CN112734340B - Method, device, equipment and storage medium for screening prediction index of express delivery quantity - Google Patents
Method, device, equipment and storage medium for screening prediction index of express delivery quantity Download PDFInfo
- Publication number
- CN112734340B CN112734340B CN202110086137.3A CN202110086137A CN112734340B CN 112734340 B CN112734340 B CN 112734340B CN 202110086137 A CN202110086137 A CN 202110086137A CN 112734340 B CN112734340 B CN 112734340B
- Authority
- CN
- China
- Prior art keywords
- prediction
- index
- indexes
- screening
- factors
- 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.)
- Active
Links
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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Pure & Applied Mathematics (AREA)
- Educational Administration (AREA)
- Mathematical Physics (AREA)
- Game Theory and Decision Science (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Accounting & Taxation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Probability & Statistics with Applications (AREA)
- Life Sciences & Earth Sciences (AREA)
- Algebra (AREA)
- Finance (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A prediction index screening method, a device, equipment and a storage medium for predicting the quantity of express items, wherein the method comprises the following steps: acquiring historical data of the quantity of the parts, preprocessing the historical data and screening out data information of the quantity of the target parts; designing a prediction index for predicting the quantity of the express items according to the target quantity data information; respectively establishing regression analysis prediction models between each prediction index and the piece quantity index; and respectively carrying out part quantity prediction by adopting various regression analysis prediction models, calculating error information of part quantity prediction results of the regression analysis prediction models according to actual part quantity data information, and screening prediction indexes contained in the regression analysis prediction models with minimum errors as prediction indexes for predicting the express part quantity. The optimal prediction index is reasonably screened out by designing various prediction indexes, creating a regression analysis prediction model between the prediction index and the quantity of the express items, so that the accuracy of the quantity prediction of the express items can be improved according to the screened prediction index.
Description
Technical Field
The application relates to the technical field of express delivery, in particular to a prediction index screening method, a device, equipment and a storage medium for predicting the quantity of express delivery.
Background
With the rapid development of the logistics industry, the management and control of the business volume (express delivery volume) is related to whether the business of the logistics company can be normally performed. Therefore, it is important to predict the amount of the parts; however, the prediction of the amount of the express delivery is affected by various factors, if the prediction index for predicting the amount of the express delivery cannot be accurately provided, the purpose of accurately predicting the amount of the express delivery cannot be achieved, for example, the difference of the prediction indexes causes large prediction error of the amount of the express delivery, so how to screen the prediction index is the problem of primarily solving the prediction of the amount of the express delivery.
Disclosure of Invention
In order to solve the problem of how to screen the prediction index influencing the predicted express delivery quantity in the technical field of express delivery, the application provides a prediction index screening method, a device, equipment and a storage medium for predicting the express delivery quantity.
The technical scheme of the application is as follows:
the application provides a prediction index screening method for predicting the quantity of express items, which comprises the following steps:
acquiring historical data of the quantity of the parts, preprocessing the historical data and screening out data information of the quantity of the target parts;
designing a prediction index for predicting the quantity of the express delivery according to the target quantity data information;
respectively establishing regression analysis prediction models between each prediction index and the piece quantity index;
and respectively carrying out part quantity prediction by adopting various regression analysis prediction models, calculating error information of part quantity prediction results of the regression analysis prediction models according to actual part quantity data information, and screening prediction indexes contained in the regression analysis prediction models with minimum errors as prediction indexes for predicting the express part quantity.
Further preferably, the prediction index includes: a plurality of prediction indexes in recent factors, contemporaneous factors, periodic factors, holiday factors, e-commerce factors, month end factors, temporary holiday factors, weather factors and economic factors.
Further preferably, the establishing a regression analysis prediction model between each prediction index and the part quantity index specifically includes the steps of:
screening independent variables for predicting the quantity of the piece from each prediction index;
and taking the piece quantity as a dependent variable, and respectively establishing regression analysis prediction models between each prediction index and the piece quantity index according to the screened independent variable.
Further preferably, the step of screening the independent variable for predicting the amount of the piece from each prediction index specifically includes the steps of:
calculating the correlation among the prediction indexes;
screening out indexes with relatively smaller absolute values of correlation coefficients among indexes and relatively larger correlation coefficients of indexes and the piece quantity;
fitting the relation between the screened indexes and the piece quantity, and establishing a regression model;
and screening out indexes with P value smaller than 0.05 or smaller than 0.01 in the regression model as independent variables of the predicted piece quantity.
Further preferably, the calculating the correlation between the predictors includes calculating pearson correlation coefficients between the predictors.
Further preferably, the preprocessing the history data includes: cleaning historical data, replacing null data and processing abnormal data.
The application also provides a prediction index screening device for predicting the express delivery quantity, which comprises the following steps:
the data processing module is used for acquiring historical data of the quantity of the parts, preprocessing the historical data and screening out data information of the quantity of the target parts;
the prediction index design module is used for designing a prediction index for predicting the express delivery quantity according to the target quantity data information;
the regression model creation module is used for respectively creating regression analysis prediction models between each prediction index and the quantity index;
and the prediction index screening module is used for respectively carrying out the part quantity prediction by adopting various regression analysis prediction models, calculating error information of the part quantity prediction results of each regression analysis prediction model according to the actual part quantity data information, and screening the prediction index contained in the regression analysis prediction model with the minimum error as the prediction index for predicting the express part quantity.
Further preferably, the regression model creation model comprises a screening unit and a creation unit;
the screening unit is used for screening independent variables for predicting the quantity of the piece from all the prediction indexes;
the creating unit is used for taking the piece quantity as a dependent variable and respectively creating regression analysis prediction models between each prediction index and the piece quantity index according to the screened independent variable.
The application also provides a predictor screening device for predicting the quantity of the express delivery, which comprises a processor and a memory, wherein at least one instruction, at least one section of program, a code set or an instruction set is stored in the memory, and the at least one instruction, the at least one section of program, the code set or the instruction set is loaded and executed by the processor so as to realize the predictor screening method.
The application also provides a computer readable storage medium, wherein the computer readable storage medium is stored with a computer program, and the computer program realizes the prediction index screening method when being executed by a processor.
According to the prediction index screening method, the device, the terminal equipment and the storage medium, various prediction indexes are designed, regression analysis prediction models between the prediction indexes and the quantity indexes are established, and the prediction indexes contained in the optimal regression analysis prediction models are obtained through screening to serve as prediction indexes for predicting the quantity of the express items.
Drawings
FIG. 1 is a flowchart of a predictor screening method;
FIG. 2 is a flow chart of an index screening method;
FIG. 3 is a schematic diagram of a predictor screening apparatus;
fig. 4 is a schematic diagram of a predictor screening apparatus.
Detailed Description
The application will be described in further detail below with reference to the drawings by means of specific embodiments.
Embodiment one:
the embodiment provides a prediction index screening method for predicting the quantity of express items, and a flowchart of the method is shown in fig. 1, and specifically includes the following steps.
0036S 100: and acquiring historical data of the quantity of the parts, preprocessing the historical data and screening out data information of the quantity of the target parts.
0037.s200: and designing a prediction index for predicting the quantity of the express items according to the target quantity data information.
0039.s300: and respectively establishing regression analysis prediction models between each prediction index and the piece quantity index.
S400: and respectively carrying out part quantity prediction by adopting various regression analysis prediction models, calculating error information of part quantity prediction results of the regression partition prediction models according to actual part quantity data information, and screening prediction indexes contained in the regression analysis prediction models with minimum errors as prediction indexes for predicting express part quantity.
The above steps S100 to S400 are specifically described below.
In step S100, historical data of the part quantity is obtained, the historical data is preprocessed, and target part quantity data information is screened out, wherein the historical data of the part quantity refers to part quantity data stored in the logistics industry, and may also be part quantity data in the logistics industry within a certain period of time published by a certain statistical organization. The amount includes the amount of the received piece, or may include the amount of the sent piece, or may include the amount of the received piece and the amount of the sent piece. In the database, information of the delivery amount and the receiving amount is stored whether on line or off line. Such information may include, but is not limited to: type of piece quantity, time. The time can be stored by day, by week or according to the specific time of the logging system.
Preprocessing the acquired historical data, including: cleaning historical data, replacing null data and processing abnormal data; and cleaning the historical data, and removing unnecessary information and replacement abnormal data in the acquired historical data. Some irregular data typically needs to be filtered out before statistical analysis of the data can be performed to ensure accuracy of the analysis. Data cleansing is a process that reduces data errors and inconsistencies, mainly by detecting and deleting or correcting irregular data.
In this embodiment, the quantity is mainly predicted, so that the single number information and the address information contained in the history data can be removed. Among these historical data, null data or data with numerical anomalies (such as non-numerical representations) may occur, which are replaced with their neighbors.
Specifically, the historical data includes the receiving amount and/or the sending amount, the information of the receiving amount (with or without orders) and the sending amount of each network point can be called from the database according to different business scenes, the receiving amount of a certain network point is taken as test data, the date of the historical data is the receiving amount in 2017-2020, and the obtained historical data can be shown in the following table 1 after data cleaning.
TABLE 1
| Date of collection | Quantity of received parts |
| 2017/1/1 | XXXXXX |
| 2017/1/2 | XXXXXX |
| … | XXXXXX |
| 2020/12/31 | XXXXXX |
The abnormal data processing method may be a deletion method, a substitution method (continuous variable mean substitution, discrete variable mode substitution, and median substitution), or an interpolation method (regression interpolation, multiple interpolation), or may be a method of changing an abnormal value into a missing value and then performing a missing value interpolation operation, in addition to direct deletion. In practical applications, abnormal value processing is generally classified into NA missing values or data trimming.
In step S200, a prediction index for predicting the amount of the express delivery is designed according to the target amount data information; specifically, a plurality of predictors are designed, and preferably, the predictors include: the method comprises the steps of predicting indexes of recent factors, contemporaneous factors, periodic factors, holiday factors, electronic commerce factors, month end factors, temporary holiday factors, weather factors and economic factors, and carrying out exponential smoothing on the predicted indexes by adopting different exponential smoothing, so that the processed indexes are smoother and have smaller fluctuation.
Further, the recent factors include recent parts amount, recent parts amount smooth value, recent week average value, the contemporaneous factors include contemporaneous parts amount index in the history year, and the periodic factors include parts amount of the same day in the history week, parts amount of the same day in the history month, parts amount of the same day in the history quarter, parts amount of the same day in the history year.
Taking an index smoothing method as an example to design the prediction index of the express delivery quantity.
Assuming that the number of the packages to be predicted on the t th day is xt, the number of the packages to be predicted on the t-1 th day is xt-1, and so on, designing a prediction index:
the entry factors include the following:
1) Recent part quantity: xt-1, xt-2, …, xt-14 represent the first 1,2, …,14 days of piece count, respectively.
2) Recent piece amount smoothing: average over about 3 days, average over about 5 days, average over about 7 days, average over about 10 days, and average over about 14 days. Wherein (xt-1+xt-2+xt-3)/3 represents a near 3 day average, and so on
3) Recent week average: average about 2-8 days, average about 7-14 days.
The contemporaneous factors include the following:
the contemporaneous factors include an index of the amount of contemporaneous pieces in the historical year, e.g., calculating a contemporaneous index of the last year, such as today lxt of the last year. The last year contemporaneous index of all indexes in the recent factors can be calculated by changing the component collecting quantity in the current year time period into the component collecting quantity in the same year time period.
The periodicity factors include the following:
the periodic factors include the amount of pieces on the same day in the history week, the amount of pieces on the same day in the history month, the amount of pieces on the same day in the history quarter, the amount of pieces on the same day in the history year, for example, the amount of pieces on the same day in the last week, the same day in the last month, the same day in the last quarter, and the same day in the last year are calculated.
Holiday factors include the following:
the express delivery is affected by production and life, and has holidays such as primordial denier, qingming, noon, mid-autumn, national celebration, spring festival and the like. And respectively indicating whether the corresponding date is the index of a primordial denier, a Qingming day, an end noon, a mid-autumn festival, a national celebration, a spring festival and the like by using 0-1 variables. Holidays in the timeframe may also be invoked.
The e-commerce factors include the following:
the express cable is affected by the electric business, and has six-eight, double-nine, double-eleven, double-twelve, annual goods and other electric business sections. And respectively indicating the corresponding date by using 0-1 variables to obtain indexes such as six, eight, double nine, double eleven, double twelve, annual goods festival and the like. E-commerce holidays in the timeframe may also be invoked.
The month end factor includes the following:
the express delivery quantity is influenced by the operation target, the sales target of clients and the activities, the situation that the month end quantity is increased rapidly possibly occurs, and the month end 26-31 daily 0-1 variable is represented.
The temporary holiday includes the following:
under the influence of policy factors, temporary holidays, such as Beijing national celebration, going to the sea for blessing, etc., appear in part of the areas, and are represented by 0-1 variable.
Weather factors include the following:
the express delivery quantity is influenced by weather factors, so that the warehouse-in quantity lags behind the actual quantity.
The economic environment includes the following:
the shopping of the electronic commerce can be influenced by the economic environment, so that the express delivery quantity is influenced.
In step S300, a regression analysis prediction model between each prediction index and the quantity index is respectively established, and specifically includes the steps of:
s301: and screening independent variables for predicting the quantity of the piece from each prediction index.
Through the above-described design process of the predictors, various predictors can be designed, for example: although each of the prediction indexes is closely related to the amount of the express delivery, all the prediction indexes cannot be used for the amount prediction, so that not only is the use inconvenient, but also the prediction accuracy may be affected, and how to select the appropriate prediction index is important.
In this step, the independent variables for predicting the amount of the piece are specifically selected by the following steps, and a specific flowchart of the selection is shown in fig. 2:
s3011: and calculating the correlation among the prediction indexes.
Specifically, the Pearson correlation coefficient (Pearson correlation coefficient), also called Pearson product-moment correlation coefficient, abbreviated as PPMCC or PCCs, is used to measure the correlation (linear correlation) between two variables X and Y, and its value is between-1 and 1. Specifically, if the pearson correlation coefficient is greater than 0, it indicates that the two variables are positively correlated, that is, the greater the value of one variable, the greater the value of the other variable; if the pearson correlation coefficient is less than 0, it indicates that the two variables are negatively correlated, i.e., the larger the value of one variable, the smaller the value of the other variable will be; the greater the about absolute value of the pearson correlation coefficient, the stronger the correlation of the two variables; if the pearson correlation coefficient is equal to 0, it indicates that the two variables are not linearly related, but there may be other ways of correlation.
Therefore, in step S3011, two-to-two correlation or non-correlation between the two predictors is calculated, and the pearson correlation coefficient between the predictors is calculated to obtain the correlation between the two predictors.
S3012: and screening out indexes with relatively smaller absolute values of correlation coefficients among indexes and relatively larger correlation coefficients of indexes and the piece quantity.
Because the correlation values of each scene are different, some of the scenes are generally larger, and some of the scenes are generally smaller, the scenes can be selected according to specific practical situations, for example, a relatively larger value can be selected, and a relatively smaller value can be selected, for example, the correlations are all larger, one correlation is 0.9, and the other correlation is 0.95, and then 0.95 is selected.
S3013: fitting the relation between the screened indexes and the piece quantity, and establishing a regression model.
Namely, fitting the corresponding relation between the indexes screened in the step S3012 and the piece quantity to establish a regression model.
S3014: and screening out indexes with P value smaller than 0.05 or smaller than 0.01 in the regression model as independent variables of the predicted piece quantity.
The coefficient of each independent variable can be fitted through fitting the regression model, meanwhile, the P value of each independent variable can be fitted, and indexes with the P value smaller than 0.05 or the P value smaller than 0.01 are screened out as independent variables of the final predicted piece quantity in step S3014.
S302: and taking the piece quantity as a dependent variable, and respectively establishing regression analysis prediction models between each prediction index and the piece quantity index according to the screened independent variable.
For example, by the predictors screened in step S301, a corresponding linear regression analysis prediction model is established for each predictor.
In step S400, the component quantity predictions are performed by using various regression analysis prediction models, and error information of the component quantity prediction results of each regression analysis prediction model is calculated according to the actual component quantity data information, and the prediction index included in the regression analysis prediction model with the minimum error is selected as the prediction index for predicting the express component quantity.
Further, calculating error information of the component quantity prediction result of each regression analysis prediction model according to the actual component quantity data information, wherein the error information specifically comprises any one or more of error information of a calculated component quantity prediction error value, a component quantity prediction error rate, a model error, a measurement error, a truncation error and a rounding error; the calculation process of the piece quantity prediction error value and the piece quantity prediction error rate is as follows:
the prediction error value formula: (A-E)/E, the excess being positive and the excess being negative, wherein A represents the measured value and E represents the normal value.
The method for calculating the prediction error rate comprises the following steps:
a is the first measurement data, b is the second measurement data, c is the third measurement data, d is the fourth measurement data, e is the fifth measurement data
(a+b+c+d+e)/5=average value
Average/100 = percentage of average.
The model errors mentioned above refer to: in the process of establishing a mathematical model, complex phenomena are abstracted into the mathematical model, the influence of some secondary factors is often ignored, and the problems are simplified. There is therefore some error in the mathematical model and the actual problem, which is called model error.
The measurement error mentioned above means: the data used in modeling and concrete operations are often obtained by observation and measurement, and due to accuracy limitations, these data are generally approximate, i.e., have errors, which are referred to as measurement errors.
The truncation error mentioned above means: since the actual operation can only complete finite term or finite step operation, some operations which need to be performed by a limit or an infinite process are limited, and the infinite process is truncated, so that the generated error becomes a truncated error.
The rounding errors mentioned above refer to: in the numerical calculation process, some numbers are often rounded off due to the limitation of calculation tools, and only the first few digits are reserved as approximations of the numbers, and such rounding-off errors become rounding errors.
Through any one or more errors of the calculation, the regression analysis prediction model with the minimum error and the prediction index corresponding to the regression analysis prediction model are screened, for example, the corresponding regression analysis prediction model and the prediction index of holiday factors corresponding to the regression analysis prediction model can be screened according to the piece quantity prediction error value and the piece quantity prediction error rate, and for example, the corresponding regression analysis prediction model and the prediction index of recent factors corresponding to the regression analysis prediction model can be screened according to the model error, so that in practical application, the optimal regression analysis prediction model can be screened according to the designed target error, and the optimal prediction index for predicting the express piece quantity can be screened, so that the prediction accuracy of the piece quantity is improved through the screened prediction index.
According to the prediction index screening method provided by the embodiment, various prediction indexes are designed, regression analysis prediction models between the prediction indexes and the quantity indexes are built, and the prediction indexes contained in the optimal regression analysis prediction models are obtained through screening to serve as the prediction indexes for predicting the quantity of the express items.
Embodiment two:
based on the first embodiment, the present embodiment provides a predictor screening device for predicting the quantity of express items, and a schematic diagram thereof is shown in fig. 3, and specifically includes a data processing module 100, a predictor design module 200, a regression model creation module 300, and a predictor screening module 400.
Specifically, the data processing module 100 is configured to obtain historical data of the part quantity, pre-process the historical data, and screen out data information of the target part quantity. The historical data of the part quantity refers to the part quantity data stored in the logistics industry, and also can be the part quantity data in the logistics industry within a certain period of time published by a certain statistical organization. The amount includes the amount of the received piece, or may include the amount of the sent piece, or may include the amount of the received piece and the amount of the sent piece.
Preprocessing the acquired historical data, including: cleaning historical data, replacing null data and processing abnormal data; and cleaning the historical data, and removing unnecessary information and replacement abnormal data in the acquired historical data. Some irregular data typically needs to be filtered out before statistical analysis of the data can be performed to ensure accuracy of the analysis. Data cleansing is a process that reduces data errors and inconsistencies, mainly by detecting and deleting or correcting irregular data.
In this embodiment, the quantity is mainly predicted, so that the single number information and the address information contained in the history data can be removed. Among these historical data, null data or data with numerical anomalies (such as non-numerical representations) may occur, which are replaced with their neighbors.
The abnormal data processing method may be a deletion method, a substitution method (continuous variable mean substitution, discrete variable mode substitution and median substitution), an interpolation method (regression interpolation, multiple interpolation), or may be a deletion value of an abnormal value and then a deletion value interpolation operation.
The predictor design module 200 is configured to design a predictor for predicting the amount of the express item according to the target amount data information. Wherein, the prediction index includes: a plurality of recent factors, contemporaneous factors, periodic factors, holiday factors, e-commerce factors, end of month factors, temporary holiday factors, weather factors, economic factors. For the description of the recent factors, the contemporaneous factors, the periodic factors, the holiday factors, the e-commerce factors, the month end factors, the temporary holiday factors, the weather factors, and the economic factors, refer to embodiment one, and the description of this embodiment is omitted.
The regression model creation module 300 is configured to create regression analysis prediction models between the prediction indexes and the quantity indexes, respectively.
Specifically, the regression model creation module 300 includes a screening unit 301 and a creation unit 302, where the screening unit 301 is configured to screen an independent variable for predicting a part quantity from the prediction indexes, and the creation unit 302 is configured to set the part quantity as a dependent variable, and to respectively create a regression analysis prediction model between each prediction index and the part quantity index according to the screened independent variable.
Preferably, the filtering unit 301 further includes a calculating unit 3011, a first filtering unit 3012, a fitting unit 3013, and a second filtering unit 3014; wherein the calculating unit 301 is configured to calculate a correlation between the predictors; the first screening unit 3012 is used for screening out indexes with relatively smaller absolute values of correlation coefficients among indexes and relatively larger correlation coefficients between indexes and the piece quantity; the fitting unit 3013 is used for fitting the relation between the screened indexes and the part quantity and establishing a regression model; the second screening unit is used for screening indexes with P value smaller than 0.05 or smaller than 0.01 in the regression model as independent variables of the predicted piece quantity.
The prediction index screening module 400 is configured to perform component amount prediction by using various regression analysis prediction models, calculate error information of component amount prediction results of each regression partition prediction model according to actual component amount data information, and screen a prediction index included in the regression analysis prediction model with the minimum error as a prediction index for predicting the express component amount.
Further, calculating error information of the component quantity prediction result of each regression analysis prediction model according to the actual component quantity data information, wherein the error information specifically comprises any one or more of error information of a calculated component quantity prediction error value, a component quantity prediction error rate, a model error, a measurement error, a truncation error and a rounding error; for the description of the component amount prediction error value, the component amount prediction error rate, the model error, the measurement error, the truncation error and the rounding error, refer to the first embodiment, and the description of this embodiment is omitted.
The prediction index screening module 400 screens the regression analysis prediction model with the smallest error and the prediction index corresponding to the regression analysis prediction model according to any one or more calculated errors, for example, the regression analysis prediction model corresponding to the regression analysis prediction model and the prediction index of holiday factors corresponding to the regression analysis prediction model can be screened according to the quantity prediction error value and the quantity prediction error rate, and for example, the regression analysis prediction model corresponding to the regression analysis prediction model and the prediction index of recent factors corresponding to the regression analysis prediction model can be screened according to the model error, so that in practical application, the optimal regression analysis prediction model can be screened according to the designed target error, and the optimal prediction index for predicting the quantity of the express delivery can be screened, so that the prediction accuracy of the quantity of the express delivery can be improved through the screened prediction index.
According to the prediction index screening device provided by the embodiment, various prediction indexes are designed, regression analysis prediction models between the prediction indexes and the quantity indexes are built, and the prediction indexes contained in the optimal regression analysis prediction models are obtained through screening to serve as the prediction indexes for predicting the quantity of the express items.
Embodiment III:
based on the first embodiment and the second embodiment, the present embodiment provides a prediction index screening device for predicting the amount of an express delivery, and the schematic diagram of the device is shown in fig. 4, and the device 500 may be a tablet computer, a notebook computer or a desktop computer. The device 500 may also be referred to by other names as portable terminal, laptop terminal, desktop terminal, etc.
In general, device 500 includes a processor 5001 and a memory 5002, where processor 5001 may include one or more processing cores, such as a 4-core processor, an 8-core processor, and the like. The processor 5001 may be implemented in hardware as at least one of DSP (Digital Signal Processing ), FPGA (Field-Programmable Gate Array, field programmable gate array), PLA (Programmable Logic Array ). The processor 5001 may also include a main processor, which is a processor for processing data in an awake state, also referred to as a CPU (Central Processing Unit ); a coprocessor is a low-power processor for processing data in a standby state.
In some embodiments, the processor 5001 may integrate a GPU (Graphics Processing Unit, image processor) for taking care of rendering and rendering of content required for display by the display screen. In some embodiments, the processor 3001 may also include an AI (Artificial Intelligence ) processor for processing computing operations related to machine learning.
The memory 5002 can include one or more computer-readable storage media, which can be non-transitory. Memory 5002 may also include high-speed random access memory, as well as nonvolatile memory, such as one or more magnetic disk storage devices, flash memory storage devices. In some embodiments, a non-transitory computer readable storage medium in the memory 5002 is configured to store at least one instruction, at least one program, a set of codes, or a set of instructions for execution by the processor 5001 to implement the predictor screening method provided by the first embodiment of the present application.
Thus, the apparatus 500 of the present application performs the piece-rate prediction method provided in the first embodiment through at least one instruction, at least one program, a code set, or an instruction set, and has the advantages that: by designing various predictors, establishing a regression analysis prediction model between each predictor and a piece quantity index, and obtaining a prediction instruction contained in the optimal regression analysis prediction model through screening as a predictor for predicting the quantity of the express, compared with the existing blind selection predictor, the predictor screening method provided by the application can reasonably screen out the optimal predictor, so that the accuracy of the prediction of the quantity of the express can be improved according to the screened predictor, and further, the quantity of the express can be accurately predicted to provide a powerful data basis for orderly development of logistics work, for example, preparation of workers and vehicles can be made in advance based on the predicted piece quantity, thereby achieving the effects of reducing cost and loss.
In some embodiments, the apparatus 500 may further optionally include: a peripheral interface 5003, and at least one peripheral. The processor 5001, the memory 5002, and the peripheral interface 5003 may be connected by bus or signal lines. The individual peripheral devices may be connected to the peripheral device interface 5003 by buses, signal lines, or circuit boards.
In this embodiment, in order to implement the method for screening the prediction index, the corresponding peripheral device includes a database 5004, and further, the processor 5001 may obtain the historical information of the part quantity through the database 5004, and the processor 5001 obtains the data information of the target part quantity after processing the historical information of the part quantity, designs the prediction index according to the data information of the target part quantity, establishes a regression analysis prediction model, and screens the optimal prediction index for predicting the express quantity.
The present application also provides a computer readable storage medium, which may be a non-volatile computer readable storage medium, and which may also be a volatile computer readable storage medium. The computer-readable storage medium has stored therein instructions that, when executed on a computer, cause the computer to perform the predictor screening method of the first embodiment.
The system of the second embodiment may be stored in a computer readable storage medium if implemented in the form of software functional modules and sold or used as a stand-alone product. Based on such understanding, the technical solution of the present application may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-only memory (ROM), a random access memory (Random access memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The foregoing description of the application has been presented for purposes of illustration and description, and is not intended to be limiting. Several simple deductions, modifications or substitutions may also be made by a person skilled in the art to which the application pertains, based on the idea of the application.
Claims (6)
1. A prediction index screening method for predicting the quantity of express items is characterized by comprising the following steps:
acquiring historical data of the quantity of the parts, preprocessing the historical data and screening out data information of the quantity of the target parts;
designing a prediction index for predicting the quantity of the express delivery according to the target quantity data information;
respectively establishing regression analysis prediction models between each prediction index and the piece quantity index;
the method comprises the steps of respectively carrying out piece quantity prediction by adopting various regression analysis prediction models, calculating error information of piece quantity prediction results of the regression analysis prediction models according to actual piece quantity data information, and screening prediction indexes contained in the regression analysis prediction models with minimum errors as prediction indexes for predicting express piece quantity;
the prediction index comprises: a plurality of prediction indexes in recent factors, contemporaneous factors, periodic factors, holiday factors, electronic commerce factors, month end factors, temporary holiday factors, weather factors and economic factors;
the establishing a regression analysis prediction model between each prediction index and the piece quantity index specifically comprises the following steps:
screening independent variables for predicting the quantity of the piece from each prediction index;
taking the piece quantity as a dependent variable, and respectively establishing regression analysis prediction models between each prediction index and the piece quantity index according to the screened independent variable;
the method for screening the independent variable for predicting the quantity of the piece from each prediction index specifically comprises the following steps:
calculating the correlation among the prediction indexes;
screening out indexes with relatively smaller absolute values of correlation coefficients among indexes and relatively larger correlation coefficients of indexes and the piece quantity;
fitting the relation between the screened indexes and the piece quantity, and establishing a regression model;
and screening out indexes with P value smaller than 0.05 or smaller than 0.01 in the regression model as independent variables of the predicted piece quantity.
2. The predictor screening method of claim 1, wherein calculating the correlation between predictors comprises calculating pearson correlation coefficients between predictors.
3. The predictor screening method of claim 1, wherein the preprocessing the historical data comprises: cleaning historical data, replacing null data and processing abnormal data.
4. A prediction index sieving mechanism for predicting express delivery volume, its characterized in that includes:
the data processing module is used for acquiring historical data of the quantity of the parts, preprocessing the historical data and screening out data information of the quantity of the target parts;
the prediction index design module is used for designing a prediction index for predicting the express delivery quantity according to the target quantity data information;
the regression model creation module is used for respectively creating regression analysis prediction models between each prediction index and the quantity index;
the prediction index screening module is used for respectively carrying out part quantity prediction by adopting various regression analysis prediction models, calculating error information of part quantity prediction results of the regression analysis prediction models according to actual part quantity data information, and screening a prediction index contained in the regression analysis prediction model with the minimum error as a prediction index for predicting the express part quantity; the prediction index comprises: a plurality of prediction indexes in recent factors, contemporaneous factors, periodic factors, holiday factors, electronic commerce factors, month end factors, temporary holiday factors, weather factors and economic factors;
the regression model creation model comprises a screening unit and a creation unit;
the screening unit is used for screening independent variables for predicting the quantity of the piece from all the prediction indexes;
the creating unit is used for taking the piece quantity as a dependent variable and respectively creating regression analysis prediction models between each prediction index and the piece quantity index according to the screened independent variable;
calculating the correlation among the prediction indexes;
screening out indexes with relatively smaller absolute values of correlation coefficients among indexes and relatively larger correlation coefficients of indexes and the piece quantity;
fitting the relation between the screened indexes and the piece quantity, and establishing a regression model;
and screening out indexes with P value smaller than 0.05 or smaller than 0.01 in the regression model as independent variables of the predicted piece quantity.
5. A predictor screening device for predicting the amount of an express delivery, comprising a processor and a memory, wherein the memory stores at least one instruction, at least one program, a set of codes, or a set of instructions, the at least one instruction, the at least one program, the set of codes, or the set of instructions being loaded and executed by the processor to implement the predictor screening method of any one of claims 1 to 3.
6. A computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor implements the predictor screening method of any one of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110086137.3A CN112734340B (en) | 2021-01-21 | 2021-01-21 | Method, device, equipment and storage medium for screening prediction index of express delivery quantity |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110086137.3A CN112734340B (en) | 2021-01-21 | 2021-01-21 | Method, device, equipment and storage medium for screening prediction index of express delivery quantity |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112734340A CN112734340A (en) | 2021-04-30 |
| CN112734340B true CN112734340B (en) | 2023-09-01 |
Family
ID=75595044
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110086137.3A Active CN112734340B (en) | 2021-01-21 | 2021-01-21 | Method, device, equipment and storage medium for screening prediction index of express delivery quantity |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112734340B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105488595A (en) * | 2015-12-10 | 2016-04-13 | 四川省电力公司供电服务中心 | Leading index construction method for monthly power consumption prediction model |
| CN108053242A (en) * | 2017-12-12 | 2018-05-18 | 携程旅游信息技术（上海）有限公司 | Sight spot admission ticket ticket amount Forecasting Methodology, system, equipment and storage medium |
| CN108805343A (en) * | 2018-05-29 | 2018-11-13 | 祝恩元 | A kind of Scientech Service Development horizontal forecast method based on multiple linear regression |
| CN109657831A (en) * | 2017-10-11 | 2019-04-19 | 顺丰科技有限公司 | A kind of Traffic prediction method, apparatus, equipment, storage medium |
| CN109858674A (en) * | 2018-12-27 | 2019-06-07 | 国网浙江省电力有限公司 | Monthly load forecasting method based on XGBoost algorithm |
| JP2020140521A (en) * | 2019-02-28 | 2020-09-03 | 富士通株式会社 | Human determination prediction device, prediction program and prediction method |
| CN112070284A (en) * | 2020-08-24 | 2020-12-11 | 上海东普信息科技有限公司 | Screening method, device, equipment and storage medium for component prediction |
| CN112116166A (en) * | 2020-09-28 | 2020-12-22 | 中国建设银行股份有限公司 | Credit risk index prediction method and device |
| CN112183827A (en) * | 2020-09-15 | 2021-01-05 | 上海东普信息科技有限公司 | Method, device, equipment and storage medium for predicting express monthly pickup quantity |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110004510A1 (en) * | 2009-07-01 | 2011-01-06 | Arash Bateni | Causal product demand forecasting system and method using weather data as causal factors in retail demand forecasting |
-
2021
- 2021-01-21 CN CN202110086137.3A patent/CN112734340B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105488595A (en) * | 2015-12-10 | 2016-04-13 | 四川省电力公司供电服务中心 | Leading index construction method for monthly power consumption prediction model |
| CN109657831A (en) * | 2017-10-11 | 2019-04-19 | 顺丰科技有限公司 | A kind of Traffic prediction method, apparatus, equipment, storage medium |
| CN108053242A (en) * | 2017-12-12 | 2018-05-18 | 携程旅游信息技术（上海）有限公司 | Sight spot admission ticket ticket amount Forecasting Methodology, system, equipment and storage medium |
| CN108805343A (en) * | 2018-05-29 | 2018-11-13 | 祝恩元 | A kind of Scientech Service Development horizontal forecast method based on multiple linear regression |
| CN109858674A (en) * | 2018-12-27 | 2019-06-07 | 国网浙江省电力有限公司 | Monthly load forecasting method based on XGBoost algorithm |
| JP2020140521A (en) * | 2019-02-28 | 2020-09-03 | 富士通株式会社 | Human determination prediction device, prediction program and prediction method |
| CN112070284A (en) * | 2020-08-24 | 2020-12-11 | 上海东普信息科技有限公司 | Screening method, device, equipment and storage medium for component prediction |
| CN112183827A (en) * | 2020-09-15 | 2021-01-05 | 上海东普信息科技有限公司 | Method, device, equipment and storage medium for predicting express monthly pickup quantity |
| CN112116166A (en) * | 2020-09-28 | 2020-12-22 | 中国建设银行股份有限公司 | Credit risk index prediction method and device |
Non-Patent Citations (1)
| Title |
|---|
| 李野 等.基于相关性分析的宏观经济指标预测算法.指挥信息系统与技术.2020,第11卷(第1期),第84-88、100. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112734340A (en) | 2021-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110866786A (en) | Goods quantity prediction method and device, electronic equipment and storage medium | |
| CN112785057B (en) | Method, device, equipment and storage medium for predicting quantity of parts based on exponential smoothing | |
| CN110264270B (en) | Behavior prediction method, behavior prediction device, behavior prediction equipment and storage medium | |
| CN112070284A (en) | Screening method, device, equipment and storage medium for component prediction | |
| CN110363652A (en) | A kind of financial product pricing method, device and electronic equipment based on Price Sensitive model | |
| CN114170002A (en) | Method and device for predicting access frequency | |
| JP2023554201A (en) | Visual complexity slider for process graphs | |
| CN112734340B (en) | Method, device, equipment and storage medium for screening prediction index of express delivery quantity | |
| CN112183827A (en) | Method, device, equipment and storage medium for predicting express monthly pickup quantity | |
| CN116757318A (en) | Screening method and device of time sequence prediction model | |
| Stanula et al. | Economic assessment of stress-based payment models | |
| US11127079B2 (en) | Computer simulated monetary allocation using network of operator data objects | |
| AU2017355658A1 (en) | Quantification for investment vehicle management and insurance process management | |
| US20110270589A1 (en) | Determining ill conditioning in square linear system of equations | |
| CN113129127A (en) | Early warning method and device | |
| CN112801358A (en) | Component prediction method, device, equipment and storage medium based on model fusion | |
| CN112365070B (en) | Power load prediction method, device, equipment and readable storage medium | |
| CN110019172A (en) | Data processing method, device, storage medium and electronic equipment | |
| CN114092265B (en) | Method, device and storage medium for improving insurance policy new service value determination efficiency | |
| CN115858291A (en) | System index detection method and device, electronic equipment and storage medium thereof | |
| CN115375037A (en) | Demand prediction method and device | |
| CN113011986A (en) | Abstract method and system for insurance calculation | |
| CN115545731A (en) | Medical service project cost prediction method and device based on key factors | |
| CN113393279A (en) | Order quantity estimation method and system | |
| CN115358869A (en) | Financial product recommendation method and device, storage medium and equipment |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |