CN111915100A - High-precision freight prediction method and freight prediction system - Google Patents
High-precision freight prediction method and freight prediction system Download PDFInfo
- Publication number
- CN111915100A CN111915100A CN202010825904.3A CN202010825904A CN111915100A CN 111915100 A CN111915100 A CN 111915100A CN 202010825904 A CN202010825904 A CN 202010825904A CN 111915100 A CN111915100 A CN 111915100A
- Authority
- CN
- China
- Prior art keywords
- freight
- prediction
- data
- demand
- railway
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000012549 training Methods 0.000 claims description 32
- 238000007726 management method Methods 0.000 claims description 25
- 238000004458 analytical method Methods 0.000 claims description 19
- 238000012545 processing Methods 0.000 claims description 16
- 230000008569 process Effects 0.000 claims description 11
- 238000012360 testing method Methods 0.000 claims description 11
- 238000011161 development Methods 0.000 claims description 10
- 230000001932 seasonal effect Effects 0.000 claims description 9
- 230000000007 visual effect Effects 0.000 claims description 9
- 230000002860 competitive effect Effects 0.000 claims description 7
- 230000008676 import Effects 0.000 claims description 7
- 238000013523 data management Methods 0.000 claims description 5
- 238000012384 transportation and delivery Methods 0.000 claims description 5
- 238000004140 cleaning Methods 0.000 claims description 3
- 238000013016 damping Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 2
- 238000010219 correlation analysis Methods 0.000 claims 1
- 230000000694 effects Effects 0.000 abstract description 12
- 238000013439 planning Methods 0.000 abstract description 5
- 238000011160 research Methods 0.000 abstract description 4
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 230000007547 defect Effects 0.000 abstract description 3
- 230000007812 deficiency Effects 0.000 abstract description 3
- 230000006872 improvement Effects 0.000 abstract description 3
- 238000005728 strengthening Methods 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 22
- 238000013528 artificial neural network Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 6
- 238000013461 design Methods 0.000 description 4
- 230000010354 integration Effects 0.000 description 4
- 230000007774 longterm Effects 0.000 description 4
- 239000000654 additive Substances 0.000 description 3
- 230000000996 additive effect Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 3
- 238000012800 visualization Methods 0.000 description 3
- 238000012550 audit Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 239000000470 constituent Substances 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000013075 data extraction Methods 0.000 description 2
- 238000013079 data visualisation Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 230000001364 causal effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000004141 dimensional analysis Methods 0.000 description 1
- 238000009713 electroplating Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000007639 printing Methods 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 230000026676 system process Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- G06Q10/083—Shipping
-
- 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
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Development Economics (AREA)
- Data Mining & Analysis (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Operations Research (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a high-precision freight prediction method and a freight prediction system, which can accurately and timely obtain the freight prediction result, have the most direct effect on the decision of a railway department and the planning of a future route, are beneficial to the railway department to fully utilize railway transportation energy resources, reasonably distribute railway vehicles and other facility equipment, and reduce unnecessary time cost and other operation cost, thereby achieving the effect of saving cost and improving the final income of the railway department. Meanwhile, it can be known that the accuracy of the prediction result is more important, and if the prediction result is inaccurate, corresponding loss is generated to the freight department, so that improvement and innovation of the traditional method are important, the defects and deficiencies are made up, or a new method is introduced, so that the method has practical significance and strong social and economic values for strengthening the research on the railway freight volume prediction.
Description
Technical Field
The invention relates to the technical field of traffic prediction intelligent management, in particular to a high-precision freight prediction method and a freight prediction system.
Background
Since the new century, the economy of China is rapidly developed, and the railway transportation capacity of China is greatly improved. The high-speed railway is used as a modern transportation sign, is vigorously developed in China, provides great convenience for the work and life of people, and makes outstanding contribution to the rising of the economy of China. China is the country with the largest high-speed rail construction and operation scale in the world and the country with the most complex railway operation scene and external environment, and road network transportation organization and operation guarantee technology with Chinese characteristics and ultra-large-scale and ultra-strong freight demand is formed.
The future railway freight requirements are determined by predicting the railway freight requirements and adjusting key influence factors of the railway freight requirements, the railway management department can master the development trend of the future railway freight market in time, and make appropriate institutional reform, research and make countermeasures suitable for the development of the China railway transportation industry. Finally, the comprehensive competitive energy of the China railway transportation system is enhanced, so that the social and economic benefits of the China railway transportation system are improved.
Therefore, aiming at the new characteristics of the transportation demand under the high-speed rail network forming condition, the freight demand evolution law is researched, the transportation demand prediction method considering the transportation service attribute is provided, and the design of the railway freight transportation product is supported. A multi-scene freight demand forecasting system is constructed, functions of data acquisition, data management, demand forecasting, visualization and the like are realized, and the method has important significance for railway departments.
Disclosure of Invention
The embodiment of the invention provides a high-precision freight prediction method and a freight prediction system, which are used for solving the technical problems in the prior art.
In order to achieve the purpose, the invention adopts the following technical scheme.
A high accuracy freight prediction method comprising:
acquiring historical data of the railway freight volume, establishing a target function aiming at monthly transportation requirements, and dividing the historical data of the railway freight volume into a training sample set and a test sample set;
establishing and training a monthly freight demand prediction model based on the training sample set and the test sample set;
and establishing and solving an annual freight demand prediction model based on the monthly transport demand prediction model to obtain a railway freight capacity prediction result.
Preferably, the objective function for monthly transportation demand is xn＝f(xn-1,xn-2,xn-N)(1)；
Based on the training sample set and the testing sample set, the method for establishing and training the monthly freight demand prediction model comprises the following steps:
based on the training sample set, samples are extracted for multiple times, and a training matrix is established and solved
Where each column extracts a sample once, the last row is the desired output, xNIs the value of the objective function.
Preferably, the establishing and solving an annual freight demand prediction model based on the monthly transportation demand prediction model, and the obtaining of the railway freight volume prediction result comprises:
establishing an annual freight demand prediction function yt+k＝(a(t)+b(t)k)ct+k(3) (ii) a In the formula (I), the compound is shown in the specification,a (t) represents intercept, b (t) represents trend, c (t) is seasonal factor of multiplication model, and the formula is respectively used
b(t)＝β[a(t)-a(t-1)]+ (1-. beta.) b (t-1) (5) and
obtaining that alpha, beta and gamma are between 0 and 1 and are damping factors;
establish and solve formula
And obtaining a railway freight volume prediction result.
In a second aspect, the present invention provides a high-precision freight prediction system, including:
the data import submodule is used for acquiring and processing historical data of the railway freight volume;
the data management submodule is used for storing, summarizing and modifying various data;
the prediction model management submodule is used for executing the freight prediction method to obtain a railway freight volume prediction result; the system is also used for obtaining corresponding training data sets according to different model parameters and inputting the training data sets into a monthly transportation demand prediction model and/or an annual freight demand prediction model to obtain a prediction result;
the demand forecasting submodule is used for forecasting the demand according to the forecasting result;
and the visual display submodule is used for visually outputting the demand forecast.
Preferably, the process of acquiring and processing the historical data of the railway freight volume by the data import submodule comprises the following steps:
acquiring basic data of one or more cargo freight volumes;
collecting, extracting, converting, processing, cleaning and storing the basic data of the one or more cargo quantities;
the process of the demand forecasting sub-module for forecasting the demand according to the forecasting result comprises the following steps:
forecasting monthly and/or annual demand according to the cargo type, and forming a report; the report forms include reports, charts and maps.
Preferably, the prediction by the demand prediction sub-module comprises: forecasting the development trend of annual and monthly railway freight requirements; predicting the development trend of the bulk cargo delivery quantity; forecasting the freight sending quantity of the urban railway; the transport demand forecast for competitive goods categories among the major ODs of a typical aisle.
Preferably, the visual input types of the visual display sub-module comprise historical data display bulk cargo sending quantity analysis display, railway freight demand gray correlation degree analysis display, urban railway freight sending quantity prediction result display and typical channel main OD competitive cargo item type transportation demand prediction result display.
The technical scheme provided by the embodiment of the invention shows that the high-precision freight prediction method and the freight prediction system provided by the invention can accurately and timely obtain the freight prediction result, have the most direct effect on the decision of the railway department and the planning of the future route, are beneficial to the railway department to fully utilize railway freight resources, reasonably distribute railway vehicles and other facility equipment, reduce unnecessary time cost and other operation cost, further achieve the effect of saving cost and improve the final income of the railway department. Meanwhile, it can be known that the accuracy of the prediction result is more important, and if the prediction result is inaccurate, corresponding loss is generated to the freight department, so that improvement and innovation of the traditional method are important, the defects and deficiencies are made up, or a new method is introduced, so that the method has practical significance and strong social and economic values for strengthening the research on the railway freight volume prediction.
Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a process flow diagram of a high accuracy freight prediction method provided by the present invention;
FIG. 2 is a system flow diagram of a high accuracy freight forecast system provided by the present invention;
FIG. 3 is a general architecture diagram of a high accuracy freight forecast system provided by the present invention;
FIG. 4 is a system architecture diagram of a high-precision freight prediction system according to the present invention;
FIG. 5 is a logic block diagram of a high accuracy freight forecast system provided by the present invention;
fig. 6 is a functional structure diagram of a high-precision freight prediction system provided by the present invention.
In the figure:
501. the data import submodule 502, the data management submodule 503, the prediction model management submodule 504, the demand prediction submodule 505 and the visualization display submodule.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the accompanying drawings are illustrative only for the purpose of explaining the present invention, and are not to be construed as limiting the present invention.
As used herein, the singular forms "a", "an", "the" and "the" include plural referents unless the context clearly dictates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Further, "connected" or "coupled" as used herein may include wirelessly connected or coupled. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
For the convenience of understanding the embodiments of the present invention, the following description will be further explained by taking several specific embodiments as examples in conjunction with the drawings, and the embodiments are not to be construed as limiting the embodiments of the present invention.
Referring to fig. 1, the high-precision freight prediction method provided by the invention comprises the following steps:
acquiring historical data of the railway freight volume, establishing a target function aiming at monthly transportation requirements, and dividing the historical data of the railway freight volume into a training sample set and a test sample set;
establishing and training a monthly freight demand prediction model based on the training sample set and the test sample set;
and establishing and solving an annual freight demand prediction model based on the monthly transport demand prediction model to obtain a railway freight capacity prediction result.
Railway freight demand forecasts can generally be divided into two categories, long term forecasts, which are generally forecasted at intervals of days, months, or even years, and short term forecasts, which are generally forecasted at intervals of no more than one day, which can be an hour, half day, or day. It is necessary to provide real-time freight demand information to a decision maker so that the decision maker can quickly decide a plan, so that traffic flow prediction in a short time is very important. And the long-term prediction can provide data support for aspects such as long-term railway transportation planning of the country, so the long-term prediction also has great significance for decision-making departments.
In the embodiments provided by the present invention, the time units for prediction are divided into months and years.
Specifically, the prediction model to be established adopts a neural network, and the neural network has strong nonlinear mapping capability and can theoretically realize any complex causal relationship. In the embodiment, an Elman neural network is adopted, which is a typical local regression network, and compared with a forward neural network, the Elman neural network has one more connecting layer in structure, so that the past state can be memorized, and the Elman neural network is particularly suitable for processing time series problems.
The target (mapping) function established based on the Elman neural network and combined with the historical data of the railway freight volume can be expressed as
xn＝f(xn-1,xn-2,xn-N) (1)
For the selected railway freight volume data, the selected railway freight volume data is firstly divided into training samples and testing samples. Taking training samples as an example, x is extracted1～xNForming a first sample in which (x)1,x2,…,xN+1) Is an independent variable, xNIs the objective function value; decimating x2～xN+1Forming a second sample in which (x)2,x3,…,xN) Is an independent variable, xN+1And (4) forming the following training matrix for the objective function value by analogy:
where each column is one sample and the last row is the desired output. And the Elman neural network inputs the training samples into the Elman network for training, so that a trained network can be obtained.
Decimating x1～xNForm the first sample, (x)1,x2,…,xN-1) Is an independent variable, xNThe objective function value is obtained by analogy. The size of N has a large influence on the learning degree of the network. Theoretically, the larger N, the better the degree of learning can cover a change in the amount of shipment for a longer time, but the less the data set, the lower the prediction accuracy. Since the data volume of monthly transportation demand forecast is only 25 data in 2016 (9 months) to 2018 (9 months), the N value is not suitable to be too large, so that the N value is limited to [2,6 ]]。
The forecast of the monthly transportation demand takes the data of the railway freight volume in 2016, 9 and 2018 and 3 months as training data, and takes the data of the railway freight volume in 2018, 4 and 2018 and 9 months as test data. And determining the time span of the sample set in the Elman neural network model by calculating the errors of training and testing data under different N values.
When the data volume of the monthly transportation demand prediction is increased, different N values are selected by training different data through the prediction model management module so that the prediction effect is better.
In the embodiment provided by the invention, the electroplating demand prediction model is established by selecting a Holt-Winters model to carry out monthly freight volume prediction according to the characteristics of periodicity and volatility of the railway freight volume, and extrapolating the result to the annual railway freight volume. The Holt-Winters model is suitable for time series with both tendency and seasonal fluctuation characteristics, and the specific models thereof are mainly divided into two types: the addition model and the multiplication model have the same idea, and divide the time series variation factors into 4 types, wherein T represents a trend component, S represents a seasonal component, C represents a cyclic component, and I represents a random variation component. The multiplication model is expressed as Y ═ T × S × C × I, and the influence between the four constituent variables is mutual; while the additive model is denoted as Y ═ T + S + C + I, the effects between the four constituent variables are independent.
For a time series Y, an addition model or a multiplication model may be employed. The additive model assumes that the absolute value of the effect of the seasonal factor on the sequence Y is constant, while the multiplicative model assumes that the proportion of the effect of the seasonal factor on the sequence Y is constant. If Y is the total amount index, the influence value of the seasonal factor is generally related to the base number of the total amount, and the multiplication model is more suitable for practical situations. If the sequence is a relative index or a developmental index, it is more reasonable to use an additive model. Here, the Holt-Winters multiplication model is selected for prediction of railway freight volume. The smoothing sequence and the prediction function are shown in the following formula (3):
yt+k＝(a(t)+b(t)k)ct+k (3)
wherein, a (t) represents intercept, b (t) represents trend, c (t) is seasonal factor of multiplication model, and the following three formulas are shown:
b(t)＝β[a(t)-a(t-1)]+(1-β)b(t-1) (5)
alpha, beta and gamma are between 0 and 1 and are damping factors. The predicted value is calculated by the formula (3-7)
The prediction result of the railway freight volume is obtained by solving this equation (7).
In the monthly data, s is taken as 12, and the seasonal factor is estimated using the last s.
When the data volume of the annual transportation demand forecast is increased, different s values are selected by training different data through the forecast model management module so that the forecast effect is better.
In a second aspect, the present invention provides a high-precision freight forecast system, as shown in the system flow of fig. 2, including metadata management, data collection, data integration processing, forecast of freight demands in different scenarios, and visualization interface display.
Metadata management is the basis of the whole system process, on one hand, the definition of metadata is completed through planning and designing, and the meaning and standard of data resources are determined; on one hand, the preparation and formulation work of data and system specifications in stages of data acquisition, summarization, integration, analysis and application and the like is completed through the metadata management function; and all-round description is carried out on various data existing in each link and each stage through metadata in the whole service process, so that data information in the system can be read and managed in the whole process.
In the data acquisition stage, firstly, the acquisition task is flexibly adjusted according to the actual condition of data through acquisition task configuration, and task attributes such as an acquired object, acquisition time, acquisition period, audit level and the like are set. And provide metadata management with the need to supplement the change with metadata. The data acquisition function realizes the statistical data acquisition of each road company through the service functions of data acquisition, exchange processing, data summarization, import loading and the like, and the acquired data enters the system and can be summarized.
And the collected data enters the system, and various data can be summarized in the next step. Data summarization is a comprehensive method for overall data assurance and data audit, and is used for auditing the correctness and validity of data. For the demand forecasting work, the function of summarizing and examining data is important, and the data is the basis of demand forecasting when being correct and effective. The data summarization service mainly realizes the automatic summarization function of the collected data through a summarization rule customized by the system, generates a summarized data report and provides export printing. Then, the data is converted by the data conversion loading service and then stored in a database of the system, and original data of a reporting unit is finally saved as files.
The data stored in the system database becomes purified basic data after the auditing and evaluating operation of the data integration processing link, and various processing results are generated through integration processing, and the processing results are also stored in the database. And performing daily service operations such as data maintenance, update, state management and the like on the purified basic data through a database management function.
After the basic data are stored in the database, in the comprehensive analysis application stage, the freight demand prediction is carried out by extracting, converting and loading the purified basic data in the database to different scenes, the visual interface display is carried out, various metadata information required by the analysis demand prediction is perfected, and the updating and the expansion of the model and the warehouse data are carried out at regular time/at any time.
A series of analysis service functions such as data comprehensive analysis application and the like can utilize a GIS platform, a BI tool and various statistical analysis software to perform inquiry, analysis, mining and prediction application in various forms. And the system platform integrates related data analysis tools to realize configuration management. The data required by analysis, the tools, algorithms and functions required by the analysis and the display form of the analysis result can be defined by self, so that an extensible analysis platform is provided for the user to perform analysis activities by self.
As shown in fig. 3, the overall architecture of the freight demand forecasting system is divided into six hierarchical architectures, namely a basic data layer, a data acquisition layer, an application support layer, data visualization, an application layer, and a platform management and control layer.
(1) Data acquisition layer
The basic data of the related freight volume acquired from the railway bureau are mainly collected, extracted, converted, processed, cleaned, stored and the like from a data structure and a data scale, so that the construction of a system database is completed, and the acquired data are integrally stored according to the database design. The key part is the ETL process, and the main links in the ETL process are data extraction, data conversion and processing and data loading. And extracting required data from the data source by a user, cleaning the data, and finally loading the data into the data source according to a predefined model.
An ETL system needs to be able to complete the periodic automatic loading of daily data within a limited time, support the loading of initial data and historical data, and meet the requirements for future expansion. Dozens or more target data tables and source data of the same number in the system mean the complexity of an ETL program, the operation efficiency of the system needs to be fully considered due to huge data volume, and a flexible, simple and clear program structure is required for developing the complex program conveniently; the requirement for optimizing the efficiency of the program often requires personalized design for different data. Therefore, the design of the ETL must strike a balance between manageability and program performance of the development.
Data extraction refers to the process of extracting data from a data source. In practical application, a relational database is frequently adopted as a data source, and other data sources also have file forms, such as txt files, excel files, xml files, PDF files, and the like.
Since the data extracted from the data source does not necessarily completely satisfy the requirements of the destination library, such as inconsistency of data format, data input error, incomplete data, and so on, the extracted data is subjected to data conversion and processing.
Loading the converted and processed data into the corresponding destination database is the final step of the ETL process.
(2) Using a supporting layer
Data are extracted and packaged, through related components and an algorithm model, the data are deeply mined through the algorithm model, multi-dimensional analysis display and query are conducted on data of a data acquisition layer through a BI tool, values in the data are found, and data analysis and prediction are achieved.
(3) Data visualization
The platform is constructed according to designed system targets and functions, an analysis model and a prediction model monitored by a freight market are efficiently and seamlessly integrated with a GIS platform, an integrated visual model component is assisted, and visual display is realized on an application layer by combining related technologies.
(4) Application layer
The application layer enables a system user to conveniently and quickly access various data in the system and perform various analysis and prediction operations through modes such as report forms, charts, map display and the like, check analysis results of a large data platform and form report forms. For example, the user checks the predicted arrival and sending quantity of the railway goods in the future monthly classification of each road bureau, the predicted arrival and sending quantity of the railway goods in the future monthly classification of each province, the predicted freight quantity in the future nationwide and the like through the application layer.
The front-end display platform adopts a B/S framework, is composed of a whole set of components or services, and is connected through a powerful Web-based communication framework, so that different application requirements of an application user are met. The components can be independent or can be mutually converted and adjusted, and all the components can be realized in one interface.
(5) Platform management and control
The method relates to the fields of metadata management, data quality management, data security management, platform operation and maintenance and the like, and aims to ensure timeliness, legality, integrity, consistency, auditability and security of data and management of a platform.
The monitoring alarm is used for monitoring the application condition of resources such as CPU, memory, network and the like of each module, and when the resource accounts for the application and exceeds a set threshold value, an alarm is sent to a system administrator.
The user management is the role of the application user in the management platform, and the application access authority of each user is configured.
The cluster management is to configure the starting, stopping and checking states of each module in the cluster, and can configure the specific parameters of the modules at the same time.
When the machine fails, the modular system automatically switches to the backup machine.
Metadata management includes the development of vocabularies, the definition of data elements and entities, rules and algorithms, and data characteristics.
And establishing a strict data quality standard for metadata and data in data processing, and strictly controlling data to be stored in a warehouse.
The technical architecture diagram of the system is shown in fig. 4, which is specifically described as follows:
(1) the network layer adopts the technologies of load balancing, transmission encryption and the like.
(2) The data storage adopts technologies such as a relational database, a memory database, file storage and the like.
(3) The data transmission layer adopts the technologies of message queue, RESTFul, Json and the like.
(4) The platform layer application component takes Spring as a core and divides each layer of the system in a classical MVC mode. SpringMVC is used as a control layer, JDBCtemplate is used as a data access layer, and Json is used as a data transmission format for interaction of an internal network and an external network.
(5) The display layer adopts Jquery, ExtJS and other page display technologies.
As shown in fig. 5 and 6, the freight prediction system provided by the present invention includes five main components:
the data import submodule 501 is used for acquiring and processing historical data of railway freight volume;
the data management submodule 502 is used for storing, summarizing and modifying various data;
the prediction model management submodule 503 is configured to execute the freight prediction method to obtain a prediction result of the railway freight volume; the system is also used for obtaining corresponding training data sets according to different model parameters and inputting the training data sets into a monthly transportation demand prediction model and/or an annual freight demand prediction model to obtain a prediction result;
the demand forecasting submodule 504 is used for forecasting demands according to forecasting results, and the specific demand forecasting comprises the development trend forecasting of annual and monthly railway freight demands; predicting the development trend of the bulk cargo delivery quantity; forecasting the freight sending quantity of the urban railway; forecasting the transportation demand of competitive goods among main OD (origin destination) of a typical channel;
the visualized display submodule 505 is used for performing visualized output on demand prediction, and comprises historical data display of bulk cargo sending quantity analysis display, railway freight demand gray correlation degree analysis display and urban railway freight sending quantity prediction result display; and the main OD of the typical channel has the functions of displaying the transportation demand prediction result of competitive goods and the like.
In conclusion, the high-precision freight prediction method and the freight prediction system provided by the invention can accurately and timely obtain the freight prediction result, have the most direct effect on the decision of the railway department and the planning of the future route, are beneficial to the railway department to fully utilize railway freight resources, reasonably distribute railway vehicles and other facility equipment, reduce unnecessary time cost and other operation cost, and further achieve the effect of saving cost and improve the final income of the railway department. Meanwhile, it can be known that the accuracy of the prediction result is more important, and if the prediction result is inaccurate, corresponding loss is generated to the freight department, so that improvement and innovation of the traditional method are important, the defects and deficiencies are made up, or a new method is introduced, so that the method has practical significance and strong social and economic values for strengthening the research on the railway freight volume prediction.
Those of ordinary skill in the art will understand that: the figures are merely schematic representations of one embodiment, and the blocks or flow diagrams in the figures are not necessarily required to practice the present invention.
From the above description of the embodiments, it is clear to those skilled in the art that the present invention can be implemented by software plus necessary general hardware platform. Based on such understanding, the technical solutions of the present invention may be embodied in the form of software products, which may be stored in a storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and include instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the embodiments or some parts of the embodiments of the present invention.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for apparatus or system embodiments, since they are substantially similar to method embodiments, they are described in relative terms, as long as they are described in partial descriptions of method embodiments. The above-described embodiments of the apparatus and system are merely illustrative, and the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are also included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (7)
1. A highly accurate freight prediction method, comprising:
acquiring historical data of the railway freight volume, establishing a target function aiming at monthly transportation requirements, and dividing the historical data of the railway freight volume into a training sample set and a test sample set;
establishing and training a monthly freight demand prediction model based on the training sample set and the test sample set;
and establishing and solving an annual freight demand prediction model based on the monthly transport demand prediction model to obtain a railway freight capacity prediction result.
2. The freight prediction method of claim 1, where the objective function for monthly transport demand is xn＝f(xn-1,xn-2,xn-N) (1)；
The establishing and training of the monthly freight demand prediction model based on the training sample set and the testing sample set comprises the following steps:
based on the training sample set, extracting samples for multiple times, establishing and solving a training matrix
(2) (ii) a Where each column extracts a sample once, the last row is the desired output, xNIs the value of the objective function.
3. The freight prediction method according to claim 1, wherein the building and solving an annual freight demand prediction model based on the monthly transportation demand prediction model to obtain the railway freight volume prediction result comprises:
establishing an annual freight demand prediction function yt+k＝(a(t)+b(t)k)ct+k(3) (ii) a Wherein a (t) represents intercept, b (t) represents trend, and c (t) is seasonal factor of multiplication model, respectively
b(t)＝β[a(t)-a(t-1)]+ (1-. beta.) b (t-1) (5) and
obtaining that alpha, beta and gamma are between 0 and 1 and are damping factors;
establish and solve formula
And obtaining a railway freight volume prediction result.
4. A high accuracy freight forecast system, comprising:
the data import submodule is used for acquiring and processing historical data of the railway freight volume;
the data management submodule is used for storing, summarizing and modifying various data;
a prediction model management submodule for executing the freight prediction method according to any one of claims 1 to 3 to obtain a railway freight volume prediction result; the system is also used for obtaining corresponding training data sets according to different model parameters and inputting the training data sets into a monthly transportation demand prediction model and/or an annual freight demand prediction model to obtain a prediction result;
the demand forecasting submodule is used for forecasting the demand according to the forecasting result;
and the visual display submodule is used for visually outputting the demand forecast.
5. The freight prediction system of claim 4, wherein the data import sub-module obtains and processes historical data of railway freight volume by:
acquiring basic data of one or more cargo freight volumes;
collecting, extracting, converting, processing, cleaning and storing the basic data of the one or more cargo quantities;
the process of the demand forecasting sub-module for forecasting the demand according to the forecasting result comprises the following steps:
forecasting monthly and/or annual demand according to the cargo type, and forming a report; the report forms include reports, charts and maps.
6. The shipment forecasting system of claim 4, wherein the forecasting by the demand forecasting sub-module comprises: forecasting the development trend of annual and monthly railway freight requirements; predicting the development trend of the bulk cargo delivery quantity; forecasting the freight sending quantity of the urban railway; the transport demand forecast for competitive goods categories among the major ODs of a typical aisle.
7. The freight prediction system of claim 4, wherein the visual input types of the visual display sub-module include historical data display of bulk cargo delivery volume analysis display, grey correlation analysis display of railway freight demand, urban railway freight delivery volume prediction display, and transportation demand prediction display of competitive cargo categories among the main ODs of typical aisles.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010825904.3A CN111915100B (en) | 2020-08-17 | 2020-08-17 | High-precision freight prediction method and freight prediction system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010825904.3A CN111915100B (en) | 2020-08-17 | 2020-08-17 | High-precision freight prediction method and freight prediction system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111915100A true CN111915100A (en) | 2020-11-10 |
| CN111915100B CN111915100B (en) | 2024-04-05 |
Family
ID=73279651
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010825904.3A Active CN111915100B (en) | 2020-08-17 | 2020-08-17 | High-precision freight prediction method and freight prediction system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111915100B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113326976A (en) * | 2021-05-08 | 2021-08-31 | 东南大学 | Port freight volume online prediction method and system based on time-space correlation |
| CN113609208A (en) * | 2021-08-11 | 2021-11-05 | 中铁第四勘察设计院集团有限公司 | Railway OD data analysis visualization system and method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2012073835A (en) * | 2010-09-29 | 2012-04-12 | Railway Technical Research Institute | Evaluation method for railroad freight transportation effect |
| CN109492812A (en) * | 2018-11-12 | 2019-03-19 | 北京航天智造科技发展有限公司 | Logistics Freight Demand Forecasting method based on support vector machines, attribute reduction and genetic algorithm |
| CN110309948A (en) * | 2019-05-09 | 2019-10-08 | 上汽安吉物流股份有限公司 | Complete vehicle logistics order forecast method and device, logistics system and computer-readable medium |
| US20190392363A1 (en) * | 2018-06-22 | 2019-12-26 | Applied Materials, Inc. | Predicting arrival time of components based on historical receipt data |
| CN111507541A (en) * | 2020-04-30 | 2020-08-07 | 南京福佑在线电子商务有限公司 | Goods quantity prediction model construction method, goods quantity measurement device and electronic equipment |
-
2020
- 2020-08-17 CN CN202010825904.3A patent/CN111915100B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2012073835A (en) * | 2010-09-29 | 2012-04-12 | Railway Technical Research Institute | Evaluation method for railroad freight transportation effect |
| US20190392363A1 (en) * | 2018-06-22 | 2019-12-26 | Applied Materials, Inc. | Predicting arrival time of components based on historical receipt data |
| CN109492812A (en) * | 2018-11-12 | 2019-03-19 | 北京航天智造科技发展有限公司 | Logistics Freight Demand Forecasting method based on support vector machines, attribute reduction and genetic algorithm |
| CN110309948A (en) * | 2019-05-09 | 2019-10-08 | 上汽安吉物流股份有限公司 | Complete vehicle logistics order forecast method and device, logistics system and computer-readable medium |
| CN111507541A (en) * | 2020-04-30 | 2020-08-07 | 南京福佑在线电子商务有限公司 | Goods quantity prediction model construction method, goods quantity measurement device and electronic equipment |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113326976A (en) * | 2021-05-08 | 2021-08-31 | 东南大学 | Port freight volume online prediction method and system based on time-space correlation |
| CN113326976B (en) * | 2021-05-08 | 2022-11-15 | 东南大学 | Port freight volume online prediction method and system based on time-space correlation |
| CN113609208A (en) * | 2021-08-11 | 2021-11-05 | 中铁第四勘察设计院集团有限公司 | Railway OD data analysis visualization system and method |
| CN113609208B (en) * | 2021-08-11 | 2023-05-12 | 中铁第四勘察设计院集团有限公司 | Railway OD data analysis visualization system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111915100B (en) | 2024-04-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ballestar et al. | Impact of robotics on manufacturing: A longitudinal machine learning perspective | |
| CN102708149A (en) | Data quality management method and system | |
| CN105574593B (en) | Track state static detection and control system and method based on cloud computing and big data | |
| Baloui Jamkhaneh et al. | Inspection error and its effects on single sampling plans with fuzzy parameters | |
| CN108846691A (en) | Regional grain and oil market price monitoring analysing and predicting system and monitoring method | |
| CN111915099B (en) | Freight traffic prediction system with visualized freight information | |
| CN111915100B (en) | High-precision freight prediction method and freight prediction system | |
| CN105761141A (en) | Intelligent monitoring analysis platform | |
| CN103971023A (en) | Automatic quality evaluating system and method in research and development process | |
| CN112508671A (en) | Enterprise financial data processing method, system, device and medium | |
| CN110543996A (en) | job salary assessment method, apparatus, server and storage medium | |
| CN113127538B (en) | High-precision spare part demand prediction method | |
| CN113127537B (en) | Spare part demand prediction method integrating time sequence prediction model and machine learning model | |
| CN112001539B (en) | High-precision passenger transport prediction method and passenger transport prediction system | |
| DE102007013340A1 (en) | System and method for managing environmental impacts of companies | |
| CN112256681A (en) | Air traffic control digital index application system and method | |
| CN111027799A (en) | National enterprise productivity analysis system | |
| CN102855354A (en) | Industrial-enterprise-oriented online multi-dimensional energy consumption data statistics modeling method | |
| Salvino et al. | A direct test of direct democracy: New England town meetings | |
| CN114781855A (en) | DEA model-based logistics transmission efficiency analysis method, device, equipment and medium | |
| Zhao et al. | Hadoop-Based Power Grid Data Quality Verification and Monitoring Method | |
| CN117171145B (en) | Analysis processing method, equipment and storage medium for enterprise management system data | |
| Li | Economic Development Analysis of China’s Coastal Industries Based on Deep Learning and SDA Model Weighting | |
| Álvarez-Chaves et al. | Machine Learning Methods for Predicting the Admissions and Hospitalisations in the Emergency Department of a Civil and Military Hospital | |
| Priantari et al. | Designing statistical data processing system platform based on SOA |
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 |