CN107767094A - A kind of logistics ladder freight charges optimization method and device - Google Patents
A kind of logistics ladder freight charges optimization method and device Download PDFInfo
- Publication number
- CN107767094A CN107767094A CN201710865080.0A CN201710865080A CN107767094A CN 107767094 A CN107767094 A CN 107767094A CN 201710865080 A CN201710865080 A CN 201710865080A CN 107767094 A CN107767094 A CN 107767094A
- Authority
- CN
- China
- Prior art keywords
- distribution
- delivery
- data
- model
- test set
- 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
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
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/083—Shipping
- G06Q10/0834—Choice of carriers
- G06Q10/08345—Pricing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/083—Shipping
- G06Q10/0835—Relationships between shipper or supplier and carriers
- G06Q10/08355—Routing methods
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Operations Research (AREA)
- Data Mining & Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- Pure & Applied Mathematics (AREA)
- Computational Mathematics (AREA)
- Human Resources & Organizations (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Quality & Reliability (AREA)
- Mathematical Physics (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Bioinformatics & Computational Biology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Probability & Statistics with Applications (AREA)
- Evolutionary Biology (AREA)
- Algebra (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
Abstract
The invention discloses logistics ladder freight charges optimization method and device, it is related to field of computer technology.One embodiment of this method includes：History allocation data is received, to be divided into allocation data training set and allocation data test set；Regression modeling is carried out according to allocation data training set, the dispatching model of foundation is then applied to allocation data test set；It is determined that the error rate of the test set dispatching cost value of dispatching model output, with the dispatching model after being adjusted.The embodiment can solve the problem that the ladder freight rate of home-delivery center is coarse, and dispense the irrational problem of expense.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a logistics step freight optimization method and a logistics step freight optimization device.
Background
Currently, a large logistics network is divided into 2 levels, the first level is a branch transportation link from a regional distribution center to each county-level service store (the service store without transportation strength is carried by a local carrier), and the second level is a home delivery link from the county-level service store to a customer. In the actual operation process, the large delivery freight rate system is influenced by a plurality of factors such as the length of delivery distance, the living standard of local residents, the coverage rate of electric businesses, the natural geographic environment (mountainous areas/plains), the traffic condition, the bargaining condition of carriers and the like, so that different step freight rate systems exist in the delivery centers of each area.
In the process of implementing the invention, the inventor finds that at least the following problems exist in the prior art: the step freight rate of the distribution center is rough, the distribution volume of most branch lines can exceed 20 square along with the increase of the large piece coverage and the increase of the customer demand, and at the moment, the rough freight rate table is difficult to achieve the cost optimization. Meanwhile, the transportation cost of each branch line under the same ladder is very different sometimes, and it is difficult for related personnel to consider whether the price of each branch line is relatively reasonable on the whole, especially under the condition that branch lines are more.
Disclosure of Invention
In view of this, embodiments of the present invention provide a method and an apparatus for optimizing logistics step freight, which can solve the problems of rough step freight rate and unreasonable distribution cost of a distribution center.
To achieve the above object, according to an aspect of an embodiment of the present invention, there is provided a logistics step freight optimization method, including: receiving historical distribution data to divide the historical distribution data into a distribution data training set and a distribution data test set; performing regression modeling according to the distribution data training set, and then applying the established distribution model to the distribution data test set; error rates of test set distribution cost values output by the distribution model are determined to obtain an adjusted distribution model.
Optionally, the receiving historical distribution data to divide into a distribution data training set and a distribution data test set includes: forming an original distribution data set according to the received historical distribution data; cleaning the original distribution data set to obtain a modeling distribution data set; the modeling distribution data set is divided into a distribution data training set and a distribution data testing set.
Optionally, the delivery data includes delivery distance, amount of the delivery party, and delivery cost; the distribution distance and the distribution amount are input values, and the distribution cost is an output value.
Optionally, the performing regression modeling according to the delivery data training set, and then applying the established delivery model to the delivery data test set includes: performing regression modeling on the distribution data training set according to a linear regression model in a machine learning library to obtain a trained distribution model; and applying the distribution model to a distribution data test set to obtain a test set distribution cost value output by the distribution model.
Optionally, when performing regression modeling on the distribution data training set, the method includes: firstly, standardizing each characteristic data; and, the delivery distance and the delivery amount in the delivery data training set are extended to the two-dimensional feature space.
Optionally, the determining an error rate of the test set distribution cost value output by the distribution model to obtain the adjusted distribution model comprises: comparing the test set distribution cost value output by the distribution model with the historical distribution cost value in the distribution data test set to obtain the error rate of the test set distribution cost value output by the distribution model; judging whether the error rate of the output test set distribution expense value is less than or equal to a preset error rate threshold value or not, and if the error rate is less than or equal to the preset error rate threshold value, obtaining an adjusted distribution model; otherwise, adjusting the distribution model parameters, and carrying out regression modeling according to the distribution data training set until the error rate of the output test set distribution expense value is less than or equal to a preset error rate threshold value.
Optionally, the adjusted delivery model is:
where D represents the input delivery distance and V represents the input amount of the delivery party.
In addition, according to an aspect of an embodiment of the present invention, there is provided a logistics step freight optimization apparatus, including: the data preprocessing module is used for receiving historical distribution data to divide the historical distribution data into a distribution data training set and a distribution data testing set; the modeling module is used for carrying out regression modeling according to the distribution data training set and then applying the established distribution model to the distribution data testing set; and the model adjusting module is used for determining the error rate of the test set distribution expense value output by the distribution model so as to obtain the adjusted distribution model.
Optionally, the data preprocessing module receives historical delivery data to divide the historical delivery data into a delivery data training set and a delivery data testing set, and includes: forming an original distribution data set according to the received historical distribution data; cleaning the original distribution data set to obtain a modeling distribution data set; the modeling distribution data set is divided into a distribution data training set and a distribution data testing set.
Optionally, the delivery data includes delivery distance, amount of the delivery party, and delivery cost; the distribution distance and the distribution amount are input values, and the distribution cost is an output value.
Optionally, the modeling module performs regression modeling according to the delivery data training set, and then applies the established delivery model to the delivery data test set includes: performing regression modeling on the distribution data training set according to a linear regression model in the machine learning library to obtain a trained distribution model; and applying the distribution model to a distribution data test set to obtain a test set distribution cost value output by the distribution model.
Optionally, when performing regression modeling on the distribution data training set, the modeling module includes: firstly, standardizing each characteristic data; and, the delivery distance and the amount of the delivery in the delivery data training set are expanded to the two-dimensional feature space.
Optionally, the model adjustment module determining an error rate of the test set delivery cost value output by the delivery model to obtain an adjusted delivery model comprises: comparing the test set distribution cost value output by the distribution model with the historical distribution cost value in the distribution data test set to obtain the error rate of the test set distribution cost value output by the distribution model; judging whether the error rate of the output test set distribution expense value is less than or equal to a preset error rate threshold value or not, and if the error rate is less than or equal to the preset error rate threshold value, obtaining an adjusted distribution model; otherwise, adjusting the distribution model parameters, and carrying out regression modeling according to the distribution data training set until the error rate of the output test set distribution expense value is less than or equal to a preset error rate threshold value.
Optionally, the adjusted delivery model is:
where D represents the input delivery distance and V represents the input amount of the delivery party.
According to another aspect of the embodiments of the present invention, there is also provided an electronic device, including:
one or more processors;
a storage device to store one or more programs,
when executed by the one or more processors, cause the one or more processors to implement the method of any of the embodiments described above.
According to another aspect of the embodiments of the present invention, there is also provided a computer readable medium, on which a computer program is stored, which when executed by a processor implements the method of any of the above embodiments.
One embodiment of the above invention has the following advantages or benefits: because the technical means of logistics stepped freight optimization based on the regression model is adopted, the technical problems of rough stepped freight rate and unreasonable distribution cost are solved, a more detailed stepped freight rate table can be obtained, and the technical effect of finding abnormal distribution cost can be achieved.
Further effects of the above-mentioned non-conventional alternatives will be described below in connection with the embodiments.
Drawings
The drawings are included to provide a better understanding of the invention and are not to be construed as unduly limiting the invention. Wherein:
FIG. 1 is a schematic diagram of the main flow of a logistic staging freight optimization method according to an embodiment of the invention;
fig. 2 is a schematic view of a main flow of a logistics step freight optimization method according to a reference embodiment of the present invention;
FIG. 3 is a diagram illustrating the summary of the delivery costs output according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of the main modules of a logistics step freight optimization apparatus according to an embodiment of the present invention;
FIG. 5 is an exemplary system architecture diagram in which embodiments of the present invention may be employed;
fig. 6 is a schematic block diagram of a computer system suitable for use in implementing a terminal device or server of an embodiment of the invention.
Detailed Description
Exemplary embodiments of the present invention are described below with reference to the accompanying drawings, in which various details of embodiments of the invention are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the invention. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
Fig. 1 is a method for optimizing logistics step freight, according to an embodiment of the present invention, as shown in fig. 1, the method for optimizing logistics step freight includes:
step S101, receiving historical distribution data to divide the historical distribution data into a distribution data training set and a distribution data testing set.
In an embodiment, each of the historical dispatch data includes at least one attribute. Further, data from the data warehouse and the business side is received, and the attributes of each data include, but are not limited to, date, delivery order number, destination, delivery distance (m), amount (side) of the delivery side, delivery cost (element), etc., to form an original data set. Preferably, each of the historical delivery data includes a delivery distance, a number of delivery parties, and a delivery cost. The distribution distance and the distribution amount are input values, and the distribution cost is an output value.
Preferably, the original distribution data set may be formed based on the received historical distribution data. The raw delivery data set is then cleaned to obtain a modeled delivery data set. And finally, dividing the modeling distribution data set into a distribution data training set and a distribution data testing set.
Preferably, the cleaning of the original distribution data set specifically includes data culling in which at least one attribute in the original distribution data set is empty. Further, any one of the three fields of 'delivery distance', 'amount of delivery' and 'delivery cost' is blank data record. Alternatively, data records with a "delivery cost" field empty may be discarded.
In addition, the modeled delivery data set is preferably randomly sampled according to a scale of 1: 75% are the delivery data training set and 25% are the delivery data testing set.
And S102, performing regression modeling according to the distribution data training set, and applying the established distribution model to the distribution data test set.
As an embodiment, a linear regression model ElasticNet in a sklern machine learning library in Python language is used to perform regression modeling on a distribution data training set, so as to obtain a trained distribution model lineren, where the distribution model lineren is a linear elastic network model and may be described as lineren (linear ElasticNet). And then, applying the linear EN distribution model to a distribution data test set to obtain a test set distribution cost value output by the linear EN distribution model.
Where ElasticNet is a model of linear regression (a statistical analysis method that uses mathematical statistical regression analysis to determine the quantitative relationship of interdependencies between two or more variables) using L1 and L2 priors as regularizers. Such a combination is used for sparse models with very little non-zero weight, and elastic networks are very useful when multiple features are correlated with one another. In addition, the Python language is an object-oriented, interpreted computer programming language. Skearn in Python language provides skearn.
Preferably, in the process of performing regression modeling on the distribution data training set, standardScaler () is used to normalize each feature data, so as to avoid errors with different dimensions. The original features (delivery distance and delivery amount) were extended to a two-dimensional feature space using polymomial features (degree =2, include _bias = true). The search space for the optimal solution for l1_ ratio is: [0.1,0.3,0.5,0.7,0.99,1]. Other parameters may be defaulted.
Step S103, determining the error rate of the test set distribution expense value output by the distribution model to obtain the adjusted distribution model.
In an embodiment, the test set distribution cost value output by the linear en distribution model may be compared with the historical distribution cost value in the distribution data test set to obtain an error rate of the test set distribution cost value output by the linear en distribution model. Preferably, the error rate threshold is 2.01%.
Further, it may be determined whether the error rate of the output test set distribution cost value is less than or equal to a predetermined error rate threshold, and if so, an adjusted linear en distribution model may be obtained. Otherwise, adjusting the linear EN distribution model parameters and returning to the step S102.
Furthermore, the ElasticNet model under Sklearn package is mainlyDependent on 2 training result parameters: one is alpha = a + b and one is L1_ ratio = a/(a + b), where a and b correspond to the L1 canonical component coefficient and L2 canonical component coefficient, respectively, in the ElasticNet model. The adjusted model parameters are: alpha =0.3162, l1_ ratio =0.99. Preferably, the adjusted second-order (degree = 2) linear model y = W T The coefficient W of X is: w = [1612.40,398.33,1504.78, -23.78,317.22, -24.12]X is (1, D, V, D) 2 ,DV,V 2 ) Wherein D represents the input distance and V represents the input square quantity, as follows:
according to the various embodiments, the logistics step freight rate optimization method provided by the invention can obtain a more detailed step freight rate table, and can also find abnormal distribution routes possibly existing in history, so that a more detailed and reasonable step freight rate table is made, and a foundation is laid for subsequent refined operation.
Fig. 2 is a schematic diagram of a main flow of a logistics step freight rate optimization method according to a reference embodiment of the present invention, and the logistics step freight rate optimization method may include:
step S201, receiving historical distribution data to form an original distribution data set.
In an embodiment, historical data may be received, and each of the historical dispatch data includes at least one attribute. Further, data from the data warehouse and business parties is received, with attributes of each data including, but not limited to, "date, delivery order number, destination, delivery distance (m), amount (party) of delivery, delivery cost (element)" etc., forming the original data set.
Preferably, each of the historical delivery data includes a delivery distance, a delivery amount, and a delivery cost. The distribution distance and the distribution amount are input values, and the distribution cost is an output value.
That is, the delivery distance and the delivery volume are taken as features input in the machine learning algorithm model, and the delivery cost is taken as the label labels.
Step S202, cleaning the original distribution data set to obtain a modeling distribution data set.
In an embodiment, the original delivery data set may be cleaned, specifically including data culling in which at least one attribute in the original data set is empty. Preferably, any one of the three fields of 'delivery distance', 'amount of delivery' and 'delivery cost' is deleted to be an empty data record. Another preferred option is to cull data records with an empty "shipping cost" field.
Step S203, the modeling distribution data set is divided into a distribution data training set and a distribution data testing set.
Preferably, the modeled dispatch data set is randomly sampled according to a ratio of 1: 75% are the delivery data training set and 25% are the delivery data testing set.
And S204, performing regression modeling on the distribution data training set by using a linear regression model ElasticNet under a sklern machine learning library in a Python language to obtain a trained distribution model linerEN.
Preferably, in the process of performing regression modeling on the distribution data training set, standardScaler () is used to normalize each feature data, so as to avoid errors with different dimensions. The original features (delivery distance and delivery amount) were extended to a two-dimensional feature space using polymomial features (degree =2, include _bias = true). The search space for the optimal solution for l1_ ratio is: [0.1,0.3,0.5,0.7,0.99,1]. Other parameters may be defaulted.
And S205, applying the linear EN distribution model to a distribution data test set to obtain a test set distribution cost value output by the linear EN distribution model.
In step S206, it is determined whether the error rate of the output test set distribution cost value is less than or equal to a preset error rate threshold, if so, step S207 is performed, otherwise, the linear en distribution model parameter is adjusted and the step S204 is returned.
In an embodiment, the test set distribution cost value output by the linear en distribution model may be compared with the historical distribution cost value in the distribution data test set to obtain an error rate of the test set distribution cost value output by the linear en distribution model. Preferably, the error rate threshold is 2.01%.
Further, the ElasticNet model under sklern package depends mainly on 2 training result parameters: one is alpha = a + b and one is L1_ ratio = a/(a + b), where a and b correspond to the L1 canonical component coefficient and L2 canonical component coefficient, respectively, in the ElasticNet model. The adjusted model is: alpha =0.3162, l1_ ratio =0.99.
And step S207, obtaining the adjusted linear EN distribution model.
It should be noted that, according to the adjusted linear en distribution model obtained in step S207, a more detailed distribution freight ladder table can be output. Meanwhile, the adjusted linear EN distribution model can correct the distribution freight in the existing historical data.
Further, when a more detailed distribution freight step table needs to be output through the adjusted linear en distribution model, the specific implementation process is as follows:
and inputting a more detailed two-dimensional (distribution distance and distribution amount) step freight rate table into the adjusted linear EN distribution model, and outputting a detailed and reasonable distribution cost value by the linear EN distribution model. Preferably, a model output value having a delivery fee value of 0 may be excluded. For example: the more detailed two-dimensional ladder freight rate table expands the distance kilometer number range to (10 km, 500km) with the interval of 10km increasing progressively; the range of the amount V of the dispensing recipe is expanded to (1 m) 3 ，120m 3 ) At an interval of 1m 3 Increasing progressively; and (4) giving the delivery and transportation cost which should be paid in the detailed table through the adjusted linear EN delivery model.
Further, when the adjusted linear en delivery model is needed to correct the delivery freight in the existing historical data, the specific implementation process is as follows:
inputting the existing distribution distance and distribution amount data in the historical data into the adjusted linear EN distribution model, giving out a new distribution freight value by the adjusted linear EN distribution model, sequencing the distribution freight values from small to large, taking the 95-quantile cost value as feeFlag, outputting the record of the distribution freight value in the historical data, which is larger than the standard value feeFlag, as an abnormal value, and carrying out subsequent error correction.
FIG. 3 is a diagram illustrating a summary result of output delivery costs according to an embodiment of the present invention, and the left-most side is a historical delivery detail diagram; the middle part is a corresponding distribution detailed graph under the new model, and the scatter diagram of the new model is more regular than the historical detailed graph; the right-most side is the detailed step tariff (see table 1).
Table 1: detailed step tariff
It is therefore very intuitive to find which < distance-square > branches have high price off-spectra and which branches have had too high a cost value. In addition, according to the detailed ladder freight rate table, more detailed freight can be calculated for the subsequent operation details, and the subsequent refined operation is further perfected. That is to say, the logistics step freight optimization method can output a more detailed and reasonable step freight rate table (each distribution distance and distribution amount can be accurately obtained), so that the overall optimization of the distribution cost is realized, and a foundation is laid for subsequent fine operation. Meanwhile, the method can output branch delivery prices which may have abnormalities in history, and remind related personnel of paying attention, so that the problem of possible waste in the delivery link is solved.
In addition, the present invention can refer to the specific implementation contents of the logistics step freight rate optimization method in the embodiment, which have been described in detail in the above logistics step freight rate optimization method, so that the repeated contents are not described again.
Fig. 4 is a logistics step freight optimization apparatus according to an embodiment of the invention, and as shown in fig. 4, the logistics step freight optimization apparatus 400 includes a data preprocessing module 401, a modeling module 402, and a model adjusting module 403. The data preprocessing module 401 receives historical distribution data, and divides the historical distribution data into a distribution data training set and a distribution data testing set. The modeling module 402 performs regression modeling based on the delivery data training set and then applies the built delivery model to the delivery data test set. The model adjustment module 403 determines an error rate of the test set distribution cost values output by the distribution model to obtain an adjusted distribution model.
In a preferred embodiment, the data preprocessing module 401 may receive historical delivery data and then form an original delivery data set based on the received historical delivery data. Meanwhile, the original delivery data set may be cleaned to obtain the modeled delivery data set. And finally, dividing the modeling distribution data set into a distribution data training set and a distribution data testing set. Preferably, each of the historical delivery data includes a delivery distance, a delivery amount, and a delivery cost. The distribution distance and the distribution amount are input values, and the distribution cost is an output value.
Further, cleaning the original distribution data set, specifically including removing data with at least one attribute being null in the original data set. Further, any of the three fields of "delivery distance", "amount of delivery" and "delivery cost" is excluded, and there is a data record with one of the fields being empty. Alternatively, data records with a "delivery cost" field empty may be discarded. In addition, the modeled delivery data set is preferably randomly sampled according to a scale of 1: 75% are delivery data training sets and 25% are delivery data testing sets.
In another preferred embodiment, the modeling module 402 uses a linear regression model, namely ElasticNet, in Python language under a sklern machine learning library to perform regression modeling on the distribution data training set, so as to obtain a trained distribution model, namely lineren. And then, applying the linear EN distribution model to a distribution data test set to obtain a test set distribution cost value output by the linear EN distribution model.
Further, in the process of performing regression modeling on the distribution data training set, the modeling module 402 normalizes each feature data by using StandardScaler () to avoid errors with different dimensions. The original features (delivery distance and delivery amount) were extended to a two-dimensional feature space using polymomial features (degree =2, include _bias = true). The search space for the optimal solution for l1_ ratio is: [0.1,0.3,0.5,0.7,0.99,1]. Other parameters may be defaulted.
As another embodiment, the model adjustment module 403 may compare the test set distribution cost value output by the linear en distribution model with the historical distribution cost value in the distribution data test set to obtain the error rate of the test set distribution cost value output by the linear en distribution model. Preferably, the error rate threshold is 2.01%.
Further, the model adjusting module 403 may further determine whether the error rate of the output test set distribution cost value is less than or equal to a preset error rate threshold, and if the error rate of the output test set distribution cost value is less than or equal to the preset error rate threshold, an adjusted linear en distribution model is obtained. Otherwise, adjusting the parameters of the linear EN distribution model, and carrying out regression modeling according to the distribution data training set until the error rate of the output test set distribution expense value is less than or equal to a preset error rate threshold value.
Further, the adjusted model is: alpha =0.3162, l1_ ratio =0.99.
It should be noted that, in the implementation of the logistics step freight rate optimization device of the present invention, the above logistics step freight rate optimization method has been described in detail, so that the repeated content is not described herein.
Fig. 5 shows an exemplary system architecture 500 of a logistics step freight optimization method or a logistics step freight optimization apparatus to which an embodiment of the present invention can be applied.
As shown in fig. 5, the system architecture 500 may include terminal devices 501, 502, 503, a network 504, and a server 505. The network 504 is the medium used to provide communication links between terminal devices 501, 502, 503 and the server 505. Network 504 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.
The user may use the terminal devices 501, 502, 503 to interact with a server 505 over a network 504 to receive or send messages or the like. The terminal devices 501, 502, 503 may have installed thereon various communication client applications, such as shopping applications, web browser applications, search applications, instant messaging tools, mailbox clients, social platform software, etc. (by way of example only).
The terminal devices 501, 502, 503 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like.
The server 505 may be a server providing various services, such as a back-office management server (for example only) that provides support for shopping-like websites browsed by users using the terminal devices 501, 502, 503. The backend management server may analyze and perform other processing on the received data such as the product information query request, and feed back a processing result (for example, target push information, product information — just an example) to the terminal device.
It should be noted that the logistics step freight rate optimization method provided by the embodiment of the present invention is generally executed by the server 505, and accordingly, the logistics step freight rate optimization device is generally disposed in the server 505.
It should be understood that the number of terminal devices, networks, and servers in fig. 5 are merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.
Referring now to FIG. 6, a block diagram of a computer system 600 suitable for use with a terminal device implementing an embodiment of the invention is shown. The terminal device shown in fig. 6 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present invention.
As shown in fig. 6, the computer system 600 includes a Central Processing Unit (CPU) 601 that can perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM) 602 or a program loaded from a storage section 608 into a Random Access Memory (RAM) 603. In the RAM603, various programs and data necessary for the operation of the system 600 are also stored. The CPU 601, ROM 602, and RAM603 are connected to each other via a bus 604. An input/output (I/O) interface 605 is also connected to bus 604.
The following components are connected to the I/O interface 605: an input portion 606 including a keyboard, a mouse, and the like; an output portion 607 including a display such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker; a storage section 608 including a hard disk and the like; and a communication section 609 including a network interface card such as a LAN card, a modem, or the like. The communication section 609 performs communication processing via a network such as the internet. The driver 610 is also connected to the I/O interface 605 as needed. A removable medium 611 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 610 as necessary, so that the computer program read out therefrom is mounted in the storage section 608 as necessary.
In particular, according to the embodiments of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer-readable medium, the computer program comprising program code for performing the method illustrated by the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication section 609 and/or installed from the removable medium 611. The computer program performs the above-described functions defined in the system of the present invention when executed by the Central Processing Unit (CPU) 601.
It should be noted that the computer readable medium shown in the present invention can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present invention, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In the present invention, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems that perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The modules described in the embodiments of the present invention may be implemented by software or hardware. The described modules may also be provided in a processor, which may be described as: a processor includes a data pre-processing module, a modeling module, and a model adjustment module, where the names of the modules do not in some cases constitute a limitation on the modules themselves.
As another aspect, the present invention also provides a computer-readable medium that may be contained in the apparatus described in the above embodiments; or may be separate and not incorporated into the device. The computer readable medium carries one or more programs which, when executed by a device, cause the device to comprise: receiving historical distribution data to divide the historical distribution data into a distribution data training set and a distribution data testing set; performing regression modeling according to the distribution data training set, and then applying the established distribution model to the distribution data test set; and determining the error rate of the test set distribution cost value output by the distribution model so as to obtain the adjusted distribution model.
According to the technical scheme of the embodiment of the invention, a more detailed ladder freight rate table can be obtained, and abnormal delivery prices possibly existing in history can be found, so that the more detailed and reasonable ladder freight rate table is made, and a foundation is laid for subsequent fine operation.
The above-described embodiments should not be construed as limiting the scope of the invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions can occur, depending on design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (16)
1. A logistics step freight optimization method is characterized by comprising the following steps:
receiving historical distribution data to divide the historical distribution data into a distribution data training set and a distribution data test set;
performing regression modeling according to the distribution data training set, and then applying the established distribution model to the distribution data test set;
and determining the error rate of the test set distribution cost value output by the distribution model so as to obtain the adjusted distribution model.
2. The method of claim 1, wherein receiving historical dispatch data for partitioning into a dispatch data training set and a dispatch data test set comprises:
forming an original distribution data set according to the received historical distribution data;
cleaning the original distribution data set to obtain a modeling distribution data set;
the modeling distribution data set is divided into a distribution data training set and a distribution data testing set.
3. The method of claim 1 or 2, wherein the delivery data includes delivery distance, amount of the delivery party, and delivery cost; the distribution distance and the distribution amount are input values, and the distribution cost is an output value.
4. The method of claim 1, wherein the regression modeling based on a training set of delivery data and then applying the established delivery model to a test set of delivery data comprises:
performing regression modeling on the distribution data training set according to a linear regression model in a machine learning library to obtain a trained distribution model;
and applying the distribution model to a distribution data test set to obtain a test set distribution cost value output by the distribution model.
5. The method of claim 4, wherein the regression modeling of the delivery data training set comprises:
firstly, standardizing each characteristic data; and, the delivery distance and the amount of the delivery in the delivery data training set are expanded to the two-dimensional feature space.
6. The method of any of claims 1-5, wherein determining an error rate of test set distribution cost values output by the distribution model to obtain the adjusted distribution model comprises:
comparing the test set distribution cost value output by the distribution model with the historical distribution cost value in the distribution data test set to obtain the error rate of the test set distribution cost value output by the distribution model;
judging whether the error rate of the output test set distribution expense value is less than or equal to a preset error rate threshold value or not, and if the error rate of the output test set distribution expense value is less than or equal to the preset error rate threshold value, obtaining the adjusted distribution model; otherwise, adjusting the distribution model parameters, and carrying out regression modeling according to the distribution data training set until the error rate of the output test set distribution expense value is less than or equal to a preset error rate threshold value.
7. The method of claim 6, wherein the adjusted delivery model is:
where D represents the input delivery distance and V represents the input amount of the delivery party.
8. A logistics step freight optimization device is characterized by comprising:
the data preprocessing module is used for receiving historical distribution data to divide the historical distribution data into a distribution data training set and a distribution data testing set;
the modeling module is used for carrying out regression modeling according to the distribution data training set and then applying the established distribution model to the distribution data testing set;
and the model adjusting module is used for determining the error rate of the test set distribution expense value output by the distribution model so as to obtain the adjusted distribution model.
9. The apparatus of claim 8, wherein the data pre-processing module receives historical delivery data for partitioning into a delivery data training set and a delivery data test set, comprising:
forming an original distribution data set according to the received historical distribution data;
cleaning the original distribution data set to obtain a modeling distribution data set;
and dividing the modeling distribution data set into a distribution data training set and a distribution data test set.
10. The apparatus of claim 8 or 9, wherein the delivery data includes a delivery distance, a delivery party amount, and a delivery cost; the distribution distance and the distribution amount are input values, and the distribution cost is an output value.
11. The apparatus of claim 8, wherein the modeling module performs regression modeling based on a delivery data training set and then applies the established delivery model to a delivery data test set comprises:
performing regression modeling on the distribution data training set according to a linear regression model in the machine learning library to obtain a trained distribution model;
and applying the distribution model to a distribution data test set to obtain a test set distribution cost value output by the distribution model.
12. The apparatus of claim 11, wherein the modeling module, when performing regression modeling on the distribution data training set, comprises:
firstly, standardizing each characteristic data; and, the delivery distance and the delivery amount in the delivery data training set are extended to the two-dimensional feature space.
13. The apparatus of any of claims 8-12, wherein the model adjustment module determines an error rate of the testset delivery cost values output by the delivery model to obtain an adjusted delivery model comprises:
comparing the test set distribution cost value output by the distribution model with the historical distribution cost value in the distribution data test set to obtain the error rate of the test set distribution cost value output by the distribution model;
judging whether the error rate of the output test set distribution expense value is less than or equal to a preset error rate threshold value or not, and if the error rate of the output test set distribution expense value is less than or equal to the preset error rate threshold value, obtaining the adjusted distribution model; otherwise, adjusting the distribution model parameters, and carrying out regression modeling according to the distribution data training set until the error rate of the output test set distribution expense value is less than or equal to a preset error rate threshold value.
14. The apparatus of claim 13, wherein the adjusted delivery model is:
where D represents the inputted delivery distance and V represents the inputted amount of the delivery party.
15. An electronic device, comprising:
one or more processors;
a storage device for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement the method of any one of claims 1-7.
16. A computer-readable medium, on which a computer program is stored, which, when being executed by a processor, carries out the method according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710865080.0A CN107767094A (en) | 2017-09-22 | 2017-09-22 | A kind of logistics ladder freight charges optimization method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710865080.0A CN107767094A (en) | 2017-09-22 | 2017-09-22 | A kind of logistics ladder freight charges optimization method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107767094A true CN107767094A (en) | 2018-03-06 |
Family
ID=61266637
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710865080.0A Pending CN107767094A (en) | 2017-09-22 | 2017-09-22 | A kind of logistics ladder freight charges optimization method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107767094A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110414875A (en) * | 2018-04-26 | 2019-11-05 | 北京京东振世信息技术有限公司 | Capacity data processing method, device, electronic equipment and computer-readable medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105869017A (en) * | 2016-03-29 | 2016-08-17 | 上海携程商务有限公司 | Method and system for predicting ticket prices |
| CN106779240A (en) * | 2016-12-30 | 2017-05-31 | 中国民航信息网络股份有限公司 | The Forecasting Methodology and system of civil aviaton's market macroscopic view index |
| CN107808256A (en) * | 2017-11-20 | 2018-03-16 | 四川大学 | A kind of regional high voltage distribution network based on chance constrained programming turns supplier's method |
-
2017
- 2017-09-22 CN CN201710865080.0A patent/CN107767094A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105869017A (en) * | 2016-03-29 | 2016-08-17 | 上海携程商务有限公司 | Method and system for predicting ticket prices |
| CN106779240A (en) * | 2016-12-30 | 2017-05-31 | 中国民航信息网络股份有限公司 | The Forecasting Methodology and system of civil aviaton's market macroscopic view index |
| CN107808256A (en) * | 2017-11-20 | 2018-03-16 | 四川大学 | A kind of regional high voltage distribution network based on chance constrained programming turns supplier's method |
Non-Patent Citations (2)
| Title |
|---|
| 方启稳: "基于GA-BP算法的公路货运定价模型研究", 《中国优秀硕士学位论文全文数据库 经济与管理科学辑》 * |
| 郑层林: "大宗散货海铁联运的铁路运费测算", 《行政事业资产与财务》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110414875A (en) * | 2018-04-26 | 2019-11-05 | 北京京东振世信息技术有限公司 | Capacity data processing method, device, electronic equipment and computer-readable medium |
| CN110414875B (en) * | 2018-04-26 | 2022-09-06 | 北京京东振世信息技术有限公司 | Capacity data processing method and device, electronic equipment and computer readable medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111047243B (en) | Method and device for determining logistics distribution cost | |
| CN110751497A (en) | Commodity replenishment method and device | |
| CN107633358B (en) | Facility site selection and distribution method and device | |
| CN110648089A (en) | Method and device for determining delivery timeliness of articles | |
| CN110858347A (en) | Method and device for logistics distribution and order distribution | |
| CN110689157A (en) | Method and device for determining call relation | |
| CN113259144A (en) | Storage network planning method and device | |
| CN110866625A (en) | Promotion index information generation method and device | |
| CN113988768B (en) | Inventory determination method and device | |
| CN114663015A (en) | Replenishment method and device | |
| CN109978421B (en) | Information output method and device | |
| CN112418258A (en) | Feature discretization method and device | |
| CN114445102A (en) | Quotation data processing method and device | |
| CN110738508A (en) | data analysis method and device | |
| CN112784212B (en) | Inventory optimization method and device | |
| CN107767094A (en) | A kind of logistics ladder freight charges optimization method and device | |
| CN109902847B (en) | Method and device for predicting amount of orders in branch warehouse | |
| CN113450042A (en) | Method and device for determining replenishment quantity | |
| CN115099865A (en) | Data processing method and device | |
| CN110858335A (en) | Method and device for calculating sales promotion elasticity | |
| CN110956478A (en) | Method and device for determining goods input quantity | |
| CN113762674B (en) | Order distribution method and device | |
| CN113806047A (en) | Data operation optimization processing method and device | |
| CN113780611A (en) | Inventory management method and device | |
| CN112950240A (en) | Distribution method and device |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180306 |
|
| RJ01 | Rejection of invention patent application after publication |