Supply chain logistics information management system based on artificial intelligence

Abstract

The invention discloses a supply chain logistics information management system based on artificial intelligence, and particularly relates to the technical field of artificial intelligence, and the system comprises a linear fitting module, a nonlinear fitting module, a dual-mode construction module, a distribution verification module, and a management intervention module. The non-linear fitting module obtains a non-linear special coefficient according to the non-linear scene information, the dual-mode construction module constructs a dual-mode management model according to the linear conventional coefficient and the non-linear special coefficient and outputs a time cost delay index, and the distribution verification module classifies delay states in a supply chain logistics information management process. According to the supply chain logistics information management system and method, time cost prediction is carried out on customs declaration and tax payment links which are most time-consuming and labor-consuming in the supply chain logistics information management process, and the management effectiveness and real-time performance of the whole logistics information management chain are improved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and more specifically, to a supply chain logistics information management system based on artificial intelligence. Background Art

[0002] The AI-based supply chain logistics information management system is a supply chain logistics management solution that integrates AI technology. Through intelligent algorithms and big data analysis, it monitors, optimizes and manages the supply chain logistics process in real time, improves logistics efficiency, reduces costs, and realizes intelligent management of the supply chain.

[0003] Existing supply chain logistics information management systems based on artificial intelligence usually build scenarios based on continuous linear data, but logistics information data may have complex nonlinear relationships. Simple linear models cannot meet the management requirements under nonlinear scenarios. There are great differences in logistics customs declaration and tax payment links, which in turn affects the subsequent ground distribution preparation links. Therefore, it is an urgent problem to build a prediction model based on artificial intelligence technology that can meet both linear and nonlinear scenarios, predict the time cost of customs declaration and tax payment links, and then verify the real-time conditions of ground distribution.

[0004] In order to solve the above defects, a technical solution is now proposed. Summary of the invention

[0005] The purpose of the present invention is to provide a supply chain logistics information management system based on artificial intelligence to solve the deficiencies in the background technology.

[0006] In order to achieve the above-mentioned object, the present invention provides the following technical solutions: a supply chain logistics information management system based on artificial intelligence, comprising a linear fitting module, a nonlinear fitting module, a dual-mode construction module, a distribution verification module, and a management intervention module;

[0007] The linear fitting module is used to collect linear scenario information of the supply chain logistics information management system, obtain linear conventional coefficients according to the linear scenario information, and transmit the linear conventional coefficients to the dual-mode construction module;

[0008] The nonlinear fitting module is used to collect nonlinear scenario information of the supply chain logistics information management system, obtain nonlinear special coefficients according to the nonlinear scenario information, and transmit the nonlinear special coefficients to the dual-mode construction module;

[0009] The dual-mode construction module is used to construct a dual-mode management model based on linear conventional coefficients and nonlinear special coefficients, evaluate the delay status of customs declaration and tax payment in the supply chain logistics information management process, and output the time cost delay index;

[0010] The delivery verification module is used to classify the delay status in the supply chain logistics information management process by comparing the time cost delay index with the preset time cost delay threshold;

[0011] The management intervention module is used to perform intervention management based on the grading results.

[0012] Preferably, linear scenario information and nonlinear scenario information are extracted by collecting data, and the collected data include cargo type, customs policy status, region code, seasonal status, cargo quantity, and cargo value. The cargo is divided into different types according to the nature of the cargo, and the cargo type is converted into numerical form using unique hot coding. The customs policy status includes policy changes, unchanged policies, and sampling ratios. The customs policy status is converted into numerical form using unique hot coding. The seasonal status is expressed as a date and converted into time series data.

[0013] Preferably, the method for obtaining linear scene information is:

[0014] Obtain the interaction fluctuation coefficient of the supply chain logistics information management system within the cycle time T, and integrate the interaction fluctuation coefficients of several supply chain logistics information management systems within the cycle time T into a data set, and mark the data set as XX = {xx p}, where p = {1, 2, 3, ..., o}, and o is a positive integer, xx p It represents the interaction fluctuation coefficient of the supply chain logistics information management system within the pth cycle time T. The calculation expression of the interaction fluctuation coefficient is: Among them, X1 represents the type of goods, X2 is a categorical variable, which is converted into a numerical value {X 1，1 , X 1，2 , ..., X 1，i}, where i is the numerical sequence number of the one-hot encoding and is a positive integer, X2 represents the seasonal state, and represents the date.

[0015] Preferably, the method for obtaining the linear normal coefficient according to the linear scene information is as follows:

[0016] Calculate the standard deviation of the interaction fluctuation coefficient of the supply chain logistics information management system within several cycle times T. The calculation expression is: In the formula, xxa is the average value of the interaction fluctuation coefficient of the supply chain logistics information management system within several cycle times T, and the calculation expression is:

[0017] The least squares method is used to fit the linear regression model and calibrate the customs policy state as C ps , area code is A c , the quantity of goods is G q , the value of the goods is G v, then the expression of the linear regression model is Where, L nc are linear normal coefficients, β1, β2, and β3 are fc , (G q +G v ), and β1, β2, β3 are all positive numbers, ∈ is a very small positive number representing the error term.

[0018] Preferably, the method for obtaining nonlinear scene information is:

[0019] Obtain the nonlinear composite coefficient of the supply chain logistics information management system within the cycle time T, and integrate the nonlinear composite coefficients of several supply chain logistics information management systems within the cycle time T into a data set, and mark the data set as Non = {non u}, where u = {1, 2, 3, ..., y}, and y is a positive integer, non u It represents the nonlinear composite coefficient of the supply chain logistics information management system within the u-th cycle time T. The calculation expression of the nonlinear composite coefficient is: Where X2 is the seasonal state converted into a time series form, P c is the number of customs policy changes, and P c Is a non-negative number.

[0020] Preferably, the method for obtaining the nonlinear special coefficient according to the nonlinear scene information is:

[0021] Calculate the standard deviation of the nonlinear composite coefficient of the supply chain logistics information management system within several cycle times T. The calculation expression is: In the formula, nona is the average value of the nonlinear composite coefficient of the supply chain logistics information management system within several cycle times T, and the calculation expression is:

[0022] Construct a random forest model, for each tree T of the random forest r Train them separately and generate prediction values, average the prediction values ​​of all trees and output the average prediction value. The mathematical model for calculating the average prediction value is: Where N sp is the nonlinear special coefficient, T r (A) is the predicted value of the rth tree in the random forest, r is the tree number and r is a positive integer, w is the total number of trees, A is the input feature vector, and the input features include cargo type, customs policy status, region code, seasonal status, cargo quantity, and cargo value.

[0023] Preferably, the construction logic of the dual-mode management model is:

[0024] The dual-mode management model is defined by combining weighted average and rule setting. When a specific condition is triggered, the rule setting selection model is used; when the specific condition is not triggered, the weighted average model is used. The expression of the dual-mode management model is Where, T cd is the time cost delay index, ωω1 and ω2 are the weight coefficients of the linear general coefficient and the nonlinear special coefficient respectively, and ω1 and ω2 are both positive numbers. If Rule(S)=False, the weighted average result of linear regression and nonlinear fitting is output. If Rule(S)=True, the result of nonlinear fitting is output, where Rule(S) is the logical rule defined according to the input feature S, and In the formula, when the cargo type is dangerous goods or precision instruments, or the seasonal status is holidays, S ty The value is 1, otherwise, S ty The value is 0.

[0025] Preferably, the logic for grading the delay status in the supply chain logistics information management process is:

[0026] The preset time cost delay threshold is T th , the calculated time cost delay index T cd With time cost delay threshold T th For comparison, when the time cost delay index T cd Greater than or equal to the time cost delay threshold T th When , the process is marked as delayed and the delay state is defined as the diffusion level;

[0027] When the time cost delay index T cd Less than the time cost delay threshold T th , the marking process is not delayed and the delay state is defined as the contraction level.

[0028] Preferably, the logic of intervention management according to the grading results is:

[0029] When the delay status is at the diffusion level, inform the management personnel to postpone the ground delivery and wait; when the delay status is at the contraction level, inform the management personnel to depart on time for the ground delivery.

[0030] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0031] This application optimizes the management of supply chain logistics information by designing linear models suitable for normal conditions and nonlinear models for special conditions respectively, ensuring that under special circumstances, time cost predictions are made for the most time-consuming and labor-intensive customs declaration and tax payment links in the supply chain logistics information management process, thereby improving the management effectiveness and real-time performance of the entire logistics information management chain, enhancing the accuracy and applicability of time cost predictions, reducing the cost and accuracy of predictions using linear models under normal circumstances, and improving the interpretability of fitting predictions using nonlinear models under special circumstances, thereby effectively improving the management efficiency of supply chain logistics information. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0033] Figure 1 It is a system module diagram of the present invention. DETAILED DESCRIPTION

[0034] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] Example 1: Please refer to Figure 1 As shown, the present invention is a supply chain logistics information management system based on artificial intelligence, including a linear fitting module, a nonlinear fitting module, a dual-mode construction module, a distribution verification module, and a management intervention module;

[0036] The linear fitting module is used to collect linear scenario information of the supply chain logistics information management system, obtain linear conventional coefficients according to the linear scenario information, and transmit the linear conventional coefficients to the dual-mode construction module;

[0037] The nonlinear fitting module is used to collect nonlinear scenario information of the supply chain logistics information management system, obtain nonlinear special coefficients according to the nonlinear scenario information, and transmit the nonlinear special coefficients to the dual-mode construction module;

[0038] The dual-mode construction module is used to construct a dual-mode management model based on linear conventional coefficients and nonlinear special coefficients, evaluate the delay status of customs declaration and tax payment in the supply chain logistics information management process, and output the time cost delay index;

[0039] The delivery verification module is used to classify the delay status in the supply chain logistics information management process by comparing the time cost delay index with the preset time cost delay threshold;

[0040] The management intervention module is used to perform intervention management based on the grading results.

[0041] Nonlinear data means that the relationship between variables does not conform to a simple linear model. The relationship between the dependent variable and the independent variable cannot be represented by a straight line, but rather by a curve or a complex pattern. This means that the relationship between the data is more complex, and there may be multiple factors acting at the same time, and the effect of the effect will change as conditions change. For the same type of logistics goods, the type, quantity and value of the goods are roughly the same, the customs policy is stable and the sampling pattern is regular, and the tax payment rate basically does not fluctuate. The time for customs declaration and tax payment basically increases linearly with the number of goods, and a simple linear regression model can be used for prediction. However, for different types of goods, for example, there are both low-value, low-risk ordinary goods and high-value, high-risk special goods. Customs policies will be adjusted in different time periods, and sometimes there will be special situations such as sampling inspections. The calculation of taxes and fees involves a complex combination of tax types, which leads to a sharp increase in tax payment time under certain conditions. In such a complex situation, the relationship between the data is no longer a linear relationship, but presents nonlinear characteristics, which is difficult to accurately predict with a simple linear model;

[0042] Characteristics of nonlinear data include: changes in the independent variable do not cause the dependent variable to change at a fixed rate; there are complex interactions between multiple independent variables; when a variable exceeds a certain threshold, the impact on the dependent variable changes dramatically, while the impact is small below the threshold.

[0043] Linear scenario information and nonlinear scenario information are extracted through data collection. The collected data include cargo type, customs policy status, region code, seasonal status, cargo quantity, and cargo value. The cargo is divided into different types according to the nature of the cargo, and the cargo type is converted into numerical form using unique hot coding. The customs policy status includes policy changes, unchanged policies, and sampling ratio. The customs policy status is converted into numerical form using unique hot coding. The seasonal status is expressed as a date and converted into time series data.

[0044] By obtaining the linear scenario information of the supply chain logistics information management system, the effectiveness of the linear model of the supply chain logistics information management system is evaluated, and the linear conventional coefficient is obtained to measure the linear effectiveness of the supply chain logistics information management system;

[0045] The linear fitting module is used to collect linear scenario information of the supply chain logistics information management system, obtain linear conventional coefficients according to the linear scenario information, and transmit the linear conventional coefficients to the dual-mode construction module;

[0046] The method for obtaining linear scene information is:

[0047] Obtain the interaction fluctuation coefficient of the supply chain logistics information management system within the cycle time T, and integrate the interaction fluctuation coefficients of several supply chain logistics information management systems within the cycle time T into a data set, and mark the data set as XX = {xx p}, where p = {1, 2, 3, ..., o}, and o is a positive integer, xx p It represents the interaction fluctuation coefficient of the supply chain logistics information management system within the pth cycle time T. The calculation expression of the interaction fluctuation coefficient is: Among them, X1 represents the type of goods, X2 is a categorical variable, which is converted into a numerical value {X 1，1 , X 1，2 , ..., X 1，i}, where i is the numerical sequence number of the one-hot encoding and i is a positive integer, X2 represents the seasonal state, and represents the date;

[0048] It should be noted that the type of goods is provided by the supplier or manufacturer at the time of shipment, the customs policy status is obtained through the customs department, government statistics website and professional international trade data platform, the regional code is generated based on the logistics information of the place of shipment and destination of the goods, the quantity of goods is obtained from the supplier's delivery note, and the value of the goods is obtained from the supplier's invoice or purchase contract.

[0049] The method for obtaining linear general coefficients based on linear scene information is as follows:

[0050] Calculate the standard deviation of the interaction fluctuation coefficient of the supply chain logistics information management system within several cycle times T. The calculation expression is: In the formula, xxa is the average value of the interaction fluctuation coefficient of the supply chain logistics information management system within several cycle times T, and the calculation expression is:

[0051] The least squares method is used to fit the linear regression model and calibrate the customs policy state as C ps , area code is A c , the quantity of goods is G q , the value of the goods is G v , then the expression of the linear regression model is Where, L nc are linear normal coefficients, β1, β2, and β3 are fc , (G q +G v ), and β1, β2, β3 are all positive numbers, ∈ is a very small positive number representing the error term.

[0052] In the process of calculating the linear normal coefficients, the linear scenario information is obtained by interactively combining the cargo type and seasonal status, which combines the characteristic significance of the cargo type and seasonal status, effectively and deeply captures the cyclical laws of different types of cargo, and improves the accuracy of prediction using linear models.

[0053] The nonlinear fitting module is used to collect nonlinear scenario information of the supply chain logistics information management system, obtain nonlinear special coefficients according to the nonlinear scenario information, and transmit the nonlinear special coefficients to the dual-mode construction module;

[0054] The method for obtaining nonlinear scene information is:

[0055] Obtain the nonlinear composite coefficient of the supply chain logistics information management system within the cycle time T, and integrate the nonlinear composite coefficients of several supply chain logistics information management systems within the cycle time T into a data set, and mark the data set as Non = {non u}, where u = {1, 2, 3, ..., y}, and y is a positive integer, non u It represents the nonlinear composite coefficient of the supply chain logistics information management system within the u-th cycle time T. The calculation expression of the nonlinear composite coefficient is: Where X2 is the seasonal state converted into a time series form, P c is the number of customs policy changes, and p c is a non-negative number;

[0056] The method for obtaining nonlinear special coefficients based on nonlinear scene information is:

[0057] Calculate the standard deviation of the nonlinear composite coefficient of the supply chain logistics information management system within several cycle times T. The calculation expression is: In the formula, nona is the average value of the nonlinear composite coefficient of the supply chain logistics information management system within several cycle times T, and the calculation expression is:

[0058] Construct a random forest model, for each tree T of the random forest r Train them separately and generate prediction values, average the prediction values ​​of all trees and output the average prediction value. The mathematical model for calculating the average prediction value is: Where N sp is the nonlinear special coefficient, T r (A) is the predicted value of the rth tree in the random forest, r is the tree number and r is a positive integer, w is the total number of trees, A is the input feature vector, and the input features include cargo type, customs policy status, region code, seasonal status, cargo quantity, and cargo value.

[0059] The dual-mode construction module is used to construct a dual-mode management model based on linear conventional coefficients and nonlinear special coefficients, evaluate the delay status of customs declaration and tax payment in the supply chain logistics information management process, and output the time cost delay index;

[0060] The construction logic of the dual-mode management model is:

[0061] The dual-mode management model is defined by combining weighted average and rule setting. When a specific condition is triggered, the rule setting selection model is used; when the specific condition is not triggered, the weighted average model is used. The expression of the dual-mode management model is Where, T cd is the time cost delay index, ω1 and ω2 are the weight coefficients of the linear general coefficient and the nonlinear special coefficient respectively, and ω1 and ω2 are both positive numbers. If Rule(S)=False, the weighted average result of linear regression and nonlinear fitting is output. If Rule(S)=True, the result of nonlinear fitting is output, where Rule(S) is the logical rule defined according to the input feature S, and In the formula, when the cargo type is dangerous goods or precision instruments, or the seasonal status is holidays, S ty The value is 1, otherwise, S ty The value is 0.

[0062] The delivery verification module is used to classify the delay status in the supply chain logistics information management process by comparing the time cost delay index with the preset time cost delay threshold;

[0063] The preset time cost delay threshold is T th , the calculated time cost delay index T cd With time cost delay threshold T th For comparison, when the time cost delay index T cd Greater than or equal to the time cost delay threshold T th When , the process is marked as delayed and the delay state is defined as the diffusion level;

[0064] When the time cost delay index T cd Less than the time cost delay threshold T th , the marking process is not delayed and the delay state is defined as the contraction level.

[0065] The management intervention module is used to perform intervention management according to the grading results;

[0066] When the delay status is at the diffusion level, inform the management personnel to postpone the ground delivery and wait; when the delay status is at the contraction level, inform the management personnel to depart on time for the ground delivery.

[0067] This application optimizes the management of supply chain logistics information by designing linear models suitable for normal conditions and nonlinear models for special conditions respectively, ensuring that under special circumstances, time cost predictions are made for the most time-consuming and labor-intensive customs declaration and tax payment links in the supply chain logistics information management process, thereby improving the management effectiveness and real-time performance of the entire logistics information management chain, enhancing the accuracy and applicability of time cost predictions, reducing the cost and accuracy of predictions using linear models under normal circumstances, and improving the interpretability of fitting predictions using nonlinear models under special circumstances, thereby effectively improving the management efficiency of supply chain logistics information.

[0068] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0069] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of computer program goods. The computer program goods include one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0070] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0071] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0072] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0073] If the functions are implemented in the form of software functional units and sold or used as independent goods, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can essentially or in other words, the part that contributes to the prior art or the part of the technical solution can be embodied in the form of software goods, and the computer software goods are stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0074] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. Supply chain logistics information management system based on artificial intelligence, characterized by: It includes linear fitting module, nonlinear fitting module, dual-mode construction module, distribution verification module and management intervention module; The linear fitting module is used to collect linear scenario information of the supply chain logistics information management system, obtain linear conventional coefficients according to the linear scenario information, and transmit the linear conventional coefficients to the dual-mode construction module; The nonlinear fitting module is used to collect nonlinear scenario information of the supply chain logistics information management system, obtain nonlinear special coefficients according to the nonlinear scenario information, and transmit the nonlinear special coefficients to the dual-mode construction module; The dual-mode construction module is used to construct a dual-mode management model based on linear conventional coefficients and nonlinear special coefficients, evaluate the delay status of customs declaration and tax payment in the supply chain logistics information management process, and output the time cost delay index; The delivery verification module is used to classify the delay status in the supply chain logistics information management process by comparing the time cost delay index with the preset time cost delay threshold; The management intervention module is used to perform intervention management based on the grading results.

2. The supply chain logistics information management system based on artificial intelligence according to claim 1 is characterized in that: Linear scenario information and nonlinear scenario information are extracted through data collection. The collected data include cargo type, customs policy status, region code, seasonal status, cargo quantity, and cargo value. The cargo is divided into different types according to the nature of the cargo, and the cargo type is converted into numerical form using unique hot coding. The customs policy status includes policy changes, unchanged policies, and sampling ratio. The customs policy status is converted into numerical form using unique hot coding. The seasonal status is expressed as a date and converted into time series data.

3. The supply chain logistics information management system based on artificial intelligence according to claim 2 is characterized in that: The method for obtaining linear scene information is: Obtain the interaction fluctuation coefficient of the supply chain logistics information management system within the cycle time T, and integrate the interaction fluctuation coefficients of several supply chain logistics information management systems within the cycle time T into a data set, and mark the data set as XX = {xx p }, where p = {1, 2, 3, ..., o}, and o is a positive integer, xx p It represents the interaction fluctuation coefficient of the supply chain logistics information management system within the pth cycle time T. The calculation expression of the interaction fluctuation coefficient is Among them, X1 represents the type of goods, X2 is a categorical variable, which is converted into a numerical value {X 1,1 ,X 1,2 ,…,X 1,i }, where i is the numerical sequence number of the one-hot encoding and is a positive integer, X2 represents the seasonal state, and represents the date.

4. The supply chain logistics information management system based on artificial intelligence according to claim 3 is characterized in that: The method for obtaining linear general coefficients based on linear scene information is as follows: Calculate the standard deviation of the interaction fluctuation coefficient of the supply chain logistics information management system within several cycle times T. The calculation expression is: In the formula, xxa is the average value of the interaction fluctuation coefficient of the supply chain logistics information management system within several cycle times T, and the calculation expression is: The least squares method is used to fit the linear regression model and calibrate the customs policy state as C ps , area code is A c , the quantity of goods is G q , the value of the goods is G v , then the expression of the linear regression model is Where, L nc are linear normal coefficients, β1, β2, and β3 are fc , (G q +G v ), and β1, β2, β3 are all positive numbers, ∈ is a very small positive number representing the error term.

5. The supply chain logistics information management system based on artificial intelligence according to claim 2 is characterized in that: The method for obtaining nonlinear scene information is: Obtain the nonlinear composite coefficient of the supply chain logistics information management system within the cycle time T, and integrate the nonlinear composite coefficients of several supply chain logistics information management systems within the cycle time T into a data set, and mark the data set as Non = {{non u }, where u = {1,2,3,…,y}, and y is a positive integer, non u It represents the nonlinear composite coefficient of the supply chain logistics information management system within the u-th cycle time T. The calculation expression of the nonlinear composite coefficient is: Where X2 is the seasonal state converted into a time series form, P c is the number of customs policy changes, and P c Is a non-negative number.

6. The supply chain logistics information management system based on artificial intelligence according to claim 5 is characterized in that: The method for obtaining nonlinear special coefficients based on nonlinear scene information is: Calculate the standard deviation of the nonlinear composite coefficient of the supply chain logistics information management system within several cycle times T. The calculation expression is: In the formula, nona is the average value of the nonlinear composite coefficient of the supply chain logistics information management system within several cycle times T, and the calculation expression is: Construct a random forest model, for each tree T of the random forest r Train them separately and generate prediction values, average the prediction values ​​of all trees and output the average prediction value. The mathematical model for calculating the average prediction value is: Where N sp is the nonlinear special coefficient, T r (A) is the predicted value of the rth tree in the random forest, r is the tree number and r is a positive integer, w is the total number of trees, A is the input feature vector, and the input features include cargo type, customs policy status, region code, seasonal status, cargo quantity, and cargo value.

7. The supply chain logistics information management system based on artificial intelligence according to claim 6 is characterized in that: The construction logic of the dual-mode management model is: The dual-mode management model is defined by combining weighted average and rule setting. When a specific condition is triggered, the rule setting selection model is used; when the specific condition is not triggered, the weighted average model is used. The expression of the dual-mode management model is Where, T cd is the time cost delay index, ω1 and ω2 are the weight coefficients of the linear general coefficient and the nonlinear special coefficient respectively, and ω1 and ω2 are both positive numbers. If Rule(S)=False, the weighted average result of linear regression and nonlinear fitting is output. If Rule(S)=True, the result of nonlinear fitting is output, where Rule(S) is the logical rule defined according to the input feature S, and In the formula, when the cargo type is dangerous goods or precision instruments, or the seasonal status is holidays, S ty The value is 1, otherwise, S ty The value is 0.

8. The supply chain logistics information management system based on artificial intelligence according to claim 7 is characterized in that: The logic for grading the delay status in the supply chain logistics information management process is: The preset time cost delay threshold is T th , the calculated time cost delay index T cd With time cost delay threshold T th For comparison, when the time cost delay index T cd Greater than or equal to the time cost delay threshold T th When , the process is marked as delayed and the delay state is defined as the diffusion level; When the time cost delay index T cd Less than the time cost delay threshold T th , the marking process is not delayed and the delay state is defined as the contraction level.

9. The supply chain logistics information management system based on artificial intelligence according to claim 8 is characterized in that: The logic of intervention management based on the grading results is: When the delay status is at the diffusion level, inform the management personnel to postpone the ground delivery and wait; when the delay status is at the contraction level, inform the management personnel to depart on time for the ground delivery.

Images