CN112070292A - Quantity prediction method, device, equipment and storage medium - Google Patents

Abstract

The invention discloses a method, a device, equipment and a storage medium for predicting quantity, aiming at the problems that the quantity of an express delivery is predicted mainly by adopting a manual prediction or rough method in the current logistics industry, the difference between the quantity of the express delivery and the actual quantity is large, and the predicted quantity is inaccurate, the historical data of the quantity is processed to obtain test data suitable for creating a double-index smooth model, the quantity is predicted in a short term by utilizing the double-index smooth model, the accuracy of quantity prediction is improved, a powerful data basis is provided for the orderly development of logistics work, and therefore the work efficiency of logistics enterprises is improved.

Description

Quantity prediction method, device, equipment and storage medium

technical field

The invention belongs to the technical field of business volume forecasting, and in particular relates to a piece volume forecasting method, device, equipment and storage medium.

Background technique

Prediction is the core application of big data, and big data prediction expands the traditional meaning of prediction to "existing measurement". The advantage of big data forecasting is that it transforms a very difficult forecasting problem into a relatively simple description problem, which is simply beyond the reach of traditional small data sets. From the perspective of forecasting, the results obtained from big data forecasts are not only simple and objective conclusions for dealing with real business, but can also be used to help enterprises make business decisions. The collected data can also be planned to guide the development of greater consumption power .

Time series data mining takes the data formed by the state of things at different times as the research object. By analyzing and researching the characteristics of time series data, it reveals the law of development and change of things, which is used to guide people's social, economic, military and life. and other activities. Time series mining is of great significance to the development of human society, technology and economy, and is gradually becoming one of the research hotspots of data mining.

With the rapid development of the logistics industry, the management and control of the business volume (express delivery volume) is related to the normal operation of the logistics company's business. Therefore, it is particularly important to predict the number of pieces.

For the problem of piece volume forecasting in the field of logistics, the piece volume always changes with time. At present, the industry mainly adopts manual forecasting or rough methods to predict the delivery volume of express delivery, which is quite different from the actual piece volume, and the predicted piece volume is inaccurate. , which is not conducive to the business development of the company.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a piece quantity forecasting method, device, equipment and storage medium, which adopts a time series forecasting model to improve the accuracy of piece quantity forecasting.

For solving the above problems, the technical scheme of the present invention is:

A piece quantity forecasting method, comprising:

Step S1: obtaining historical data of the amount of pieces, preprocessing the historical data, and selecting a weekly target data set of at least one historical period;

Step S2: Adjust the weekly target data set to eliminate the weekly variation trend;

Step S3: performing a data stationarity check on the weekly target data set for which the change trend has been eliminated to obtain a stable weekly target data set;

Step S4: Based on the stable weekly target data set, a double exponential smoothing model is created to predict the piece volume, and output the weekly piece volume forecast value.

According to an embodiment of the present invention, the step S1 further includes:

Clean historical data, replace empty data and abnormal data;

Analyze the weekly change trend of historical data to obtain a weekly change data set;

Data smoothing is performed on the weekly variation data set to obtain the weekly target data set.

According to an embodiment of the present invention, the step S2 further includes:

Use the following calculation formula to adjust the data in the weekly target data set to eliminate the weekly variation trend:

Among them, a _i,j is the piece quantity value of the jth day in the i-th week, i is a positive integer greater than 1, j=1,2,3,4,5,6,7; a _i-1,j is The piece quantity value of the jth day in the i-1th week; s _i-1 is the sum of the piece quantity value of the 7th day of the i-1th week.

According to an embodiment of the present invention, the step S3 further includes:

Use time series plots or autocorrelation plots to detect data stationarity of weekly target datasets;

Adjust the unstable data to obtain a stable weekly target data set.

According to an embodiment of the present invention, the step S4 further includes:

The double exponential smoothing model is:

Y _{t +T} = at + b _t ·T

为t期的一次指数平滑值；

分别为t期和t-1期的二次指数平滑值；a为平滑系数；Y_t+T为t+T期的预测值，T为由t期向后推移的期数。in,

is an exponential smoothing value for the t period;

are the quadratic exponential smoothing values of the t period and the t-1 period respectively; a is the smoothing coefficient; Y _t+T is the predicted value of the t+T period, and T is the number of periods shifted backward from the t period.

According to an embodiment of the present invention, after step S4, the method further includes: comparing the predicted value of the weekly piece quantity with the actual value of the piece quantity, calculating the error; and adjusting the parameters of the double exponential smoothing model according to the error.

A piece quantity prediction device, comprising:

The data preprocessing module is used to obtain the historical data of the piece volume, preprocess the historical data, and select the weekly target data set of at least one historical period;

Trend removal module, used to adjust the weekly target data set to eliminate the weekly trend:

The stationarity detection module is used to check the data stationarity of the weekly target data set that eliminates the changing trend, and obtain a stable weekly target data set;

The model creation module is used to create a double exponential smoothing model based on the stable weekly target data set to predict the piece volume, and output the predicted value of the weekly piece volume.

According to an embodiment of the present invention, the piece quantity prediction device further includes: a model checking module for comparing the size of the weekly piece quantity prediction value and the actual piece quantity value, calculating the error; and adjusting the parameters of the double exponential smoothing model according to the error.

A piece quantity forecasting device, comprising:

a memory and at least one processor with instructions stored in the memory, the memory and the at least one processor interconnected by wires;

The at least one processor invokes the instructions in the memory, so that the piece quantity prediction device executes the piece quantity prediction method in an embodiment of the present invention.

A computer-readable storage medium stores a computer program on the computer-readable storage medium, and when the computer program is executed by a processor, implements the method for predicting a piece quantity in an embodiment of the present invention.

Compared with the prior art, the present invention has the following advantages and positive effects due to the adoption of the above technical solutions:

The method for predicting the quantity of pieces in an embodiment of the present invention mainly adopts manual forecasting or rough methods to predict the quantity of express delivery in the current logistics industry, which is quite different from the actual quantity of pieces, and the predicted piece quantity is inaccurate. It can process the historical data of the quantity to obtain the test data suitable for creating the double exponential smoothing model, and use the double exponential smoothing model to make short-term forecast of the piece volume, improve the accuracy of the piece volume forecast, and provide a strong data foundation for the orderly development of logistics work. , so as to improve the efficiency of logistics enterprises.

Description of drawings

FIG. 1 is a flow chart of a method for predicting a piece quantity in an embodiment of the present invention;

Fig. 2 is the trend curve diagram of elimination week in an embodiment of the present invention;

FIG. 3 is a block diagram of an apparatus for predicting a piece quantity in an embodiment of the present invention;

FIG. 4 is a schematic diagram of a piece quantity prediction device in an embodiment of the present invention.

Detailed ways

A method, device, device and storage medium for piece quantity prediction proposed by the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The advantages and features of the present invention will become apparent from the following description and claims.

Example 1

Referring to FIG. 1, the method for predicting the quantity of pieces in this embodiment includes:

Step S1: Acquire historical data of the piece volume, preprocess the historical data, and select a weekly target data set of at least one historical period.

In this embodiment, the historical piece volume data refers to piece volume data stored in the logistics industry, and may also be piece volume data in the logistics industry within a certain period of time published by a statistical agency. The number of pieces includes the volume of incoming mail, and it can also include the amount of outgoing mail. In the database, whether it is online or offline, the information on the number of deliveries and receipts will be stored. The information may include, but is not limited to, the type and time of the piece quantity. The time can be stored by day, by week, or by the specific time entered into the system.

Preprocessing the acquired historical data, including: cleaning the historical data, replacing empty data and abnormal data; analyzing the weekly change trend of the historical data to obtain the weekly change data set; performing data smoothing on the weekly change data set to obtain the weekly target data set.

Among them, the historical data is cleaned, unnecessary information in the acquired historical data is removed, and abnormal data is replaced. Usually, before performing statistical analysis on the data, some irregular data needs to be filtered out to ensure the accuracy of the analysis. Data cleaning is a process of reducing data errors and inconsistencies, mainly by detecting and removing or correcting irregular data.

In this embodiment, the prediction is mainly based on the quantity of pieces, so the tracking number information and address information contained in the historical data can be removed. In these historical data, there may be empty data or data with abnormal numerical value (such as non-numeric representation), and these empty data or data with abnormal numerical value are replaced with their adjacent data.

Specifically, the historical data includes the volume of receipts and/or the volume of deliveries, and information on the volume of receipts (with or without orders) and delivery volumes at each outlet can be retrieved from the database according to different business scenarios. The receipt volume of the outlets is the test data, and the date of the historical data is 2017/07/01-2017/07/30. After data cleaning, the obtained historical data can be shown in Table 1 below.

When smoothing the historical data, it is necessary to analyze the change trend of the historical data to obtain a data set with a change trend, such as a weekly change data set with a weekly change trend.

Smoothing the weekly variation data set can reduce the influence of statistical errors on the forecast results of the piece volume, and obtain the weekly target data set. The data can be smoothed in the following ways:

weighted moving average

The basic principle of the weighted moving average method is: the weight of the central data in the averaged interval is the largest, and the weight of the data farther from the center is smaller. The weight coefficient can adopt the principle of least squares, so that the smoothed data approximates the original data with the minimum mean square error.

smooth function smoothing method

Call function: Z=smooth(Y, span, method), where Z represents the smoothed data vector, Y represents the smoothed original data vector, span represents the number of smoothed points, and method represents the smoothing method (including moving moving average, lowess linearity) weighted smoothing, loess quadratic weighted smoothing, etc.).

The above exemplifies two data smoothing methods. Of course, the smoothing of data is not limited to these two methods, and other methods can also be used, such as using Gaussian function to smooth data or SG filtering method, etc. No introduction.

Step S2: Adjust the weekly target data set to eliminate the weekly variation trend.

Specifically, the weekly target data sets are arranged in time series, and a curve is drawn, such as curve a in Figure 2 (a curve formed by connecting dots). In order to improve the accuracy of the data, the value of the piece size is normalized. As can be seen from curve a, the weekly change trend is from Monday (2017/7/4) to Sunday (2017/7/10), and the number of pieces shows an obvious downward trend. This downward trend is eliminated for the weekly target data set, and the curve a is as flat as possible, as shown in the curve b (curve formed by connecting square points) in Figure 2. Compared with the curve a, the change trend of the curve b is obviously reduced, and the overall characteristic is stable.

The method for eliminating the weekly variation trend in this embodiment is: adjusting the weekly target data set through the following calculation formula, so that the data tends to be stable.

Among them, a _i,j is the piece quantity value of the jth day in the i-th week, i is a positive integer greater than 1, j=1,2,3,4,5,6,7; a _i-1,j is The piece quantity value of the jth day in the i-1th week; s _i-1 is the sum of the piece quantity value of the 7th day of the i-1th week. According to the above calculation formula, the values in the weekly target data set of the piece quantity are adjusted one by one. The curve converted from the final weekly target data set is shown as curve b in Figure 2.

Step S3: Perform a data stationarity check on the weekly target data set whose change trend has been eliminated to obtain a stable weekly target data set.

This embodiment provides two methods for detecting the stationarity of data, namely, time sequence graph detection and autocorrelation graph detection.

Among them, the time series diagram detection follows the principle that the mean value and variance of the stationary time series are constant. The stationary series in the time series diagram shows the characteristics of random fluctuation around a constant value, and the fluctuation range is limited, and there is no obvious trend or periodicity. If a sequence shows an obvious trend or periodicity in the time series diagram, it means that the sequence is not stationary, not stationary.

In autocorrelation graph detection, as the delay period k increases, the autocorrelation coefficient p of the stationary time series decays rapidly to zero. The autocorrelation coefficients of non-stationary series usually decay slowly towards zero. Typical autocorrelation diagram of non-stationary series: The autocorrelation diagram shows obvious triangular symmetry; it is located on the side of the zero axis, which has the typical characteristics of monotonic trend series, or has obvious sinusoidal fluctuation law.

According to the principle of the above time sequence diagram and autocorrelation diagram, the weekly target data set is subjected to time sequence diagram detection or autocorrelation diagram detection to judge the data stationarity. If a non-stationary sequence is found, adjust the non-stationary sequence (such as a linearly growing trend, a new stationary (trend-eliminating) time series can be formed by first-order difference) until the sequence in the weekly target data set tends to be stationary.

Step S4: Based on the stable weekly target data set, a double exponential smoothing model is created to predict the piece volume, and output the weekly piece volume forecast value.

Exponential smoothing is a method of time series forecasting for univariate data, including one-time exponential smoothing, second-order exponential smoothing, etc. Among them, the formula of the linear quadratic exponential smoothing method is:

分别为t期和t–1期的二次指数平滑值；a为平滑系数。在

和

已知的条件下，二次指数平滑法的预测模型(即双指数平滑模型)为：where:

are the quadratic exponential smoothing values for periods t and t–1, respectively; a is the smoothing coefficient. exist

and

Under the known conditions, the prediction model of the double exponential smoothing method (that is, the double exponential smoothing model) is:

Y _{t +T} = at + b _t ·T

In the formula: Y _t+T is the predicted value of the period t+T, and T is the number of periods shifted from the period t to the back.

Create a double exponential smoothing model according to the above formula, and perform logarithm, exponent, square, difference, and integral processing of the weekly target data set, and five curves are obtained as the data basis of the double exponential smoothing model.

Then, select a set of values of model parameters including historical data duration, model prediction duration, and data starting point, such as historical data duration=7 days, model prediction duration=1 day, data starting point=monthly in the weekly target data set 1 day's data value. These parameters are written into the double exponential smoothing model, and the piece quantity is forecasted to obtain a piece quantity forecast value.

In order to make the obtained prediction results more accurate, different model parameters can be used to perform multiple piece quantity predictions. The following model parameters for one-time volume forecasting are desirable: historical data duration = 15 days, model prediction duration = 3 days, data starting point = data value on the 5th day of each month in the weekly target data set, and write these parameters into the double exponential sliding model Carry out a piece quantity forecast to obtain another piece quantity forecast value.

In this way, a plurality of piece quantity prediction values can be obtained, and each piece quantity forecast value is compared with the actual piece quantity value, and the error is calculated; and the parameters of the double exponential smoothing model are adjusted according to the error.

The piece quantity forecasting method in this embodiment mainly adopts manual forecasting or rough method to predict the quantity of express delivery in the current logistics industry, which is quite different from the actual piece quantity, and the predicted piece quantity is inaccurate. The historical data is processed to obtain test data suitable for creating a double exponential smoothing model, and the double exponential smoothing model is used to make short-term forecasts on the quantity of pieces, so as to improve the accuracy of the forecast of the quantity of pieces, and provide a strong data foundation for the orderly development of logistics work. Improve the efficiency of logistics enterprises.

Embodiment 2

The present invention also provides a piece quantity prediction device, referring to FIG. 3 , the device includes:

The data preprocessing module 1 is used to obtain the historical data of the piece volume, preprocess the historical data, and select the weekly target data set of at least one historical period;

Trend elimination module 2, which is used to adjust the weekly target data set and eliminate the weekly variation trend:

The stationarity detection module 3 is used to check the data stationarity of the weekly target data set that eliminates the change trend, and obtain a stable weekly target data set;

The model creation module 4 is used to create a double exponential smoothing model to predict the piece volume based on the stable weekly target data set, and output the predicted value of the weekly piece volume;

The model checking module 5 is used to compare the predicted value of the weekly piece quantity with the actual value of the piece quantity, calculate the error, and adjust the parameters of the double exponential smoothing model according to the error.

The specific contents and implementation methods of the above-mentioned data preprocessing module 1, trend elimination module 2, stationarity detection module 3, model creation module 4, and model verification module 5 are as described in Embodiment 1, and will not be repeated here.

Embodiment 3

The above Embodiment 2 describes in detail the piece quantity prediction apparatus of the present invention from the perspective of modular functional entities, and the following describes the piece quantity prediction apparatus of the present invention in detail from the perspective of hardware processing.

Referring to FIG. 4 , the piece quantity prediction apparatus 500 may vary greatly due to different configurations or performances, and may include one or more processors (central processing units, CPU) 510 (eg, one or more processors) and memory 520, one or more storage media 530 (eg, one or more mass storage devices) that store applications 533 or data 532. Among them, the memory 520 and the storage medium 530 may be short-term storage or persistent storage. The program stored in the storage medium 530 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations on the piece quantity prediction apparatus 500 .

Further, the processor 510 may be configured to communicate with the storage medium 530 to execute a series of instruction operations in the storage medium 530 on the piece quantity prediction device 500 .

Quantity prediction device 500 may also include one or more power supplies 540, one or more wired or wireless network interfaces 550, one or more input and output interfaces 560, and/or, one or more operating systems 531, such as Windows Server , Vista, etc.

Those skilled in the art can understand that the structure of the piece quantity prediction device shown in FIG. 4 does not constitute a limitation on the piece quantity prediction device, and may include more or less components than the one shown, or combine some components, or different Component placement.

The present invention also provides a computer-readable storage medium. The computer-readable storage medium may be a non-volatile computer-readable storage medium, and the computer-readable storage medium may also be a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium, and when the instructions are executed on a computer, the instructions cause the computer to execute the steps of the method for predicting a piece quantity in the first embodiment.

If the modules in the second embodiment are implemented in the form of software function modules and sold or used as independent products, they may be stored in a computer-readable storage medium. Based on such understanding, the technical solution of the present invention essentially or the part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of software, and the computer software is stored in a storage medium, including Several instructions are used to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: U disk, removable hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk and other media that can store program codes .

Those skilled in the art can clearly understand that, for the convenience and brevity of description, for the specific working process of the apparatus and equipment described above, reference may be made to the corresponding process in the foregoing method embodiments, which will not be repeated here.

The embodiments of the present invention have been described in detail above with reference to the accompanying drawings, but the present invention is not limited to the above-mentioned embodiments. Even if various changes are made to the present invention, if these changes fall within the scope of the claims of the present invention and the technical equivalents thereof, they still fall within the protection scope of the present invention.

Claims (10)

1. A method for predicting a quantity, comprising:

step S1: acquiring historical data of the quantity, preprocessing the historical data, and selecting a weekly target data set of at least one historical period;

step S2: adjusting a weekly target data set to eliminate a weekly variation trend;

step S3: carrying out data stability verification on the week target data set with the change trend eliminated to obtain a stable week target data set;

step S4: and based on the stable weekly target data set, creating a bi-exponential smoothing model to predict the parts, and outputting a predicted value of the weekly parts.

2. The component prediction method according to claim 1, wherein the step S1 further includes:

cleaning historical data, and replacing null data and abnormal data;

analyzing the weekly variation trend of the historical data to obtain a weekly variation data set;

and performing data smoothing treatment on the weekly change data set to obtain a weekly target data set.

3. The component prediction method according to claim 2, wherein the step S2 further includes:

and adjusting data in the weekly target data set by adopting the following calculation formula to eliminate the weekly variation trend:

wherein, a_i，jIs the dose value on day j in week i, i is a positive integer greater than 1, j is 1,2,3,4,5,6, 7; a is_i-1，jIs the dose value on day j in week i-1; s_i-17 days in i-1 weekThe sum of the component values.

4. The component prediction method according to claim 1, wherein the step S3 further includes:

detecting data stationarity of a week target data set by adopting a time sequence diagram or an autocorrelation diagram;

and adjusting the unstable data to obtain a stable weekly target data set.

5. The component prediction method according to claim 1, wherein the step S4 further includes:

the bi-exponential smoothing model is:

Y_t+T＝a_t+b_t.T

wherein S is_t ⁽¹⁾Is a first exponential smoothing value of the t period;

respectively the secondary exponential smoothing values of the t period and the t-1 period; a is a smoothing coefficient; y is_t+TIs the predicted value of T + T period, and T is the number of periods moving backwards from T period.

6. The component prediction method according to claim 1, wherein said step S4 is further followed by: comparing the predicted value of the weekly component quantity with the actual value of the component quantity, and calculating an error; and adjusting parameters of the dual-exponential smoothing model according to the error.

7. A quantity prediction apparatus, comprising:

the data preprocessing module is used for acquiring historical data of the quantity, preprocessing the historical data and selecting a weekly target data set of at least one historical period;

the trend elimination module is used for adjusting the weekly target data set and eliminating the weekly variation trend:

the stability detection module is used for carrying out data stability verification on the weekly target data set with the change trend eliminated to obtain a stable weekly target data set;

and the model creating module is used for creating a bi-exponential smoothing model to predict the parts based on the stable weekly target data set and outputting the predicted values of the weekly parts.

8. The quantity prediction device of claim 7, further comprising: the model checking module is used for comparing the predicted value of the weekly component with the actual value of the component and calculating an error; and adjusting parameters of the dual-exponential smoothing model according to the error.

9. A quantity prediction apparatus, characterized by comprising:

a memory having instructions stored therein and at least one processor, the memory and the at least one processor interconnected by a line;

the at least one processor invokes the instructions in the memory to cause the component prediction device to perform the component prediction method of any of claims 1-6.

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the method for component prediction according to any one of claims 1 to 6.

Images