteristic; and generating a flow amount prediction multitask model according to the preset loss function and each value reduction flow amount characteristic, non-value reduction flow amount characteristic and value reduction characteristic included in the sample historical order information set.

Computer program code for carrying out operations for embodiments of the present disclosure may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + +, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The units described in some embodiments of the present disclosure may be implemented by software, and may also be implemented by hardware. The described units may also be provided in a processor, and may be described as: a processor includes an acquisition unit, a feature processing unit, and a generation unit. Where the names of the units do not constitute a limitation on the units themselves in some cases, for example, the acquiring unit may also be described as a "unit that acquires a historical order information set and a historical value reduction information set of the target item for a target historical period".

The functions described herein above may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), systems on a chip (SOCs), Complex Programmable Logic Devices (CPLDs), and the like.

The foregoing description is only exemplary of the preferred embodiments of the disclosure and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the invention in the embodiments of the present disclosure is not limited to the specific combination of the above-mentioned features, but also encompasses other embodiments in which any combination of the above-mentioned features or their equivalents is made without departing from the inventive concept as defined above. For example, the above features and (but not limited to) technical features with similar functions disclosed in the embodiments of the present disclosure are mutually replaced to form the technical solution.

Claims

1. A method for generating a multi-task model for traffic prediction comprises the following steps:

acquiring a historical order information set and a historical value reduction information set of a target article in a target historical time period;

based on each order date and the historical value reduction information set which are included in the historical order information set, performing feature processing on the historical order information set and the historical value reduction information set to obtain a processed historical order information set as a sample historical order information set, wherein the sample historical order information in the sample historical order information set comprises a value reduction traffic characteristic, a non-value reduction traffic characteristic and a value reduction characteristic;

and generating a flow amount prediction multitask model according to the preset loss function and each value reduction flow amount characteristic, non-value reduction flow amount characteristic and value reduction characteristic included in the sample historical order information set.

2. The method of claim 1, wherein said performing feature processing on said historical order information set and said historical value reduction information set comprises:

and in response to the historical order information meeting the abnormal conditions of the flow rate in the historical order information set, removing the historical order information meeting the abnormal conditions of the flow rate from the historical order information set.

3. The method of claim 2, wherein said characterizing said historical order information set and said historical value reduction information set further comprises:

responding to the historical order information meeting the reflow volume backfilling condition in the historical order information set, and performing reflow volume backfilling processing on the historical order information meeting the reflow volume backfilling condition.

4. The method of claim 3, wherein said characterizing said historical order information set and said historical value reduction information set further comprises:

for each historical value reduction information in the historical value reduction information set, responding to that the value reduction attribute value included in the historical value reduction information is a non-numerical value type, and performing numerical value conversion processing on the value reduction attribute value to obtain a converted value reduction attribute value serving as a value reduction power value;

and determining each determined value reduction force value as a value reduction characteristic of sample historical order information corresponding to historical value reduction information corresponding to the value reduction force value.

5. The method of claim 4, wherein said characterizing said historical order information set and said historical value reduction information set further comprises:

determining the type of the value reduction time period of each historical value reduction information according to the value reduction time period included in each historical value reduction information in the historical value reduction information set;

and determining each determined value reduction period type as a value reduction characteristic of the sample historical order information corresponding to the historical value reduction information corresponding to the value reduction period type.

6. A method of scheduling, comprising:

in response to receiving the target value reduction information for the target item, performing the steps of:

inputting target value reduction information into a traffic prediction multitask model to obtain a predicted value reduction traffic sequence and a predicted non-value reduction traffic sequence of the target object in a value reduction period included in the target value reduction information, wherein the traffic prediction multitask model is generated by adopting the method of any one of claims 1 to 5;

generating a value reduction simulation result based on the value reduction target information included in the predicted value reduction transfer amount sequence, the predicted non-value reduction transfer amount sequence and the target value reduction information;

and executing article scheduling operation in response to the value reduction simulation result meeting the value reduction target condition corresponding to the value reduction target information.

7. The method of claim 6, wherein the steps further comprise:

and responding to the value reduction simulation result which does not meet the value reduction target condition corresponding to the value reduction target information, and sending the abnormal prompt information corresponding to the target value reduction information to the associated display equipment.

8. The method of claim 7, wherein the steps further comprise:

and in response to the value reduction simulation result not meeting the value reduction target condition corresponding to the value reduction target information and receiving modified target value reduction information corresponding to the target value reduction information, taking the modified target value reduction information as target value reduction information, and executing the step again.

9. A traffic prediction multitask model generation apparatus comprising:

an acquisition unit configured to acquire a historical order information set and a historical value reduction information set of a target item in a target historical period;

the characteristic processing unit is configured to perform characteristic processing on the historical order information set and the historical value reduction information set based on each order date and the historical value reduction information set which are included in the historical order information set, and obtain a processed historical order information set as a sample historical order information set, wherein the sample historical order information in the sample historical order information set includes a value reduction traffic characteristic, a non-value reduction traffic characteristic and a value reduction characteristic;

and the generating unit is configured to generate a traffic prediction multitask model according to a preset loss function and each value reduction traffic characteristic, non-value reduction traffic characteristic and value reduction characteristic included in the sample historical order information set.

10. A scheduling apparatus, comprising:

an execution unit configured to, in response to receiving the target value reduction information of the target item, execute the steps of:

11. An electronic device, comprising:

one or more processors;

a storage device having one or more programs stored thereon,

when executed by the one or more processors, cause the one or more processors to implement the method of any one of claims 1-5 or 6-8.

12. A computer-readable medium, on which a computer program is stored, wherein the program, when executed by a processor, implements the method of any of claims 1-5 or 6-8.