Disclosure of Invention
The invention provides a dynamic inventory allocation method based on resource flow, electronic equipment, media and program products, which are used for solving the defect of larger deviation of a predicted result in the prior art.
The invention provides a dynamic inventory allocation method based on resource flow, which is used for a dynamic inventory allocation system and comprises the following steps:
Inputting the real-time resource data stored in the inventory into a time sequence prediction model, and outputting a predicted amount of the resource data to be packaged; the time sequence prediction model is obtained by training based on historical resource sample data;
Determining a stock output rule corresponding to the resource data to be packaged according to different products corresponding to the resource data to be packaged, wherein the stock output rule is used for generating after analyzing the resource data to be packaged so as to restrict the resource data to be packaged;
and generating and outputting resource inventory data of the next period according to the predicted amount of the resource data to be packaged and the corresponding inventory output rule.
According to the inventory dynamic allocation method provided by the invention, the time sequence prediction model comprises the following steps: the system comprises an autoregressive integral moving average model and a secondary exponential smoothing model, wherein the autoregressive integral moving average model is used for autoregressive integral prediction of stable time sequence data, and the secondary exponential smoothing model is used for predicting twice exponential smoothing of the time sequence data; each model predictive value has a corresponding preset weight value;
Inputting the real-time resource data stored in the inventory into a time sequence prediction model, and outputting the predicted amount of the resource data to be packaged, wherein the method comprises the following steps:
respectively inputting the real-time resource data stored in the stock into an autoregressive integral moving average model and a secondary exponential smoothing model to respectively obtain predicted values of the two models;
and determining and outputting the predicted amount of the resource data to be packaged according to the predicted values of the two models and the preset weight values corresponding to the two predicted values.
According to the inventory dynamic allocation method provided by the invention, the training method of the time sequence prediction model comprises the following steps:
Inputting historical resource sample data with a time sequence into a time sequence prediction model to determine the order, and fitting an autoregressive integral moving average model based on the determined order to obtain a first predicted value;
calculating a primary index smooth value and a secondary index smooth value for the time sequence of the historical resource sample data, calculating a model parameter value by using the two index smooth values in the last period, bringing the model parameter value into a secondary index smooth model, and calculating a second predicted value of the secondary index smooth model;
After the first predicted value and the second predicted value are obtained, different weight values are given to different predicted values according to the relative accuracy of the first predicted value and the second predicted value, and the predicted data are compared with the acquired historical resource sample data of the next period for a plurality of times, so that model parameters are corrected.
According to the inventory dynamic allocation method provided by the invention, the inventory production rule comprises at least one of the following: the method comprises the steps of predicting resource data, sequencing of different resource data in a resource flow, correction coefficients of products corresponding to different resource data, and rule configuration and inventory proportion in different cities;
according to different products corresponding to the resource data to be packaged, determining an inventory output rule of the resource data to be packaged in a next time period, including:
Performing topology positioning in a resource stream according to different products corresponding to the resource data to be packaged, and determining the sequence of the products corresponding to the different resource data in the resource stream;
Acquiring statistical data of different resource data, performing paret analysis, calculating the duty ratio of products corresponding to the different resource data in different dimensions, determining the level of the different products, and obtaining correction coefficients of the different products; wherein, the stock stores the statistical data of different resource data in the current time period;
determining rule configuration of products corresponding to different resource data in different cities according to an input instruction of configuration information of the city rules;
and determining the inventory proportion of different products in the next time period according to the input instruction of the inventory proportion.
According to the method for dynamically allocating the inventory provided by the invention, after generating and outputting the resource inventory data of the next period, the method further comprises the following steps: and continuously adjusting parameters of the time sequence prediction model according to the quantity of the output resource inventory data and the difference value between the predicted quantity of the resource data to be packaged, so that the difference value between the predicted quantity of the resource data output by the time sequence prediction model and the quantity of the real output resource inventory data is smaller than a threshold value.
According to the inventory dynamic allocation method provided by the invention, the inventory dynamic allocation system is used in the field of house source transaction, and the resources are house source business opportunity resources.
According to the inventory dynamic allocation method provided by the invention, the resource inventory data comprises a plurality of resource inventory data packets;
after generating and outputting the next period of resource inventory data, the method further comprises:
Pushing the plurality of resource inventory data packages to different brokers, and displaying the resource inventory data packages in a buying and selling platform according to the different resource inventory data packages so as to generate the latest real-time resource data.
The invention also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the resource flow based inventory dynamic allocation method as described in any of the above when executing the program.
The present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the resource flow based inventory dynamic allocation method as described in any of the above.
The invention also provides a computer program product comprising a computer program which when executed by a processor implements the steps of a resource-flow-based inventory dynamic allocation method as described in any of the above.
According to the inventory dynamic allocation method based on resource flow, real-time resource data stored in inventory are input into a time sequence prediction model, the predicted amount of the resource data to be packaged is output, and then inventory output rules corresponding to the resource data to be packaged are determined according to different products corresponding to the resource data to be packaged, so that the resource data to be packaged is constrained, and the predicted amount of the resource data is corrected; finally, generating and outputting resource inventory data of the next period according to the predicted quantity of the resource data to be packaged and the corresponding inventory output rule, wherein compared with a method for performing inventory distribution only through historical resource data in the prior art, the prediction result is more accurate.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In addition, the inventor has found that the prior art merchandise inventory system has the following problems:
1) The applicable scenario is relatively single, and the design goal of inventory systems is generally to reduce the instances of diapause or excess inventory.
2) The effects of a diapire or overstock problem cannot be avoided.
3) The flexibility is lacking, and the scene of the linkage of the commodity inventory with wide commodity selling range is difficult to support. For example, merchandise sales contains many cities, and different cities have different sales rules; in a primary inventory estimation process, the inventory of one commodity can influence the inventory of another commodity, and the like.
By the method of the embodiment, the accuracy of the prediction result is improved.
Those skilled in the art will appreciate that the resources in the present application may be commodities for sale in a marketing scenario, or may be items or opportunities for distribution in a distribution scenario, and the present application is not limited thereto.
The embodiment builds a stock dynamic allocation system and method based on resource flow. Compared with the prior art, the method has the following improvement points:
1) Classifying the resource data from multiple dimensions, for example, classifying the resource data from multiple dimensions such as physical locations (business circles/administrative regions/cities), business lines (new houses, second houses, rentals), business types, time ranges (weeks, months) and the like according to the allocation of house source business machine resources, predicting the future inventory-packed resource data by using a quadratic exponential smoothing and autoregressive integral sliding average mixed model, comparing the predicted data with real resource data obtained by resource data recovery, and finally reducing the gap between the predicted data and the real resource data to 14%.
2) The problem of strong coupling between inventory estimation and resource allocation is solved by a top-down inventory estimation mode, and the 'bull penis effect' (namely, the principle of accelerating and amplifying the variation of the demand) is avoided.
3) Besides the scene of accurately predicting the resource inventory, the method also supports the dynamic adjustment of the resource inventory in the scene of allowing the inventory to be overstocked, avoids the problem of sale or overstock (namely, the scene that the resources have lower timeliness and can be distributed in a plurality of periods), and improves the reliability.
4) In the scenario of multi-resource inventory linkage, improvements are made on the basis of the paret analysis method (Activity Based Classification, ABC) of safety inventory calculation, not only to more strictly inventory control for class a (high value) products, but also to give a more preferential allocation according to the topological order of the resources in the resource flow. A method for dynamic allocation of inventory based on resource movement according to an embodiment of the present invention is described below with reference to fig. 1-6.
The terms related to the present embodiment are schematically explained first.
Resource data: data that can be used for sale or distribution. For example, in the field of house brokerage, business opportunity data, as a resource data, may be assigned to brokerage for use. Taking the second-hand house service as an example, if the user clicks on a house source to view or clicks on a broker head portrait to communicate, corresponding business opportunity data is generated.
Time series prediction model: and predicting the business opportunity number of the future period based on the historical sample data, and then carrying out packaging distribution based on the prediction result. Common time series prediction models such as LSTM model, autoregressive integral moving average model, etc.
Autoregressive integral moving average model (Auto-REGRESSIVE MOVING AVERAGE MODEL, ARMA): is a common time series prediction local statistical algorithm. The ARIMA model confirms the orders of AR and MA through differential operation, an autocorrelation function (ACF) and a partial autocorrelation function (PACF), and fits the ARIMA model based on the confirmed orders to obtain a prediction result.
City rule: different city rules may be set for different cities, for example, the minimum packing size of the area is set to 5, etc.
The embodiment of the invention discloses a dynamic inventory allocation method based on resource flow, which is used for a dynamic inventory allocation system, and referring to fig. 1, the method comprises the following steps:
101. Inputting the real-time resource data stored in the inventory into a time sequence prediction model, and outputting a predicted amount of the resource data to be packaged; the time sequence prediction model is obtained by training based on historical resource sample data.
The resource data is business data as an example. In this embodiment, the business data may be distributed in a platform, for example, the business data in this embodiment may be distributed to a broker in the platform, and after obtaining the business data, the broker may display the business data in a corresponding advertising space or other location capable of displaying the broker information in a property trading platform facing the user. If the user has trading intent to the house, click on broker information, such as a broker avatar or other control that can trigger communication, the user communicates with the broker for house trading.
Prediction of business opportunity data is a very critical step in this embodiment, and the more accurate the predicted business opportunity data, the more accurate the inventory prediction, and the less the impact on actual production. If the inventory forecast is small, the requirements of the brokers cannot be met; the inventory is set up in many ways and cannot be redeemed after being distributed to brokers. There are the following effects:
1) For example, 1000 business data can be generated, but 800 business data are predicted by the inventory dynamic allocation system, and 200 business data are allocated in the period (for example, a week) so as to not meet the requirements of brokers. And unallocated business data may be wasted for timeliness reasons.
2) For example, 1000 business data can be generated when inventory estimation is more, but 1200 business data are predicted by the inventory dynamic allocation system, and then 200 business data are allocated in the period (for example, a week), but if the allocated business data cannot be honored by a broker for a long time, the time cost of online work order audit of the broker is increased, and the enthusiasm of the broker is also eliminated.
Therefore, in this embodiment, the real-time resource data stored in the inventory is input into the time series prediction model to obtain the predicted amount of the resource data to be packaged, so as to obtain a more accurate predicted amount of the resource data.
Wherein the historical resource sample data is the basis of a time series prediction algorithm. High quality historical resource sample data is critical to the accuracy of the time series prediction algorithm. In the system, the historical resource sample data can be business opportunity data produced based on a resource platform, such as resource data obtained by classifying from a physical location (business district/administrative district/city), a business line (new house, second house, renting house), a business type, a time range (week, month) and other multidimensional degrees. An exemplary historical sample data classification for this embodiment is shown in FIG. 2.
Training the model through multi-dimensional resource sample data, comparing the predicted quantity of the resource data with the real resource data quantity obtained by recycling, correcting the model in real time, and finally reducing the difference between the predicted data and the real data to be within a set threshold value.
In this embodiment, the resource data to be packaged in the inventory may be stored in Hive, and read when the data is called.
102. And determining inventory output rules corresponding to the resource data to be packaged according to different products corresponding to the resource data to be packaged.
The inventory output rule is used for analyzing and generating the resource data to be packaged so as to restrict the resource data to be packaged.
Different past historical stock data are used for prediction, and new resource stock is generated from top to bottom along with resource circulation. Taking house service as an example, after the resource estimated flow is generated, the resource estimated flow is divided according to service lines, namely new houses, second hands and leases. Clustering the resource data from different dimensions, and dividing the data of the dimensions of cities, administrative areas and business circles.
There are various inventory production rules, such as resource data pre-measurement, sequence of different resource data in resource flow, correction coefficient of different resource data corresponding products, rule configuration in different cities, inventory proportion, etc. And constraining the inventory output from multiple angles, so as to constrain the resource data to be packaged, and further realize the correction of the predicted quantity of the resource data.
It should be noted that, in the present embodiment, the inventory output rule is not based on the historical resource sample data as the basis of the inventory output rule, but rather, the inventory output rule is generated based on the resource data generated in real time.
103. And generating and outputting resource inventory data of the next period according to the predicted amount of the resource data to be packaged and the corresponding inventory output rule.
In this step, the resource data pre-measurement is taken out from Hive, and inventory data of the next period is estimated according to the corresponding inventory output rule. Rules may be manually formulated, but the process of executing the rules to generate the next period of resource inventory data is automatic and dynamic.
In particular, the resource inventory data includes a plurality of resource inventory data packets. For example, a cell phone has red, green and blue models, which are 3 different resource inventory data packets.
After the resource inventory data of the next period is generated and output, the plurality of resource inventory data packages are pushed to different brokers and displayed in the buying and selling platform according to the different resource inventory data packages so as to generate the latest real-time resource data.
According to the latest real-time resource data, the index can be calculated by a correction coefficient, so that the parameters in the inventory output rule can be corrected.
And, after generating and outputting the next period of resource inventory data, the method further comprises: and continuously adjusting parameters of the time sequence prediction model according to the quantity of the output resource inventory data and the difference value between the predicted quantity of the resource data to be packaged, so that the difference value between the predicted quantity of the resource data output by the time sequence prediction model and the quantity of the real output resource inventory data is smaller than a threshold value, and further training the time sequence prediction model.
According to the inventory dynamic allocation method based on resource flow, real-time resource data stored in inventory are input into a time sequence prediction model, predicted quantity of the resource data to be packaged is output, inventory output rules corresponding to the resource data to be packaged are then determined according to different products corresponding to the resource data to be packaged, and therefore the resource data to be packaged is constrained, and correction of the predicted quantity of the resource data is achieved; finally, generating and outputting resource inventory data of the next period according to the predicted quantity of the resource data to be packaged and the corresponding inventory output rule, wherein compared with a method for performing inventory distribution only through historical resource data in the prior art, the prediction result is more accurate.
Further, the time series prediction model includes: the system comprises an autoregressive integral moving average model and a secondary exponential smoothing model, wherein the autoregressive integral moving average model is used for autoregressive integral prediction of stable time sequence data, and the secondary exponential smoothing model is used for predicting twice exponential smoothing of the time sequence data; each model predictive value has a corresponding preset weight value.
In this embodiment, although the individual sequence values constituting the time sequence of the resource-related sample data have uncertainty, the variation of the entire sequence has certain regularity, and for the non-stationary time sequence, the d-order difference can be first converted into a stationary time sequence, and then the future trend can be predicted based on the past behavior of the sequence by means of an autoregressive integral sliding average model.
In addition, considering the periodicity of the prediction band of the autoregressive integral moving average model, there is a possibility that trend prediction is opposite due to the fixed period, and a quadratic exponential smoothing model is also introduced in the embodiment, so that the method is used for predicting time series data. It first considers seasonality through one exponential smoothing and then applies exponential smoothing to estimate trends. This helps to improve accurate predictions of future data, especially for data with trend fluctuations, as a complement to the autoregressive integral moving average model.
As for the time-series prediction model of this embodiment, there are various training methods, and those skilled in the art will understand that there are different training methods corresponding to different time-series prediction models, for example, a primary exponential smoothing model, a moving average model, etc., and a training method is schematically described below.
The system predicts through an autoregressive integral moving average model, confirms the AR and MA orders through a difference operation, an autocorrelation function (ACF) and a partial autocorrelation function (PACF), and fits the autoregressive integral moving average model based on the confirmed orders to obtain a first predicted value.
And then carrying out predictive training on the secondary exponential smoothing model. Firstly, calculating a primary index smooth value and a secondary index smooth value for a time sequence of historical resource sample data, then calculating a model parameter value by utilizing the two index smooth values in the last period, and finally, bringing the parameter value into the secondary index smooth model to calculate a second predicted value of the secondary index smooth model.
After the first predicted value and the second predicted value are obtained, different weight values are given to different model predicted values according to the relative accuracy of the first predicted value and the second predicted value, and the predicted data are compared with the acquired historical resource sample data in the next period for a plurality of times, so that the parameters of the model are corrected.
In the application stage, the resource data which can be subjected to inventory packaging in the future is predicted through a secondary exponential smoothing model and an autoregressive integral moving average mixed model, and the gap between the predicted data and the actual data is reduced to 14 percent at present.
Referring to fig. 3, step 101 includes:
301. Respectively inputting the real-time resource data stored in the stock into an autoregressive integral moving average model and a secondary exponential smoothing model to respectively obtain predicted values of the two models;
302. And determining and outputting the predicted amount of the resource data to be packaged according to the predicted values of the two models and the preset weight values corresponding to the two predicted values.
Through steps 301 to 302, the prediction of the resource data prediction amount by two time series prediction models can be realized, so as to obtain a more accurate predicted value.
Further, referring to fig. 4, step 102 includes:
401. And carrying out topology positioning in the resource stream according to different products corresponding to the resource data to be packaged, and determining the sequence of the products corresponding to the different resource data in the resource stream.
Different resource data corresponds to different products, and the different products have corresponding sequences in the resource stream. Taking the second house product as an example, 4 products can be included: resource package 1, resource package 2, resource package 3, resource package 4.
As shown in fig. 5, topology positioning is performed on four products, and the analysis and confirmation are performed on the products by depending on product managers and business specialists, so that the positions of the products in the resource flow can be obtained.
402. Acquiring statistical data of different resource data, performing paret analysis, calculating the duty ratio of products corresponding to the different resource data in different dimensions, determining the level of the different products, and obtaining correction coefficients of the different products; wherein, the stock stores the statistical data of different resource data in the current time period.
Table 1 below shows the duty ratios of different resource data corresponding products in different dimensions. Different types of products should have different evaluation criteria, and the products of this embodiment are scored according to the dispensing amount and the dispensing ratio.
TABLE 1
Considering the commodity topology obtained in the last step comprehensively, two resource packages with highest scores, namely a resource package 1 and a resource package 2, can be determined to be class A products, stricter inventory control is implemented on the two resource packages, and a correction coefficient is obtained according to the contribution duty ratio and is used for subsequent inventory correction.
403. And determining rule configuration of products corresponding to different resource data in different cities according to an input instruction of configuration information of the city rules.
In this embodiment, rules can be flexibly customized for each city according to requirements, such as manually adjusting inventory proportion; if the city is expected to have a certain super-allocation proportion in addition to the normal estimated stock, the bearing coefficient of the city for the super-allocation condition can be set.
Rule customization can be customized for people and can be obtained according to product design principles, city operation requirements, broker intent feedback and data analysis.
City rules include a variety of, for example, regional minimum packing volume, whether a broker supports custom allocation restrictions, etc.
404. And determining the inventory proportion of different products in the next time period according to the input instruction of the inventory proportion.
The stock ratio can be 1 or 0.5, and can be set according to actual requirements.
And storing configuration information of the city rules by using a mysql database, synchronizing the configuration information in the mysql database to a data storage cluster in each inventory period, analyzing the corresponding configuration information by Hive, and predicting resource inventory data of the next period of resource flow.
Referring to fig. 6, fig. 6 shows a flow chart of an inventory dynamic allocation system provided by an embodiment of the invention.
As can be seen in fig. 6, the overall flow can be divided into: resource estimation, rule configuration, inventory estimation and product distribution.
In the resource estimation process, predicting the real-time resource data stored in the inventory through an autoregressive integral moving average model and a secondary exponential smoothing model, and outputting the predicted quantity of the resource data to be packaged.
In the rule configuration process, the product rule is formulated through a resource operation management platform. And after the product rule is formulated, the statistical analysis can be further carried out according to the truly allocated resource data, the duty ratio of products corresponding to different resource data in different dimensions is calculated, the grades of different products are determined, and the correction coefficients of different products are obtained.
In the inventory estimation process, the resource inventory data of the next period is generated and output according to the resource data pre-measurement to be packaged and the corresponding inventory output rule, and the resource inventory data is stored in a MySQL database and finally stored in a Hive database.
In the product distribution process, the plurality of resource inventory data packages are pushed to different brokers and displayed in the buying and selling platform according to the different resource inventory data packages so as to generate the latest real-time resource data. And adjusting parameters of the time sequence prediction model according to the quantity of the output resource inventory data and the difference between the predicted quantity of the resource data to be packaged, so that the difference between the predicted quantity of the resource data output by the time sequence prediction model and the quantity of the real output resource inventory data is smaller than a threshold value.
The following describes the resource-flow-based inventory dynamic allocation device provided by the embodiment of the invention, and the resource-flow-based inventory dynamic allocation device described below and the resource-flow-based inventory dynamic allocation method described above can be referred to correspondingly.
The embodiment of the invention discloses a dynamic inventory allocation device based on resource flow, which is used for a dynamic inventory allocation system, as shown in fig. 7, and comprises the following components:
The resource data prediction module 701 is configured to input real-time resource data stored in inventory into a time sequence prediction model, and output a predicted amount of resource data to be packaged; the time sequence prediction model is obtained by training based on historical resource sample data;
The output rule determining module 702 is configured to determine an inventory output rule corresponding to the resource data to be packaged according to different products corresponding to the resource data to be packaged, where the inventory output rule is used to constrain the resource data to be packaged when outputting the resource inventory data;
the inventory data output module 703 is configured to generate and output the inventory data of the next period according to the predicted amount of the resource data to be packaged and the corresponding inventory output rule.
Optionally, the time series prediction model includes: the system comprises an autoregressive integral moving average model and a secondary exponential smoothing model, wherein the autoregressive integral moving average model is used for autoregressive integral prediction of stable time sequence data, and the secondary exponential smoothing model is used for predicting twice exponential smoothing of the time sequence data; each model predictive value has a corresponding preset weight value;
The resource data prediction module 701 is specifically configured to:
respectively inputting the real-time resource data stored in the stock into an autoregressive integral moving average model and a secondary exponential smoothing model to respectively obtain predicted values of the two models;
and determining and outputting the predicted amount of the resource data to be packaged according to the predicted values of the two models and the preset weight values corresponding to the two predicted values.
Optionally, the inventory production rule includes at least one of: the method comprises the steps of predicting resource data, sequencing of different resource data in a resource flow, correction coefficients of products corresponding to different resource data, and rule configuration and inventory proportion in different cities;
the yield rule determining module 702 is specifically configured to:
Performing topology positioning in a resource stream according to different products corresponding to the resource data to be packaged, and determining the sequence of the products corresponding to the different resource data in the resource stream;
Acquiring statistical data of different resource data, performing paret analysis, calculating the duty ratio of products corresponding to the different resource data in different dimensions, determining the level of the different products, and obtaining correction coefficients of the different products; wherein, the stock stores the statistical data of different resource data in the current time period;
determining rule configuration of products corresponding to different resource data in different cities according to an input instruction of configuration information of the city rules;
and determining the inventory proportion of different products in the next time period according to the input instruction of the inventory proportion.
Optionally, the resource inventory data includes a plurality of resource inventory data packets;
The device also comprises a resource allocation module, a plurality of real-time resource data storage modules and a plurality of resource storage modules, wherein the resource allocation module is used for pushing the plurality of resource storage data packages to different brokers after generating and outputting the resource storage data of the next period, and displaying the resource storage data packages in a buying and selling platform according to the different resource storage data packages so as to generate the latest real-time resource data.
Optionally, the apparatus further comprises: and the training module is used for continuously adjusting the parameters of the time sequence prediction model according to the quantity of the output resource inventory data and the difference value between the predicted quantity of the resource data to be packaged after the resource inventory data of the next period is generated and output, so that the difference value between the predicted quantity of the resource data output by the time sequence prediction model and the quantity of the resource inventory data actually output is smaller than a threshold value.
Fig. 8 illustrates a physical structure diagram of an electronic device, as shown in fig. 8, which may include: processor 810, communication interface (Communications Interface) 820, memory 830, and communication bus 840, wherein processor 810, communication interface 820, memory 830 accomplish communication with each other through communication bus 840. The processor 810 may invoke logic instructions in the memory 830 to perform a resource flow based inventory dynamic allocation method, the method comprising: inputting the real-time resource data stored in the inventory into a time sequence prediction model, and outputting a predicted amount of the resource data to be packaged; the time sequence prediction model is obtained by training based on historical resource sample data; determining a stock output rule corresponding to the resource data to be packaged according to different products corresponding to the resource data to be packaged, wherein the stock output rule is used for generating after analyzing the resource data to be packaged so as to restrict the resource data to be packaged; and generating and outputting resource inventory data of the next period according to the predicted amount of the resource data to be packaged and the corresponding inventory output rule.
Further, the logic instructions in the memory 830 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product, the computer program product including a computer program, the computer program being storable on a non-transitory computer readable storage medium, the computer program, when executed by a processor, being capable of executing the resource flow-based inventory dynamic allocation method provided by the above methods, the method comprising: inputting the real-time resource data stored in the inventory into a time sequence prediction model, and outputting a predicted amount of the resource data to be packaged; the time sequence prediction model is obtained by training based on historical resource sample data; determining a stock output rule corresponding to the resource data to be packaged according to different products corresponding to the resource data to be packaged, wherein the stock output rule is used for generating after analyzing the resource data to be packaged so as to restrict the resource data to be packaged; and generating and outputting resource inventory data of the next period according to the predicted amount of the resource data to be packaged and the corresponding inventory output rule.
In yet another aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, is implemented to perform the resource flow based inventory dynamic allocation method provided by the above methods, the method comprising: inputting the real-time resource data stored in the inventory into a time sequence prediction model, and outputting a predicted amount of the resource data to be packaged; the time sequence prediction model is obtained by training based on historical resource sample data; determining a stock output rule corresponding to the resource data to be packaged according to different products corresponding to the resource data to be packaged, wherein the stock output rule is used for generating after analyzing the resource data to be packaged so as to restrict the resource data to be packaged; and generating and outputting resource inventory data of the next period according to the predicted amount of the resource data to be packaged and the corresponding inventory output rule.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.