CN117875468A - Logistics prediction method and system based on big data - Google Patents
Logistics prediction method and system based on big data Download PDFInfo
- Publication number
- CN117875468A CN117875468A CN202311431282.6A CN202311431282A CN117875468A CN 117875468 A CN117875468 A CN 117875468A CN 202311431282 A CN202311431282 A CN 202311431282A CN 117875468 A CN117875468 A CN 117875468A
- Authority
- CN
- China
- Prior art keywords
- logistics
- data
- prediction
- time
- decomposition
- 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
- 238000000034 method Methods 0.000 title claims abstract description 81
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 51
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 46
- 238000005457 optimization Methods 0.000 claims abstract description 38
- 241000255581 Drosophila <fruit fly, genus> Species 0.000 claims abstract description 20
- 238000012545 processing Methods 0.000 claims abstract description 15
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 125000004122 cyclic group Chemical group 0.000 claims description 28
- 238000004364 calculation method Methods 0.000 claims description 12
- 238000009826 distribution Methods 0.000 claims description 10
- 241000255588 Tephritidae Species 0.000 claims description 9
- 238000004140 cleaning Methods 0.000 claims description 9
- 235000013305 food Nutrition 0.000 claims description 8
- 238000006243 chemical reaction Methods 0.000 claims description 7
- 238000010276 construction Methods 0.000 claims description 7
- 238000000605 extraction Methods 0.000 claims description 7
- 238000013507 mapping Methods 0.000 claims description 7
- 238000005516 engineering process Methods 0.000 claims description 6
- 238000009499 grossing Methods 0.000 claims description 5
- 238000001514 detection method Methods 0.000 claims description 4
- 238000011156 evaluation Methods 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 4
- 230000002123 temporal effect Effects 0.000 claims description 2
- 241000255601 Drosophila melanogaster Species 0.000 claims 1
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 14
- 238000010586 diagram Methods 0.000 description 10
- 238000004891 communication Methods 0.000 description 8
- 238000003860 storage Methods 0.000 description 6
- 238000011161 development Methods 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 4
- 230000005236 sound signal Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 3
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- KLDZYURQCUYZBL-UHFFFAOYSA-N 2-[3-[(2-hydroxyphenyl)methylideneamino]propyliminomethyl]phenol Chemical compound OC1=CC=CC=C1C=NCCCN=CC1=CC=CC=C1O KLDZYURQCUYZBL-UHFFFAOYSA-N 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000000739 chaotic effect Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 201000001098 delayed sleep phase syndrome Diseases 0.000 description 1
- 208000033921 delayed sleep phase type circadian rhythm sleep disease Diseases 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000003475 lamination Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000000241 respiratory effect Effects 0.000 description 1
- 238000000547 structure data Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a logistics prediction method and a logistics prediction system based on big data, which relate to the technical field of big data processing and comprise the steps of obtaining historical logistics data and preprocessing the historical logistics data; extracting features related to the logistics plan demand, the cost material fee control quantity and the warehouse reserve quantity from the historical logistics data according to a rough set theory algorithm, and constructing a prediction model; performing time sequence decomposition on the historical logistics data by using a signal decomposition algorithm; and carrying out parameter optimization on the prediction model based on a drosophila optimization algorithm, and predicting the decomposition data based on a tree structure Parzen super-parameter optimization algorithm to obtain a logistics prediction result. The invention has the beneficial effects that the stability carries out component decomposition on the logistics data, reduces the complexity of the logistics data and has good fault tolerance performance; the prediction accuracy of the model is improved by adopting a parameter optimization algorithm, and the prediction speed and the prediction accuracy are also higher.
Description
Technical Field
The invention relates to the technical field of big data processing, in particular to a big data-based logistics prediction method and a big data-based logistics prediction system.
Background
The physical distribution forecast is to describe and analyze the physical distribution management development trend and state according to the past and modern development rules of objective matters. With the advent of economic globalization and knowledge economic age, the rapid development of high and new technology has become more and more complex in commodity structure, and logistics prediction is the basis of logistics operation and the basis of planning and control of logistics departments, and plays a very important role in logistics activities. The method can reveal the future development trend and direction of the logistics market for logistics enterprises, and forecast various situations possibly occurring in the activities of the logistics enterprises, so that the enterprises can prevent or minimize the occurrence of adverse conditions on the development of the enterprises in time.
The logistics demand prediction based on big data is an important technical means of logistics resource scheduling, and if the actual market demand cannot be accurately predicted, the supply is insufficient or excessive, so that the inventory level and the operation cost of enterprises are affected. The logistics demand data has the characteristics of strong randomness, volatility, non-stationarity and the like, so that accurate prediction of the logistics demand data is difficult.
Disclosure of Invention
The invention aims to provide a logistics prediction method and a logistics prediction system based on big data, so as to solve the problems. In order to achieve the above purpose, the technical scheme adopted by the invention is as follows:
in a first aspect, the present application provides obtaining historical logistics data, and preprocessing the historical logistics data to obtain preprocessed logistics data, where the historical logistics data includes a logistics plan demand, a cost material fee control amount and a warehouse reserve amount;
extracting features related to the logistics plan demand, the cost material fee control quantity and the warehouse reserve quantity from the historical logistics data according to a rough set theory algorithm, and constructing a prediction model;
performing time sequence decomposition on the historical logistics data by using a signal decomposition algorithm to obtain decomposition data;
and carrying out parameter optimization on the prediction model based on a drosophila optimization algorithm to obtain an optimized parameter set, inputting the optimized parameter set into the prediction model, and predicting the decomposition data based on a tree structure Parzen super-parameter optimization algorithm to obtain a logistics prediction result.
In a second aspect, the present application further provides a big data-based logistics prediction system, including an acquisition module, an extraction module, a decomposition module, and a prediction module, where:
the acquisition module is used for: the method comprises the steps of acquiring historical logistics data, and preprocessing the historical logistics data to obtain preprocessed logistics data, wherein the historical logistics data comprises logistics plan demand, cost material fee control quantity and warehouse reserve quantity;
and an extraction module: the method comprises the steps of extracting characteristics related to the logistics plan demand, the cost material fee control quantity and the warehouse reserve quantity from the historical logistics data according to a rough set theory algorithm, and constructing a prediction model;
and a decomposition module: the method comprises the steps of performing time sequence decomposition on historical logistics data by using a signal decomposition algorithm to obtain decomposition data;
and a prediction module: the method is used for carrying out parameter optimization on the prediction model based on a drosophila optimization algorithm to obtain an optimized parameter set, inputting the optimized parameter set into the prediction model, and predicting the decomposition data based on a tree structure Parzen super-parameter optimization algorithm to obtain a logistics prediction result.
The beneficial effects of the invention are as follows: according to the invention, the logistics information is obtained by connecting the urban road with the GPS for transporting logistics, the problem of traffic jam is solved, the urban traffic efficiency is improved, and the logistics data is obtained rapidly; according to the invention, the component decomposition is carried out on the logistics data through the decomposition algorithm, so that the complexity of the logistics data is reduced, and the fault tolerance performance is good; the invention adopts the parameter optimization algorithm, so that the model can learn the time sequence characteristics of the data better, and the prediction accuracy of the model is further improved. The prediction speed and the precision are also higher, and the method is also suitable for predicting the loss in the logistics industry.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be apparent from the description, or may be learned by practice of the embodiments of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims thereof as well as the appended drawings.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a big data-based logistics prediction method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a big data based logistics prediction system according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a big data-based logistics prediction apparatus according to an embodiment of the present invention.
701, an acquisition module; 7011. an obtaining unit; 7012. a matching unit; 7013. an acquisition unit; 7014. a first processing unit; 702. an extraction module; 703. a decomposition module; 7031. a construction unit; 7032. a calculation unit; 7033. an updating unit; 704. a prediction module; 7041. adopting a unit; 7042. a search unit; 7043. an iteration unit; 7044. a first building unit; 7045. a second construction unit; 7046. a second processing unit; 7047. fitting unit; 7048. a prediction unit; 800. a big data based logistics prediction device; 801. a processor; 802. a memory; 803. a multimedia component; 804. an I/O interface; 805. a communication component.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures. Meanwhile, in the description of the present invention, the terms "first", "second", and the like are used only to distinguish the description, and are not to be construed as indicating or implying relative importance.
Example 1:
the embodiment provides a logistics prediction method based on big data.
Referring to fig. 1, the method is shown to include step S100, step S200, step S300, and step S400.
S100, acquiring historical logistics data, and preprocessing the historical logistics data to obtain preprocessed logistics data, wherein the historical logistics data comprises logistics plan demand, cost material fee control quantity and warehouse reserve quantity.
It is understood that S101, S102, S103, and S104 are included in step S100, where:
s101, marking each road in the urban road according to a parallelogram by utilizing a road network matching algorithm of a boundary rectangle, and obtaining a parallelogram area corresponding to each road;
s102, matching the parallelogram area with GPS positioning data on a logistics transportation vehicle to form matched GPS positioning data;
specifically, in this embodiment, the road network matching determination algorithm of the bounding rectangle is two vertices of the rectangle. In this embodiment, the road is identified by a parallelogram, and the position of the parallelogram is determined by storing the coordinates of its 4 vertices. By the method, GPS positioning data can be quickly matched to the road network, and efficiency and accuracy are achieved.
S103, acquiring flow information of all vehicles transported through the logistics in the required urban road according to the GPS positioning data;
specifically, by matching the GPS positioning data with the logistics transportation vehicles, it is possible to acquire which traffic information on which logistics transportation vehicle is on which road.
And S104, based on a data cleaning method, carrying out data cleaning on the flow information, setting outliers, and carrying out detection screening to obtain the pretreated logistics data.
The data cleaning method comprises the steps of discarding part of data, complementing missing data, not processing the data and a true value conversion method, so as to detect, screen and clean the data by means of tools, and clean the dirty data according to a certain rule to ensure the accuracy of the subsequent analysis result.
And S200, extracting features related to the logistics plan demand, the cost material fee control quantity and the warehouse reserve quantity from the historical logistics data according to a rough set theory algorithm, and constructing a prediction model.
It will be appreciated that in this step, extraction from the historical data is based on a rough set theory algorithm, which has the advantage that it is not necessary to provide any prior knowledge beyond the data set that the problem needs to deal with, and that there is strong complementarity (especially fuzzy theory) to the theory that deals with other uncertainty problems, and that features associated with important features can be clearly extracted from the database, so that it is convenient to build a predictive model, with better relevance.
S300, performing time sequence decomposition on the historical logistics data by using a signal decomposition algorithm to obtain decomposition data.
It will be appreciated that step S300 includes steps S301, S302 and S303, wherein:
s301, constructing a Lagrangian function based on a Gaussian smoothing filtering algorithm through a quadratic penalty function term and a Lagrangian multiplier;
calculating constraint variation problems by constructing a Lagrangian function L, and expanding the expression as follows;
alpha in the formula (1) is a quadratic penalty function term, so that the accuracy of signal reconstruction can be ensured, lambda is a Lagrangian multiplier, the strictness of constraint can be ensured,for Gaussian smoothing calculations, ω k Represents the center frequency corresponding to the kth modal component, u k Represents the kth modal component, u k (t) represents the time-domain form of the kth modal component, i.e. the time-domain modal component, delta (t) represents the dirac distribution,/o>For convolution operator>For deriving symbols, ++>Is HilbertThe signal used as convolution in the transformation, x (t) represents the time domain raw signal and t represents the instant t.
S302, converting a definition domain in the Lagrangian function from time to frequency according to the historical logistics data, and calculating an extremum of the Lagrangian function;
the extremum is calculated by the lagrangian function, and the calculation formula is as follows:
equation (3) (2), x (t) represents the original signal, t is the time t, x (t) is the original signal in the time domain, x (t) is converted from the time domain to the frequency domain by Fourier transform, and the obtained resultOmega is the non-negative centre frequency, < >>I.e. the frequency domain original signal; likewise->I.e. the ith frequency domain modal component, n represents the iteration number; lambda (t) represents Lagrangian multiplier,>for the kth component, n represents the number of iterations, < -> Is composed of->Lambda (t) is obtained by Fourier transform; alpha is a quadratic penalty function term, +.>To take an infinite integral about ω, where dω is the derivative of ω.
S303, updating Lagrange multipliers, carrying out iterative computation, and if the computation result is smaller than a penalty factor, ending the computation to obtain decomposition data; otherwise, the domain in the Lagrangian function is converted from time to frequency until the penalty factor is larger, and the calculation is finished.
Updating the lagrangian multiplier λ:
lambda in formula (4) n+1 (ω) represents the frequency domain Lagrangian multiplier, n+1 represents the number of iterations, pi is the circumference ratio, ω is the non-negative center frequency, +.,to take infinite integral about ω, x (ω) represents the frequency domain raw signal and dω is differentiated for ω.
Specifically, when the penalty factor is smaller than the calculation is ended; otherwise, continuing to execute the step (1) until the decomposed data are obtained.
S400, carrying out parameter optimization on the prediction model based on a drosophila optimization algorithm to obtain an optimized parameter set, inputting the optimized parameter set into the prediction model, and predicting the decomposition data based on a tree structure Parzen super-parameter optimization algorithm to obtain a logistics prediction result.
It will be appreciated that the drosophila optimization algorithm in this step includes S401, S402 and S403, wherein:
s401, adopting a chaos technology based on Logistic mapping to enable the initial coordinate distribution of the drosophila to be uniform;
s402, randomly initializing the position of the Drosophila population, and searching the random direction and distance of food;
s403, calculating a taste concentration determination value, making fitness evaluation, performing iterative optimization, and determining an optimal solution; wherein, calculating the taste concentration determination value comprises obtaining the coordinate points of the highest, the smallest and the best fruit flies of the food smell concentration value, and marking the coordinate points as the searching position and the direction of the fruit flies of the next generation to provide guidance.
Specifically, the basic idea of the chaotic technology is: mapping the variable to be processed into a chaos variable value space, and optimizing the coordinates of the drosophila cluster by utilizing the coordinate data of the chaos variable value interval and combining the searching step length, the searching direction and the like of the drosophila. The Logistic mapping iteration equation is as follows:
x di+1 ＝μx di (1-x di )
wherein: x is x di For coordinates before iteration, x di+1 For coordinates after one iteration, μ is the control parameter. As the μ value increases, the distribution of the sequences becomes more and more uniform, and when μ is equal to 4, the distribution of the Logistic map is most uniform. The chaos value after chaos mapping is converted into a variable space of the population, so that the initial position distribution of the drosophila population is more uniform. And calculating the coordinate points of the highest, the smallest and the optimal fruit flies according to the odor concentration judgment value formula. And carrying out parameter optimization on the prediction model by adopting the method to obtain an optimized parameter set. Compared with other methods, the method has higher precision and improves the accuracy.
It will be appreciated that step S400 further includes S404, S405, S406, S407, and S408, wherein:
s404, constructing a convolution model, wherein the convolution model comprises at least three space-time cyclic convolution blocks, and each space-time cyclic convolution block comprises a time convolution block and a space convolution block;
wherein, the "space diagram convolution block" is a diagram convolution to obtain a space characteristic. Such a structure is called a space-diagram convolution block. The purpose is to extract the corresponding features from the spatial dimension by means of graph convolution. The graph convolution refers to a graph convolution neural network, namely a convolution neural network which is expanded into a graph structure, and a feedforward neural network which can process graph structure data and comprises convolution calculation and has a depth structure, namely a special convolution neural network.
S405, stacking the space-time cyclic convolution blocks to construct a dynamic space-time convolution block, wherein the dynamic space-time convolution block comprises 2 time layers and 1 space layer connected with a time domain;
and stacking according to the 2 space-time cyclic convolution blocks to construct a dynamic space-time diagram convolution block, wherein a space layer is a bridge connecting 2 time domains, so that transition between diagram convolution and time cyclic convolution is realized, and quick propagation in a space state is facilitated.
S406, lifting the corresponding output dimension of each space-time cyclic convolution block by using a bottleneck theory and using a graph roll lamination layer;
the dimension of the channel c is lifted by using the graph roll stacking layer, so that the characteristic compression and the scale compression of the data are realized.
S407, performing fitting alleviation on the space-time cyclic convolution block after lifting treatment based on a tree structure Parzen super-parameter optimization algorithm and a random discarding method;
in each space-time cyclic convolution block, a random discard process is used after each layer in order to avoid excessive fitting. The accuracy of the space-time cyclic convolution block is improved through a tree structure Parzen super-parameter optimization algorithm; after superimposing 2 dynamic space-time diagram convolution blocks, an output layer is added, which is a time-cyclic convolution layer added with a full connection layer. The output layer maps the output of the last dynamic space-time diagram convolution block with the single-step prediction, so that the final output Z can be obtained by using the model, and then the linear conversion can be usedFinally, respiratory signal values of n nodes are predicted, wherein w is a weight vector and b is a bias.
S408, carrying out logistics prediction on the space-time cyclic convolution block subjected to the overfitting alleviation according to linear conversion to obtain a logistics prediction result.
Specifically, the flow is predicted according to the conversion process of subtracting the average number from the original fraction and dividing the average number by the standard deviation, and a final prediction result is obtained.
Example 2:
as shown in fig. 2, the present embodiment provides a big data-based logistics prediction system, which includes an acquisition module 701, an extraction module 702, a decomposition module 703 and a prediction module 704, where:
the acquisition module 701: the method comprises the steps of acquiring historical logistics data, and preprocessing the historical logistics data to obtain preprocessed logistics data, wherein the historical logistics data comprises logistics plan demand, cost material fee control quantity and warehouse reserve quantity;
extraction module 702: the method comprises the steps of extracting characteristics related to the logistics plan demand, the cost material fee control quantity and the warehouse reserve quantity from the historical logistics data according to a rough set theory algorithm, and constructing a prediction model;
the decomposition module 703: the method comprises the steps of performing time sequence decomposition on historical logistics data by using a signal decomposition algorithm to obtain decomposition data;
the prediction module 704: the method is used for carrying out parameter optimization on the prediction model based on a drosophila optimization algorithm to obtain an optimized parameter set, inputting the optimized parameter set into the prediction model, and predicting the decomposition data based on a tree structure Parzen super-parameter optimization algorithm to obtain a logistics prediction result.
Specifically, the acquiring module 701 includes an acquiring unit 7011, a matching unit 7012, an acquiring unit 7013, and a first processing unit 7014, where:
obtaining unit 7011: the road network matching algorithm is used for marking each road in the urban road according to the parallelogram by utilizing the boundary rectangle, and obtaining a parallelogram area corresponding to each road;
matching section 7012: the method comprises the steps of matching the parallelogram area with GPS positioning data on a logistics transportation vehicle to form matched GPS positioning data;
acquisition unit 7013: the flow information of all the vehicles transported by the logistics in the required urban road is collected according to the GPS positioning data;
first processing unit 7014: the method is used for carrying out data cleaning on the flow information based on a data cleaning method, setting outliers and carrying out detection screening to obtain pretreated logistics data.
Specifically, the decomposition module 703 includes a construction unit 7031, a calculation unit 7032, and an update unit 7033, wherein:
construction unit 7031: the method comprises the steps of constructing a Lagrange function based on a Gaussian smoothing filtering algorithm through a quadratic penalty function term and a Lagrange multiplier;
calculation unit 7032: the method comprises the steps of converting a definition domain in the Lagrangian function from time to frequency according to historical logistics data, and calculating an extremum of the Lagrangian function;
update unit 7033: the method comprises the steps of updating Lagrangian multipliers, carrying out iterative computation, and if a computation result is smaller than a penalty factor, ending computation to obtain decomposition data; otherwise, the domain in the Lagrangian function is converted from time to frequency until the penalty factor is larger, and the calculation is finished.
Specifically, the prediction module 704 includes an adoption unit 7041, a search unit 7042, and an iteration unit 7043, where:
unit 7041 is adopted: the method is used for enabling the initial coordinate distribution of the drosophila to be uniform by adopting a chaos technology based on Logistic mapping;
search unit 7042: the method comprises the steps of randomly initializing the positions of the drosophila population, and searching random directions and distances of foods;
iteration unit 7043: the method comprises the steps of calculating a taste concentration determination value, making adaptability evaluation, performing iterative optimization, and determining an optimal solution; wherein, calculating the taste concentration determination value comprises obtaining the highest or the smallest and the best fruit fly coordinate points of the food smell concentration value, and marking the coordinate points as the searching position and the direction of the next generation fruit fly to provide guidance.
Specifically, the prediction module 704 includes a first building unit 7044, a second building unit 7045, a second processing unit 7046, a fitting unit 7047, and a prediction unit 7048, where:
first building unit 7044: for constructing a convolution model comprising at least three spatiotemporal cyclic convolution blocks, each of the spatiotemporal cyclic convolution blocks comprising a temporal convolution block and a spatial convolution block;
second building unit 7045: the method comprises the steps of stacking the space-time cyclic convolution blocks to construct a dynamic space-time convolution block, wherein the dynamic space-time convolution block comprises 2 time layers and 1 space layer connected with a time domain;
second processing unit 7046: the method is used for lifting the corresponding output dimension of each space-time cyclic convolution block by using a bottleneck theory and using a graph convolution layer;
fitting unit 7047: the method is used for performing fitting alleviation on the space-time cyclic convolution blocks after lifting processing based on a tree structure Parzen super-parameter optimization algorithm and a random discarding method;
prediction unit 7048: and the method is used for carrying out logistics prediction on the space-time cyclic convolution block subjected to the over-fitting alleviation according to linear conversion to obtain a logistics prediction result.
It should be noted that, regarding the system in the above embodiment, the specific manner in which the respective modules perform the operations has been described in detail in the embodiment regarding the method, and will not be described in detail herein.
Example 3:
corresponding to the above method embodiment, a big data based logistics prediction apparatus is further provided in this embodiment, and a big data based logistics prediction apparatus described below and a big data based logistics prediction method described above may be referred to correspondingly.
Fig. 3 is a block diagram illustrating a big data based logistics prediction apparatus 800 in accordance with an exemplary embodiment. As shown in fig. 3, the big data based logistics prediction apparatus 800 may include: a processor 801, a memory 802. The big data based logistics prediction apparatus 800 may further comprise one or more of a multimedia component 803, an i/O interface 804, and a communication component 805.
Wherein the processor 801 is configured to control the overall operation of the big data based logistics prediction apparatus 800 to complete all or part of the steps of the big data based logistics prediction method described above. The memory 802 is used to store various types of data to support the operation of the big data based logistics prediction apparatus 800, such data may include, for example, instructions for any application or method operating on the big data based logistics prediction apparatus 800, as well as application related data, such as contact data, messages, pictures, audio, video, etc. The Memory 802 may be implemented by any type or combination of volatile or non-volatile Memory devices, such as static random access Memory (Static Random Access Memory, SRAM for short), electrically erasable programmable Read-Only Memory (Electrically Erasable Programmable Read-Only Memory, EEPROM for short), erasable programmable Read-Only Memory (Erasable Programmable Read-Only Memory, EPROM for short), programmable Read-Only Memory (Programmable Read-Only Memory, PROM for short), read-Only Memory (ROM for short), magnetic Memory, flash Memory, magnetic disk, or optical disk. The multimedia component 803 may include a screen and an audio component. Wherein the screen may be, for example, a touch screen, the audio component being for outputting and/or inputting audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signals may be further stored in the memory 802 or transmitted through the communication component 805. The audio assembly further comprises at least one speaker for outputting audio signals. The I/O interface 804 provides an interface between the processor 801 and other interface modules, which may be a keyboard, mouse, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 805 is configured to perform wired or wireless communication between the big-data based logistics prediction apparatus 800 and other apparatuses. Wireless communication, such as Wi-Fi, bluetooth, near field communication (Near FieldCommunication, NFC for short), 2G, 3G or 4G, or a combination of one or more thereof, the respective communication component 805 may thus comprise: wi-Fi module, bluetooth module, NFC module.
In an exemplary embodiment, big data based logistics prediction apparatus 800 may be implemented by one or more application specific integrated circuits (Application Specific Integrated Circuit, abbreviated ASIC), digital signal processor (DigitalSignal Processor, abbreviated DSP), digital signal processing apparatus (Digital Signal Processing Device, abbreviated DSPD), programmable logic device (Programmable Logic Device, abbreviated PLD), field programmable gate array (Field Programmable Gate Array, abbreviated FPGA), controller, microcontroller, microprocessor, or other electronic component for performing the big data based logistics prediction method described above.
In another exemplary embodiment, a computer readable storage medium is also provided, comprising program instructions which, when executed by a processor, implement the steps of the big data based logistics prediction method described above. For example, the computer readable storage medium may be the memory 802 described above including program instructions executable by the processor 801 of the big data based logistics prediction apparatus 800 to perform the big data based logistics prediction method described above.
Example 4:
corresponding to the above method embodiment, there is further provided a readable storage medium in this embodiment, and a readable storage medium described below and a big data based logistics prediction method described above may be referred to correspondingly.
A readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the big data based logistics prediction method of the method embodiment described above.
The readable storage medium may be a usb 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, and the like.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present invention. Therefore, the protection scope of the invention is subject to the protection scope of the claims.
Claims (10)
1. A big data based logistics prediction method, comprising:
acquiring historical logistics data, and preprocessing the historical logistics data to obtain preprocessed logistics data, wherein the historical logistics data comprises logistics plan demand, cost material fee control quantity and warehouse reserve quantity;
extracting features related to the logistics plan demand, the cost material fee control quantity and the warehouse reserve quantity from the historical logistics data according to a rough set theory algorithm, and constructing a prediction model;
performing time sequence decomposition on the historical logistics data by using a signal decomposition algorithm to obtain decomposition data;
and carrying out parameter optimization on the prediction model based on a drosophila optimization algorithm to obtain an optimized parameter set, inputting the optimized parameter set into the prediction model, and predicting the decomposition data based on a tree structure Parzen super-parameter optimization algorithm to obtain a logistics prediction result.
2. The big data based logistics prediction method of claim 1, wherein the obtaining historical logistics data, and preprocessing the historical logistics data to obtain preprocessed logistics data, comprises:
marking each road in the urban road according to the parallelogram by utilizing a road network matching algorithm of the boundary rectangle, and obtaining a parallelogram area corresponding to each road;
matching the parallelogram area with GPS positioning data on the logistics transportation vehicle to form matched GPS positioning data;
acquiring flow information of all vehicles transported by the logistics in the required urban road according to the GPS positioning data;
and based on a data cleaning method, carrying out data cleaning on the flow information, setting outliers, and carrying out detection screening to obtain pretreated logistics data.
3. The big data based logistics prediction method of claim 1, wherein the performing time series decomposition on the historical logistics data by using a signal decomposition algorithm to obtain decomposition data comprises:
based on a Gaussian smoothing filtering algorithm, constructing a Lagrange function through a quadratic penalty function term and a Lagrange multiplier;
converting the definition domain in the Lagrangian function from time to frequency according to the historical logistics data, and calculating the extremum of the Lagrangian function;
updating Lagrangian multipliers, carrying out iterative computation, and if the computation result is smaller than the penalty factor, ending the computation to obtain decomposition data; otherwise, the domain in the Lagrangian function is converted from time to frequency until the penalty factor is larger, and the calculation is finished.
4. The big data based logistics prediction method of claim 1, wherein the drosophila-based optimization algorithm performs a parametric optimization on the prediction model, wherein the drosophila optimization algorithm comprises:
the chaos technology based on Logistic c mapping is adopted, so that the initial coordinate distribution of the drosophila melanogaster is uniform;
randomly initializing the position of the drosophila population, and searching the random direction and distance of food;
calculating a taste concentration determination value, making fitness evaluation, performing iterative optimization, and determining an optimal solution; wherein, calculating the taste concentration determination value comprises obtaining the coordinate points of the highest, the smallest and the best fruit flies of the food smell concentration value, and marking the coordinate points as the searching position and the direction of the fruit flies of the next generation to provide guidance.
5. The big data-based logistics prediction method of claim 1, wherein the tree-structure-based Parzen super-parameter optimization algorithm predicts the decomposition data to obtain a logistics prediction result, and the method comprises the following steps:
constructing a convolution model, wherein the convolution model comprises at least three space-time cyclic convolution blocks, and each space-time cyclic convolution block comprises a time convolution block and a space convolution block;
stacking the space-time cyclic convolution blocks to construct a dynamic space-time convolution block, wherein the dynamic space-time convolution block comprises 2 time layers and 1 space layer connected with a time domain;
lifting the corresponding output dimension of each space-time cyclic convolution block by using a graph convolution layer according to a bottleneck theory;
performing fitting alleviation on the space-time cyclic convolution blocks after lifting treatment based on a tree structure Parzen super-parameter optimization algorithm and a random discarding method;
and carrying out logistics prediction on the space-time cyclic convolution block subjected to the over-fitting alleviation according to linear conversion to obtain a logistics prediction result.
6. A big data based logistics prediction system, comprising:
the acquisition module is used for: the method comprises the steps of acquiring historical logistics data, and preprocessing the historical logistics data to obtain preprocessed logistics data, wherein the historical logistics data comprises logistics plan demand, cost material fee control quantity and warehouse reserve quantity;
and an extraction module: the method comprises the steps of extracting characteristics related to the logistics plan demand, the cost material fee control quantity and the warehouse reserve quantity from the historical logistics data according to a rough set theory algorithm, and constructing a prediction model;
and a decomposition module: the method comprises the steps of performing time sequence decomposition on historical logistics data by using a signal decomposition algorithm to obtain decomposition data;
and a prediction module: the method is used for carrying out parameter optimization on the prediction model based on a drosophila optimization algorithm to obtain an optimized parameter set, inputting the optimized parameter set into the prediction model, and predicting the decomposition data based on a tree structure Parzen super-parameter optimization algorithm to obtain a logistics prediction result.
7. The big-data based logistics prediction system of claim 6, wherein the acquisition module comprises:
the obtaining unit: the road network matching algorithm is used for marking each road in the urban road according to the parallelogram by utilizing the boundary rectangle, and obtaining a parallelogram area corresponding to each road;
matching unit: the method comprises the steps of matching the parallelogram area with GPS positioning data on a logistics transportation vehicle to form matched GPS positioning data;
the acquisition unit: the flow information of all the vehicles transported by the logistics in the required urban road is collected according to the GPS positioning data;
a first processing unit: the method is used for carrying out data cleaning on the flow information based on a data cleaning method, setting outliers and carrying out detection screening to obtain pretreated logistics data.
8. The big-data based logistics prediction system of claim 6, wherein the decomposition module comprises:
the construction unit comprises: the method comprises the steps of constructing a Lagrange function based on a Gaussian smoothing filtering algorithm through a quadratic penalty function term and a Lagrange multiplier;
a calculation unit: the method comprises the steps of converting a definition domain in the Lagrangian function from time to frequency according to historical logistics data, and calculating an extremum of the Lagrangian function;
an updating unit: the method comprises the steps of updating Lagrangian multipliers, carrying out iterative computation, and if a computation result is smaller than a penalty factor, ending computation to obtain decomposition data; otherwise, the domain in the Lagrangian function is converted from time to frequency until the penalty factor is larger, and the calculation is finished.
9. The big-data based logistics prediction system of claim 6, wherein the prediction module comprises:
the adoption unit comprises: the method is used for enabling the initial coordinate distribution of the drosophila to be uniform by adopting a chaos technology based on Logistic c mapping;
search unit: the method comprises the steps of randomly initializing the positions of the drosophila population, and searching random directions and distances of foods;
iteration unit: the method comprises the steps of calculating a taste concentration determination value, making adaptability evaluation, performing iterative optimization, and determining an optimal solution; wherein, calculating the taste concentration determination value comprises obtaining the highest or the smallest and the best fruit fly coordinate points of the food smell concentration value, and marking the coordinate points as the searching position and the direction of the next generation fruit fly to provide guidance.
10. The big-data based logistics prediction system of claim 6, wherein the prediction module comprises:
a first construction unit: for constructing a convolution model comprising at least three spatiotemporal cyclic convolution blocks, each of the spatiotemporal cyclic convolution blocks comprising a temporal convolution block and a spatial convolution block;
a second construction unit: the method comprises the steps of stacking the space-time cyclic convolution blocks to construct a dynamic space-time convolution block, wherein the dynamic space-time convolution block comprises 2 time layers and 1 space layer connected with a time domain;
a second processing unit: the method is used for lifting the corresponding output dimension of each space-time cyclic convolution block by using a bottleneck theory and using a graph convolution layer;
fitting unit: the method is used for performing fitting alleviation on the space-time cyclic convolution blocks after lifting processing based on a tree structure Parzen super-parameter optimization algorithm and a random discarding method;
prediction unit: and the method is used for carrying out logistics prediction on the space-time cyclic convolution block subjected to the over-fitting alleviation according to linear conversion to obtain a logistics prediction result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311431282.6A CN117875468A (en) | 2023-10-31 | 2023-10-31 | Logistics prediction method and system based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311431282.6A CN117875468A (en) | 2023-10-31 | 2023-10-31 | Logistics prediction method and system based on big data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117875468A true CN117875468A (en) | 2024-04-12 |
Family
ID=90585332
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311431282.6A Pending CN117875468A (en) | 2023-10-31 | 2023-10-31 | Logistics prediction method and system based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117875468A (en) |
-
2023
- 2023-10-31 CN CN202311431282.6A patent/CN117875468A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yu et al. | Prediction of bus travel time using random forests based on near neighbors | |
| JP6743934B2 (en) | Method, apparatus and system for estimating causal relationship between observed variables | |
| JP6179598B2 (en) | Hierarchical hidden variable model estimation device | |
| CN113366473A (en) | Method and system for automatic selection of models for time series prediction of data streams | |
| CN101710422B (en) | Image segmentation method based on overall manifold prototype clustering algorithm and watershed algorithm | |
| Yao et al. | Simulating urban growth processes by integrating cellular automata model and artificial optimization in Binhai New Area of Tianjin, China | |
| CN112396231B (en) | Modeling method and device for space-time data, electronic equipment and readable medium | |
| US10719639B2 (en) | Massively accelerated Bayesian machine | |
| Nikolić et al. | Implementation of generic algorithm in map-matching model | |
| JP6330901B2 (en) | Hierarchical hidden variable model estimation device, hierarchical hidden variable model estimation method, payout amount prediction device, payout amount prediction method, and recording medium | |
| CN116611546A (en) | Knowledge-graph-based landslide prediction method and system for target research area | |
| Parri et al. | A hybrid VMD based contextual feature representation approach for wind speed forecasting | |
| CN115438453B (en) | Method for constructing road network facility topological structure by using observation data | |
| CN116861373A (en) | Query selectivity estimation method, system, terminal equipment and storage medium | |
| CN117875468A (en) | Logistics prediction method and system based on big data | |
| KR20210042709A (en) | Method and server for stock movement prediction using corporate relation data | |
| CN113284337B (en) | OD matrix calculation method and device based on vehicle track multidimensional data | |
| Ye et al. | Application of time series method to the passenger flow prediction in the intelligent bus transportation system with big data | |
| CN114936700A (en) | Construction method and prediction method of aerosol concentration space-time distribution prediction model | |
| Loch-Dehbi et al. | Stochastic reasoning for UAV supported reconstruction of 3D building models | |
| Alamgeer et al. | Data Mining with Comprehensive Oppositional Based Learning for Rainfall Prediction. | |
| US20230419195A1 (en) | System and Method for Hierarchical Factor-based Forecasting | |
| Boyko et al. | Methodology for Estimating the Cost of Construction Equipment Based on the Analysis of Important Characteristics Using Machine Learning Methods | |
| Sikder | Application of granular computing paradigm in knowledge induction | |
| US20220237179A1 (en) | Systems and Methods for Improved Machine Learning Using Data Completeness and Collaborative Learning Techniques |
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 |