CN111222668A - Warehouse list prediction method and device, electronic equipment and readable storage medium - Google Patents
Warehouse list prediction method and device, electronic equipment and readable storage medium Download PDFInfo
- Publication number
- CN111222668A CN111222668A CN201811422336.1A CN201811422336A CN111222668A CN 111222668 A CN111222668 A CN 111222668A CN 201811422336 A CN201811422336 A CN 201811422336A CN 111222668 A CN111222668 A CN 111222668A
- Authority
- CN
- China
- Prior art keywords
- fitting
- value
- historical
- predicted
- history
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/087—Inventory or stock management, e.g. order filling, procurement or balancing against orders
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Theoretical Computer Science (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Game Theory and Decision Science (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a warehouse single quantity prediction method, a warehouse single quantity prediction device, electronic equipment and a readable storage medium, wherein real single quantities of a plurality of historical periods are obtained from historical data of a warehouse; obtaining a prediction fitting value of the next period according to the real single quantity of the plurality of historical periods; and obtaining the single-quantity predicted value of the next period according to the predicted fitting value and the preset predicted strength information corresponding to the next period, introducing the predicted strength information corresponding to the next period on the basis of performing warehouse single-quantity prediction on real single quantities of a plurality of historical periods, considering the influence of artificial control factors on the warehouse single quantity, and improving the accuracy and reliability of warehouse single-quantity prediction.
Description
Technical Field
The invention relates to the field of warehouse logistics, in particular to a warehouse single quantity prediction method and device, electronic equipment and a readable storage medium.
Background
With the rapid development of internet technology, internet-based services, such as e-commerce services of take-out, shopping, etc., are increasing. Based on these services, the user can obtain the required goods without going out. In order to ensure the reliability of the service, the supplier needs to reasonably arrange the resources such as warehouse commodity inventory, personnel and the like. For example, in severe weather, special holidays, marketing activities launched by merchants and other special situations, the commodity order volume usually has the phenomena of sudden increase of the order volume and excessive order backlog, and the suppliers need to allocate more hands in advance and store more spare commodities in advance to cope with the possible coming warehouse order pressure. Therefore, it is important to warehouse inventory prediction.
The main solution at present is to manually observe the order supply and demand conditions of each warehouse, and manually estimate the warehouse inventory based on the historical inventory to arrange warehouse personnel and inventory commodity staff. For example, if the historical single quantity in the new year period of the previous year is 4 times of the average value of the single quantity in the weekday, the single quantity in the new year period of the current year is estimated to be 5 times of the average value of the single quantity in the weekday, and corresponding standby resources are arranged.
However, the current warehouse inventory prediction method is too dependent on empirical values, and the problem that warehouse personnel redundancy is caused by the fact that orders are not sent in time or orders are reduced due to sudden increase of warehouse inventory still exists. The existing warehouse single quantity prediction method is low in reliability.
Disclosure of Invention
The embodiment of the invention provides a warehouse single quantity prediction method, a warehouse single quantity prediction device, electronic equipment and a readable storage medium, and aims to solve the problems that single quantity cannot be accurately predicted and the reliability of single quantity prediction is low.
According to a first aspect of the present invention, there is provided a method for predicting inventory, comprising:
acquiring real single quantities of a plurality of historical periods from historical data of a warehouse;
obtaining a prediction fitting value of the next period according to the real single quantity of the plurality of historical periods;
and obtaining a single-quantity predicted value of the next period according to the predicted fitting value and preset predicted intensity information corresponding to the next period.
Optionally, in a possible implementation manner of the first aspect, the obtaining a predicted fitting value of a next cycle according to a true single quantity of the plurality of history cycles includes:
determining optimized fitting values of the plurality of historical periods according to the real single quantities of the plurality of historical periods;
and performing fitting processing of the next period on the optimized fitting values of the plurality of historical periods by using a preset time sequence prediction algorithm to obtain predicted fitting values.
Optionally, in another possible implementation manner of the first aspect, the determining an optimized fitting value of the plurality of history cycles according to the true single quantity of the plurality of history cycles includes:
fitting the real single quantity of the plurality of historical periods with each historical period by a preset time sequence fitting algorithm to obtain initial fitting values corresponding to the plurality of historical periods;
and determining the optimized fitting values of the plurality of history periods according to the initial fitting values corresponding to the plurality of history periods.
Optionally, in yet another possible implementation manner of the first aspect, the determining an optimized fitting value of the multiple history cycles according to the initial fitting values corresponding to the multiple history cycles includes:
acquiring a trend fitting factor corresponding to each history period;
and determining an optimized fitting value corresponding to each history period according to the initial fitting value, the trend fitting factor and the real single quantity corresponding to each history period.
Optionally, in another possible implementation manner of the first aspect, the determining an optimized fitting value corresponding to each history cycle according to the initial fitting value, the trend fitting factor, and the true single quantity corresponding to each history cycle includes:
taking the product of the initial fitting value and the trend fitting factor corresponding to each history period as an adjusted fitting value;
obtaining relative errors between the adjustment fitting values corresponding to the historical periods and the real single quantity;
if the preset stopping condition is not met, updating the initial fitting value and the trend fitting factor, and returning to execute the product of the initial fitting value and the trend fitting factor corresponding to each history period to be used as an adjustment fitting value;
and if the preset stopping condition is met, taking the current initial fitting value as an optimized fitting value.
Optionally, in a further possible implementation manner of the first aspect, the preset stop condition includes at least one of:
when the trend fitting factor is equal to 1, the absolute value of the relative error is less than 0.15;
when the trend fitting factor is not equal to 1, the absolute value of the relative error is less than 0.1;
and the execution times of the updating reaches the preset upper limit times.
Optionally, in yet another possible implementation manner of the first aspect, the updating the initial fitting value and the trend fitting factor includes:
if the relative error is greater than or equal to 0, reducing the trend fitting factor by a preset step length to obtain a current trend fitting factor;
if the relative error is less than 0, increasing the trend fitting factor by a preset step length to obtain a current trend fitting factor;
and for each history period, taking the quotient of the real single quantity and the current trend fitting factor as a current initial fitting value.
Optionally, in a further possible implementation manner of the first aspect, the obtaining a trend fitting factor corresponding to each history period includes:
acquiring pre-stored historical single quantity predicted values and historical predicted strength information of each historical period;
obtaining a historical prediction fitting value of each historical period according to the historical single quantity prediction value and the historical prediction strength information of each historical period;
and obtaining a trend fitting factor corresponding to each history period according to the real single quantity of each history period and the history prediction fitting value.
Optionally, in yet another possible implementation manner of the first aspect, the preset timing prediction algorithm and the preset timing prediction algorithm both include an ETS algorithm.
Optionally, in another possible implementation manner of the first aspect, before the obtaining a predicted single-amount value of the next cycle according to the prediction fit value and preset predicted intensity information corresponding to the next cycle, the method further includes:
acquiring service attribute information corresponding to the next period, wherein the service attribute information corresponds to preset predicted intensity information;
and using the expected strength information corresponding to the service attribute information as the expected strength information corresponding to the next period.
Optionally, in another possible implementation manner of the first aspect, the obtaining a single-quantity predicted value of the next cycle according to the prediction fit value and preset predicted intensity information corresponding to the next cycle includes:
and taking the product of the predicted fitting value and the predicted intensity information corresponding to the next period as a single-quantity predicted value of the next period.
According to a second aspect of the present invention, there is provided a bin sheet amount prediction device including:
the acquisition module is used for acquiring real single quantities of a plurality of historical periods from historical data of the warehouse;
the processing module is used for obtaining a predicted fitting value of the next period according to the real single quantity of the plurality of historical periods;
and the output module is used for obtaining the single amount predicted value of the next period according to the predicted fitting value and the preset predicted intensity information corresponding to the next period.
Alternatively, in one possible implementation of the second aspect,
the processing module is used for determining optimized fitting values of the plurality of history cycles according to the real single quantities of the plurality of history cycles; and performing fitting processing of the next period on the optimized fitting values of the plurality of historical periods by using a preset time sequence prediction algorithm to obtain predicted fitting values.
Alternatively, in another possible implementation manner of the second aspect,
the processing module is used for performing fitting processing of each history cycle on the real single quantity of the plurality of history cycles by using a preset time sequence fitting algorithm to obtain initial fitting values corresponding to the plurality of history cycles; and determining the optimized fitting values of the plurality of history periods according to the initial fitting values corresponding to the plurality of history periods.
Alternatively, in yet another possible implementation of the second aspect,
the processing module is used for acquiring trend fitting factors corresponding to the historical periods; and determining an optimized fitting value corresponding to each history period according to the initial fitting value, the trend fitting factor and the real single quantity corresponding to each history period.
Alternatively, in yet another possible implementation form of the second aspect,
the processing module is used for taking the product of the initial fitting value corresponding to each history period and the trend fitting factor as an adjustment fitting value; obtaining relative errors between the adjustment fitting values corresponding to the historical periods and the real single quantity; if the preset stopping condition is not met, updating the initial fitting value and the trend fitting factor, and returning to execute the product of the initial fitting value and the trend fitting factor corresponding to each history period to be used as an adjustment fitting value; and if the preset stopping condition is met, taking the current initial fitting value as an optimized fitting value.
Optionally, in a further possible implementation manner of the second aspect, the preset stop condition includes at least one of:
when the trend fitting factor is equal to 1, the absolute value of the relative error is less than 0.15;
when the trend fitting factor is not equal to 1, the absolute value of the relative error is less than 0.1;
and the execution times of the updating reaches the preset upper limit times.
Alternatively, in yet another possible implementation form of the second aspect,
the processing module is used for reducing the trend fitting factor by a preset step length to obtain a current trend fitting factor if the relative error is greater than or equal to 0; if the relative error is less than 0, increasing the trend fitting factor by a preset step length to obtain a current trend fitting factor; and for each history period, taking the quotient of the real single quantity and the current trend fitting factor as a current initial fitting value.
Alternatively, in yet another possible implementation form of the second aspect,
the processing module is used for acquiring pre-stored historical single quantity predicted values and historical predicted intensity information of each historical period; obtaining a historical prediction fitting value of each historical period according to the historical single quantity prediction value and the historical prediction strength information of each historical period; and obtaining a trend fitting factor corresponding to each history period according to the real single quantity of each history period and the history prediction fitting value.
Alternatively, in yet another possible implementation form of the second aspect,
the output module is further configured to obtain service attribute information corresponding to the next period before the single predicted value of the next period is obtained according to the predicted fitting value and the preset predicted intensity information corresponding to the next period, where the service attribute information corresponds to the preset predicted intensity information; and using the expected strength information corresponding to the service attribute information as the expected strength information corresponding to the next period.
Alternatively, in yet another possible implementation form of the second aspect,
and the output module is used for taking the product of the prediction fitting value and the predicted intensity information corresponding to the next period as a single-quantity prediction value of the next period.
According to a third aspect of the present invention, there is provided an electronic apparatus comprising: memory, a processor and a computer program, the computer program being stored in the memory, the processor running the computer program to perform the warehouse inventory prediction method of the first aspect of the invention and of various possible designs of the first aspect of the invention.
According to a fourth aspect of the present invention, there is provided a readable storage medium having stored therein a computer program for implementing the warehouse inventory prediction method of the first aspect of the present invention and of various possible designs of the first aspect, when the computer program is executed by a processor.
According to the warehouse single quantity prediction method, the warehouse single quantity prediction device, the electronic equipment and the readable storage medium, the real single quantity of a plurality of historical periods is obtained from historical data of a warehouse; obtaining a prediction fitting value of the next period according to the real single quantity of the plurality of historical periods; and obtaining the single-quantity predicted value of the next period according to the predicted fitting value and the preset predicted strength information corresponding to the next period, introducing the predicted strength information corresponding to the next period on the basis of performing warehouse single-quantity prediction on real single quantities of a plurality of historical periods, considering the influence of artificial control factors on the warehouse single quantity, and improving the accuracy and reliability of warehouse single-quantity prediction.
Drawings
Fig. 1 is a schematic diagram of an application scenario provided in an embodiment of the present invention;
FIG. 2 is a schematic flow chart of a warehouse inventory prediction method according to an embodiment of the present invention;
FIG. 3 is a schematic flow chart for determining an optimized fit value according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a warehouse unit quantity prediction apparatus according to an embodiment of the present invention;
fig. 5 is a schematic diagram of a hardware structure of an electronic device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The sequence numbers of the processes in the description and the claims of the embodiments of the present invention and the drawings do not mean the execution sequence, and the execution sequence of the processes should be determined by the functions and the internal logic, and should not limit the implementation process of the embodiments of the present invention.
It should be understood that in the embodiments of the present invention, "including" and "having" and any variations thereof, it is intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
It should be understood that, in the embodiments of the present invention, "a plurality" means two or more.
It should be understood that in the embodiment of the present invention, "B corresponding to a", "a corresponds to B", or "B corresponds to a" means that B is associated with a, and B may be determined according to a. Determining B from a does not mean determining B from a alone, but may be determined from a and/or other information.
As used herein, "if" may be interpreted as "at … …" or "when … …" or "in response to a determination" or "in response to a detection", depending on the context.
It should be understood that in the embodiment of the present invention, the history period is a calculation period which already has a real single amount, and may be understood as a calculation period which is previously used for making a warehouse single amount prediction, such as a day, a week or a month.
It should be understood that in the embodiment of the present invention, the time sequence prediction algorithm performs prediction analysis on the future change trend according to the time sequence of the historical statistical data. In general, a time series consists of four variation components, such as long-term trend variations, seasonal variations, periodic variations, and random fluctuations. Simple prediction algorithms can be used to predict the first three trend changes, such as exponential smoothing algorithms, moving average algorithms, and the like.
It should be understood that in the embodiment of the present invention, the ETS algorithm is a time sequence prediction algorithm implemented in R language, and the author Rob Hyndman names the algorithm with E, T, S three letters, which can be understood as Error, Trend and seaselectivity, and can be interpreted as ExponenTial Smoothing. The former discloses three components of the ETS algorithm, and the latter describes the working principle of the ETS algorithm. To be precise, the ETS algorithm is actually a whole series of algorithms, and can be based on any combination of these three components.
The technical solution of the present invention will be described in detail below with specific examples. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments.
Fig. 1 is a schematic view of an application scenario provided in an embodiment of the present invention. In the application scenario shown in fig. 1, a user initiates a network order through a client 1, and a server 2 receives orders from a plurality of clients 1 and sends a prompt to warehouse staff to process the order, and stores the received order in a database 3 for future order prediction. For example, if a day is taken as a period, the amount of orders received each day is taken as the actual amount of orders for a historical period. The server 2 also presets a predicted strength information for reflecting the promotion strength or the purchase demand strength for each period in the database 3, and the server 2 predicts the future time amount of the warehouse amount according to the actual amount of the past days and the predicted strength information in the future. The forecasted single quantities may then be provided to warehouse management personnel for commodity replenishment quantity adjustment, personnel scheduling adjustment, system network adjustment, and the like. For example, before the shopping segment comes, the warehouse inventory of the shopping segment of this year is accurately and reliably predicted by the embodiment, and then the corresponding preparation work of the shopping segment is carried out, so that the possibility of insufficient hands or insufficient inventory during the shopping segment is reduced. The repository in this embodiment may be understood as a repository of an entity service, or may also be understood as a standby resource management system in a virtual service. In the embodiment, the server 2 executes the warehouse inventory prediction method in the following various embodiments, so that the accuracy and reliability of warehouse inventory prediction are improved.
Referring to fig. 2, which is a schematic flowchart of a warehouse inventory prediction method according to an embodiment of the present invention, an execution main body of the method shown in fig. 2 may be a software and/or hardware device, and a server is taken as an execution main body for illustration. The method shown in fig. 2 mainly includes steps S101 to S103, and specifically includes the following steps:
s101, acquiring real single quantities of a plurality of history periods from historical data of the warehouse.
It is understood that the order information is stored when the warehouse orders sent by the clients are received, and the actual pre-stored orders are obtained after the end of each period according to a plurality of historical periods, so that the order amount of the next period in the future is predicted. True single volume refers to the true order quantity. The historical period may be by day, week, month, or other preset time period. For example, with one day as a period, before preparing to start shopping festival promotion, the real singlets of the last 8 weeks, i.e., 56 days, are acquired as the real singlets of a plurality of history periods for making the singlets prediction.
And S102, obtaining a predicted fitting value of the next period according to the real single quantities of the plurality of historical periods.
It can be understood that the variation trend of the true single quantity with time is calculated by the true single quantities of a plurality of historical periods, so as to obtain a fitting curve. Each data point in the fitting curve corresponds to a time point and a single fitting value, and the last time point is extended backward by one time point according to the trend of the fitting curve to obtain the predicted fitting value of the next period.
In one implementation, the server may determine the optimal fitting values of the plurality of history periods according to the true single quantities of the plurality of history periods. For example, a preset time sequence fitting algorithm is used to perform fitting processing on each history cycle on the real single quantity of the plurality of history cycles, so as to obtain initial fitting values corresponding to the plurality of history cycles. And determining the optimized fitting values of the plurality of history periods according to the initial fitting values corresponding to the plurality of history periods. And after the optimal fitting values are obtained, fitting the optimal fitting values of the multiple historical periods in the next period by a preset time sequence prediction algorithm to obtain predicted fitting values. Wherein, the timing prediction algorithm and the timing prediction algorithm may be ETS algorithm.
In particular, the ETS algorithm may be understood as an algorithm executed by a preset ETS model in the server. The server may perform data fitting of smoothing processing on the real single quantities of the plurality of history periods by using a preset ETS model, so as to obtain an initial fitting value corresponding to each history period. And then the server adjusts the initial fitting value sequence and the fitting factors of the initial fitting values to obtain an adjusted optimized fitting value sequence. And finally, fitting the obtained optimized fitting value sequence in the next period by using the ETS model to obtain a predicted fitting value. It can be understood that a time point is added to the existing optimized fitting value sequence to perform the prediction calculation of the fitting value. The same ETS algorithm is used for fitting and predicting, so that trend fitting and predicting are guaranteed by using the same algorithm, and the consistency and accuracy of the predicted fitting value are improved.
And S103, obtaining a single amount predicted value of the next period according to the predicted fitting value and preset predicted intensity information corresponding to the next period.
It is understood that before obtaining the predicted single-quantity value of the next period, the predicted intensity information corresponding to the next period may be determined by:
when the server needs to predict the order quantity of the next period, the server firstly acquires the service attribute information corresponding to the next period, wherein the service attribute information indicates the sales promotion information of the next period. And then using the expected strength information corresponding to the service attribute information as the expected strength information corresponding to the next period. For example, in the next period, the shopping node is included, the service attribute information includes a promotion plan with great strength, each promotion degree corresponds to an expected strength information, and if the pre-planned promotion degree is higher, the expected strength information is determined to be 3. For another example, if the weather forecast of the next cycle is a typhoon weather period, the traffic attribute information includes a plan for closing the market, and the predicted intensity information is set to 0.1. Or, if the next cycle is a normal working day and the service attribute information has no special information, the expected intensity information is set to 1. It is also understood that the different types of service attribute information may be a one-to-one mapping corresponding relationship with the expected intensity information, for example, a preset corresponding relationship list, and then the expected intensity information corresponding to the service attribute information is queried according to the service attribute information.
After determining the predicted intensity information, a specific implementation manner of the step S103 may be that the server takes a product of the prediction fitting value and the predicted intensity information corresponding to the next period as a single-quantity predicted value of the next period. Therefore, the predicted strength information represents the influence of human control factors on the future order quantity in the predicted quantity of the next period.
In the warehouse inventory prediction method provided by the embodiment, the actual inventory of a plurality of history cycles is obtained from the historical data of the warehouse; obtaining a prediction fitting value of the next period according to the real single quantity of the plurality of historical periods; and obtaining the single-quantity predicted value of the next period according to the predicted fitting value and the preset predicted strength information corresponding to the next period, introducing the predicted strength information corresponding to the next period on the basis of performing warehouse single-quantity prediction on real single quantities of a plurality of historical periods, considering the influence of artificial control factors on the warehouse single quantity, and improving the accuracy and reliability of warehouse single-quantity prediction.
In the embodiment shown in fig. 2, before obtaining the predicted fitting value of the next cycle in step S102, there may be a plurality of implementation manners for determining the optimized fitting values of the plurality of history cycles according to the initial fitting values corresponding to the plurality of history cycles. In an alternative implementation manner, the server may first obtain a trend fitting factor corresponding to each of the history periods. And then determining an optimized fitting value corresponding to each history period according to the initial fitting value, the trend fitting factor and the real single quantity corresponding to each history period. Specifically, referring to fig. 3, which is a schematic flowchart illustrating a process of determining an optimal fitting value according to an embodiment of the present invention, the method shown in fig. 3 may implement the determination of the optimal fitting value through the following steps S201 to S207.
S201, obtaining a pre-stored historical single quantity predicted value and historical predicted intensity information of each historical period.
The historical single quantity predicted value can be understood as a single quantity predicted value of a historical period, and the single quantity predicted value obtained each time is stored for subsequent prediction. The historical expected strength information can be determined according to the service attribute information of each historical period, or can be directly read from the historical stored data.
S202, obtaining a historical prediction fitting value of each historical period according to the historical single quantity prediction value and the historical predicted intensity information of each historical period.
It will be appreciated that the historical predicted fit value may be the quotient of the historical single quantity predicted value and the historical predicted intensity information. The resulting historical predicted fit values are sequences corresponding to each historical period.
And S203, obtaining a trend fitting factor corresponding to each history period according to the real single quantity of each history period and the history prediction fitting value.
It will be appreciated that the trend fit factor may be the true single quantity of the history period divided by the historical predicted fit value. And obtaining a trend fitting factor corresponding to each historical period, thereby obtaining a sequence of the trend fitting factors corresponding to the historical periods.
And S204, taking the product of the initial fitting value and the trend fitting factor corresponding to each history period as an adjustment fitting value.
It can be understood that, in order to improve the matching degree of the optimized fitting values and the data trend so as to make the optimized fitting values for reflecting the history period order quantity smoother, the embodiment performs numerical optimization by circularly adjusting the initial fitting values and the trend fitting factors. And obtaining an adjusted fitting value which is more matched with the change trend of the data by multiplying the initial fitting value by the trend fitting factor.
S205, obtaining the relative error between the adjusted fitting value corresponding to each history period and the real single quantity.
The relative error corresponding to each history period can be understood as the percentage of the ratio of the difference between the adjusted fitting value and the real single quantity to the real single quantity.
And S206, if the preset stopping condition is not met, updating the initial fitting value and the trend fitting factor, and returning to execute the product of the initial fitting value and the trend fitting factor corresponding to each history period as an adjusted fitting value.
The specific way of updating the trend fitting factor may be: if the relative error is greater than or equal to 0, indicating that the current initial fitting value is equal to or greater than the true value, reducing the trend fitting factor by a preset step length to obtain a current trend fitting factor; and if the relative error is less than 0, indicating that the current initial fitting value is less than the true value, increasing the trend fitting factor by a preset step length to obtain the current trend fitting factor. The preset step size may be, for example, 0.01.
The specific way to update the initial fitting values in the loop may be: and for each history period, taking the quotient of the real single quantity and the current trend fitting factor as a current initial fitting value so as to use the initial fitting value in the next cycle.
Wherein the preset stop condition may be at least one of:
when the trend fitting factor is equal to 1, the absolute value of the relative error is less than 0.15;
when the trend fitting factor is not equal to 1, the absolute value of the relative error is less than 0.1;
and the execution times of the updating reaches the preset upper limit times.
Wherein the limitation of the number of executions may prevent overfitting.
And S207, if the preset stopping condition is met, taking the current initial fitting value as an optimized fitting value.
In the embodiment, the initial fitting values and the trend fitting factors are adjusted in a circulating manner, so that the initial fitting values are gradually and smoothly optimized, and each optimized fitting value approaching the variation trend of the real value is obtained, and the accuracy of the single quantity predicted value is improved.
The table is an example of warehouse single quantity prediction process data provided by the embodiment of the present invention, and in the example shown in the table one below, a variety of historical data of a warehouse is stored in advance in the database, and the server obtains a pre-stored historical single quantity prediction value 600 and historical predicted strength information 1.2 from the database.
Table one shows only the historical single quantity predicted value and the historical predicted intensity information for one day as an example, but in practice, the historical data for the previous 56 days should be acquired: 56 historical single quantity predicted values, 56 historical predicted intensity information and 56 true single quantities. And performing the same calculation on each historical data to obtain 56 trend fitting factors, and inputting 56 real single quantities into the ETS model to obtain an initial fitting value of the next period. If the next cycle is one day, then only one initial fit value is obtained 570; if the next cycle is 7 days, then 7 initial fit values can be obtained, and the initial fit value for the first day of the next cycle is 570.
Specifically, the present embodiment obtains the historical predicted fitting value 500 as the quotient of 600 and 1.2 for the historical single quantity predicted value 600 and the historical predicted intensity information 1.2 on the example date. The true single quantity 700 of the history period on the example date is obtained, and the historical trend fitting factor of 1.4 is obtained by the quotient of the true single quantity 700 and the historical prediction fitting value 500. Inputting 56 real single quantities corresponding to the previous 56 days into a preset ETS model for data fitting to obtain an initial fitting value or a first initial fitting value 570 of the next period. And obtaining an adjusted fitting value 798 from each initial fitting value and the corresponding trend fitting factor. And obtaining a relative error between the adjusted fitting value 798 and the true single quantity of 0.14, wherein the relative error does not meet the stopping condition, and therefore, performing adjustment circulation on the initial fitting value and the trend fitting factor until the initial fitting value and the trend fitting factor which meet the stopping condition are obtained, and stopping continuous adjustment. For example, 503.6 obtained by adjusting the initial fitting value 570 for 3 times meets the preset stop condition, and then 503.6 is used as the finally obtained optimized fitting value to be input into the ETS model for single quantity prediction in the next period, so as to obtain a single quantity prediction value.
Watch 1
Referring to fig. 4, which is a schematic structural diagram of a warehouse inventory prediction apparatus according to an embodiment of the present invention, the warehouse inventory prediction apparatus 40 shown in fig. 4 mainly includes:
an obtaining module 41, configured to obtain, from the historical data of the warehouse, real single quantities of multiple historical periods;
a processing module 42, configured to obtain a predicted fitting value of the next cycle according to the real single quantities of the multiple history cycles;
and an output module 43, configured to obtain a single quantity predicted value of the next period according to the predicted fitting value and preset predicted intensity information corresponding to the next period.
The warehouse inventory prediction apparatus 40 of the embodiment shown in fig. 4 can be correspondingly used to perform the steps of the method embodiment shown in fig. 2, and the implementation principle and the technical effect are similar, and are not described herein again.
Optionally, the processing module 42 is configured to determine an optimized fitting value of the plurality of history cycles according to the true single quantity of the plurality of history cycles; and performing fitting processing of the next period on the optimized fitting values of the plurality of historical periods by using a preset time sequence prediction algorithm to obtain predicted fitting values.
Optionally, the processing module 42 is configured to perform fitting processing on each history cycle on the real single quantity of the multiple history cycles by using a preset time sequence fitting algorithm, so as to obtain initial fitting values corresponding to the multiple history cycles; and determining the optimized fitting values of the plurality of history periods according to the initial fitting values corresponding to the plurality of history periods.
Optionally, the processing module 42 is configured to obtain a trend fitting factor corresponding to each history period; and determining an optimized fitting value corresponding to each history period according to the initial fitting value, the trend fitting factor and the real single quantity corresponding to each history period.
Optionally, the processing module 42 is configured to use a product of the initial fitting value and the trend fitting factor corresponding to each history cycle as an adjusted fitting value; obtaining relative errors between the adjustment fitting values corresponding to the historical periods and the real single quantity; if the preset stopping condition is not met, updating the initial fitting value and the trend fitting factor, and returning to execute the product of the initial fitting value and the trend fitting factor corresponding to each history period to be used as an adjustment fitting value; and if the preset stopping condition is met, taking the current initial fitting value as an optimized fitting value.
Optionally, the preset stop condition includes at least one of:
when the trend fitting factor is equal to 1, the absolute value of the relative error is less than 0.15;
when the trend fitting factor is not equal to 1, the absolute value of the relative error is less than 0.1;
and the execution times of the updating reaches the preset upper limit times.
Optionally, the processing module 42 is configured to reduce the trend fitting factor by a preset step length to obtain a current trend fitting factor if the relative error is greater than or equal to 0; if the relative error is less than 0, increasing the trend fitting factor by a preset step length to obtain a current trend fitting factor; and for each history period, taking the quotient of the real single quantity and the current trend fitting factor as a current initial fitting value.
Optionally, the processing module 42 is configured to obtain a pre-stored historical single quantity predicted value and historical predicted intensity information of each historical period; obtaining a historical prediction fitting value of each historical period according to the historical single quantity prediction value and the historical prediction strength information of each historical period; and obtaining a trend fitting factor corresponding to each history period according to the real single quantity of each history period and the history prediction fitting value.
Optionally, the output module 43 is further configured to, before the predicted single-quantity value of the next period is obtained according to the predicted fitting value and the preset predicted intensity information corresponding to the next period, obtain service attribute information corresponding to the next period, where the service attribute information indicates promotion information of the next period, and the service attribute information corresponds to the preset predicted intensity information; and using the expected strength information corresponding to the service attribute information as the expected strength information corresponding to the next period.
Optionally, the output module 43 is configured to take a product of the prediction fit value and the expected intensity information corresponding to the next period as a single quantity prediction value of the next period.
The warehouse inventory prediction device provided by the embodiment is used for acquiring real inventory of a plurality of history cycles from historical data of a warehouse through an acquisition module; the processing module is used for obtaining a predicted fitting value of the next period according to the real single quantity of the plurality of historical periods; the output module is used for obtaining the single-quantity predicted value of the next period according to the predicted fitting value and the preset predicted intensity information corresponding to the next period, and on the basis of warehouse single-quantity prediction by using real single quantities of a plurality of historical periods, the predicted intensity information corresponding to the next period is introduced, so that the influence of artificial control factors on warehouse single-quantity is considered, and the accuracy and reliability of warehouse single-quantity prediction are improved.
Referring to fig. 5, which is a schematic diagram of a hardware structure of an electronic device according to an embodiment of the present invention, the electronic device 50 includes: a processor 51, a memory 52 and computer programs; wherein
A memory 52 for storing the computer program, which may also be a flash memory (flash). The computer program is, for example, an application program, a functional module, or the like that implements the above method.
A processor 51 for executing the computer program stored in the memory to implement the steps performed by the electronic device in the above method. Reference may be made in particular to the description relating to the preceding method embodiment.
Alternatively, the memory 52 may be separate or integrated with the processor 51.
When the memory 52 is a device independent of the processor 51, the electronic apparatus may further include:
a bus 53 for connecting the memory 52 and the processor 51.
The present invention also provides a readable storage medium, in which a computer program is stored, which, when being executed by a processor, is adapted to implement the methods provided by the various embodiments described above.
The readable storage medium may be a computer storage medium or a communication medium. Communication media includes any medium that facilitates transfer of a computer program from one place to another. Computer storage media may be any available media that can be accessed by a general purpose or special purpose computer. For example, a readable storage medium is coupled to the processor such that the processor can read information from, and write information to, the readable storage medium. Of course, the readable storage medium may also be an integral part of the processor. The processor and the readable storage medium may reside in an Application Specific Integrated Circuits (ASIC). Additionally, the ASIC may reside in user equipment. Of course, the processor and the readable storage medium may also reside as discrete components in a communication device. The readable storage medium may be a read-only memory (ROM), a random-access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.
The present invention also provides a program product comprising execution instructions stored in a readable storage medium. The at least one processor of the device may read the execution instructions from the readable storage medium, and the execution of the execution instructions by the at least one processor causes the device to implement the methods provided by the various embodiments described above.
In the above embodiments of the electronic device, it should be understood that the Processor may be a Central Processing Unit (CPU), other general-purpose processors, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of a method disclosed in connection with the present invention may be embodied directly in a hardware processor, or in a combination of the hardware and software modules within the processor.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (14)
1. A warehouse single quantity prediction method is characterized by comprising the following steps:
acquiring real single quantities of a plurality of historical periods from historical data of a warehouse;
obtaining a prediction fitting value of the next period according to the real single quantity of the plurality of historical periods;
and obtaining a single-quantity predicted value of the next period according to the predicted fitting value and preset predicted intensity information corresponding to the next period.
2. The method of claim 1, wherein obtaining a predictive fit value for a next cycle from a true single quantity for the plurality of historical cycles comprises:
determining optimized fitting values of the plurality of historical periods according to the real single quantities of the plurality of historical periods;
and performing fitting processing of the next period on the optimized fitting values of the plurality of historical periods by using a preset time sequence prediction algorithm to obtain predicted fitting values.
3. The method of claim 2, wherein determining the optimized fit value for the plurality of history cycles based on the true single quantity of the plurality of history cycles comprises:
fitting the real single quantity of the plurality of historical periods with each historical period by a preset time sequence fitting algorithm to obtain initial fitting values corresponding to the plurality of historical periods;
and determining the optimized fitting values of the plurality of history periods according to the initial fitting values corresponding to the plurality of history periods.
4. The method of claim 3, wherein determining the optimized fit values for the plurality of history cycles according to the initial fit values for the plurality of history cycles comprises:
acquiring a trend fitting factor corresponding to each history period;
and determining an optimized fitting value corresponding to each history period according to the initial fitting value, the trend fitting factor and the real single quantity corresponding to each history period.
5. The method of claim 4, wherein determining the optimized fit value for each history cycle according to the initial fit value, the trend fit factor and the true single quantity for each history cycle comprises:
taking the product of the initial fitting value and the trend fitting factor corresponding to each history period as an adjusted fitting value;
obtaining relative errors between the adjustment fitting values corresponding to the historical periods and the real single quantity;
if the preset stopping condition is not met, updating the initial fitting value and the trend fitting factor, and returning to execute the product of the initial fitting value and the trend fitting factor corresponding to each history period to be used as an adjustment fitting value;
and if the preset stopping condition is met, taking the current initial fitting value as an optimized fitting value.
6. The method of claim 5, wherein the preset stop condition comprises at least one of:
when the trend fitting factor is equal to 1, the absolute value of the relative error is less than 0.15;
when the trend fitting factor is not equal to 1, the absolute value of the relative error is less than 0.1;
and the execution times of the updating reaches the preset upper limit times.
7. The method of claim 5, wherein said updating the initial fit values and the trend fit factors comprises:
if the relative error is greater than or equal to 0, reducing the trend fitting factor by a preset step length to obtain a current trend fitting factor;
if the relative error is less than 0, increasing the trend fitting factor by a preset step length to obtain a current trend fitting factor;
and for each history period, taking the quotient of the real single quantity and the current trend fitting factor as a current initial fitting value.
8. The method of claim 4, wherein obtaining a trend fit factor corresponding to each of the historical periods comprises:
acquiring pre-stored historical single quantity predicted values and historical predicted strength information of each historical period;
obtaining a historical prediction fitting value of each historical period according to the historical single quantity prediction value and the historical prediction strength information of each historical period;
and obtaining a trend fitting factor corresponding to each history period according to the real single quantity of each history period and the history prediction fitting value.
9. The method of claim 3, wherein the predetermined timing prediction algorithm and the predetermined timing prediction algorithm each comprise an ETS algorithm.
10. The method according to claim 1, further comprising, before the obtaining the predicted single-volume value for the next cycle according to the prediction fit value and the preset predicted intensity information corresponding to the next cycle:
acquiring service attribute information corresponding to the next period, wherein the service attribute information corresponds to preset predicted strength information;
and using the expected strength information corresponding to the service attribute information as the expected strength information corresponding to the next period.
11. The method according to any one of claims 1 to 10, wherein the obtaining a predicted single-amount value of the next cycle according to the prediction fit value and the preset predicted intensity information corresponding to the next cycle comprises:
and taking the product of the predicted fitting value and the predicted intensity information corresponding to the next period as a single-quantity predicted value of the next period.
12. A warehouse sheet amount prediction device, comprising:
the acquisition module is used for acquiring real single quantities of a plurality of historical periods from historical data of the warehouse;
the processing module is used for obtaining a predicted fitting value of the next period according to the real single quantity of the plurality of historical periods;
and the output module is used for obtaining the single amount predicted value of the next period according to the predicted fitting value and the preset predicted intensity information corresponding to the next period.
13. An electronic device, comprising: a memory, a processor, and a computer program stored in the memory, the processor executing the computer program to perform the warehouse inventory prediction method of any of claims 1-11.
14. A readable storage medium, having stored thereon a computer program which, when executed by a processor, is adapted to carry out the warehouse inventory prediction method of any of claims 1 to 11.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811422336.1A CN111222668A (en) | 2018-11-27 | 2018-11-27 | Warehouse list prediction method and device, electronic equipment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811422336.1A CN111222668A (en) | 2018-11-27 | 2018-11-27 | Warehouse list prediction method and device, electronic equipment and readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111222668A true CN111222668A (en) | 2020-06-02 |
Family
ID=70827215
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811422336.1A Pending CN111222668A (en) | 2018-11-27 | 2018-11-27 | Warehouse list prediction method and device, electronic equipment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111222668A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112561562A (en) * | 2020-11-13 | 2021-03-26 | 广汽蔚来新能源汽车科技有限公司 | Order issuing method and device, computer equipment and storage medium |
| CN113436733A (en) * | 2021-08-26 | 2021-09-24 | 肾泰网健康科技(南京)有限公司 | Characteristic construction method of hemodialysis scheme generation model based on fusion experience |
| CN114331567A (en) * | 2022-03-03 | 2022-04-12 | 未来地图(深圳)智能科技有限公司 | Order popularity prediction method, prediction system and storage medium |
-
2018
- 2018-11-27 CN CN201811422336.1A patent/CN111222668A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112561562A (en) * | 2020-11-13 | 2021-03-26 | 广汽蔚来新能源汽车科技有限公司 | Order issuing method and device, computer equipment and storage medium |
| CN113436733A (en) * | 2021-08-26 | 2021-09-24 | 肾泰网健康科技(南京)有限公司 | Characteristic construction method of hemodialysis scheme generation model based on fusion experience |
| CN113436733B (en) * | 2021-08-26 | 2021-11-30 | 肾泰网健康科技(南京)有限公司 | Characteristic construction method of hemodialysis scheme generation model based on fusion experience |
| CN114331567A (en) * | 2022-03-03 | 2022-04-12 | 未来地图(深圳)智能科技有限公司 | Order popularity prediction method, prediction system and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111429048B (en) | Method, device and equipment for determining replenishment information | |
| US10748072B1 (en) | Intermittent demand forecasting for large inventories | |
| WO2019001120A1 (en) | Method and system for processing dynamic pricing data of commodity | |
| CN109325825B (en) | Flow data regulation and control method and device | |
| CN111222668A (en) | Warehouse list prediction method and device, electronic equipment and readable storage medium | |
| WO2020082611A1 (en) | Method for carrying out deep learning on basis of blockchain platform and electronic device | |
| CN107767092B (en) | Processing method, display method and device of commodity object information | |
| CN110555578A (en) | sales prediction method and device | |
| CN112070301A (en) | Method, system and equipment for adjusting power consumption of user | |
| US20200311749A1 (en) | System for Generating and Using a Stacked Prediction Model to Forecast Market Behavior | |
| CN111932314A (en) | Method, device and equipment for pushing recommended content and readable storage medium | |
| US20170178221A1 (en) | Method and system for inventory availability prediction | |
| CN114663015A (en) | Replenishment method and device | |
| CN114140252A (en) | Resource allocation method of target object and related device | |
| KR20210014432A (en) | Method for accurate estimating demand of Raw Materials Using System for estimating demand of Raw Materials | |
| CN113627847A (en) | Method and device for generating replenishment list | |
| US20210304243A1 (en) | Optimization of markdown schedules for clearance items at physical retail stores | |
| US20100082405A1 (en) | Multi-period-ahead Forecasting | |
| CN116128408A (en) | Intelligent replenishment method, system and computer readable medium | |
| CN111144634A (en) | Method and device for predicting power price | |
| CN113781120A (en) | Construction method of sales amount prediction model and sales amount prediction method | |
| CN114091770A (en) | Prediction analysis method, device, equipment and storage medium for material demand plan | |
| CN114219247A (en) | Processing method and device of equipment consumables, computer equipment and storage medium | |
| CN109614386B (en) | Data processing method, device, server and computer readable storage medium | |
| CN114254841A (en) | Resource limit processing method, computing device, user device and server |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |