Process node-based supply chain intelligent management and control method and system

Abstract

This invention discloses a supply chain intelligent management and control method and system based on process nodes, belonging to the field of supply chain management technology. It includes: a login management module for user security protection based on acquired relevant information; a logistics route planning module for generating candidate routes and selecting a transportation route from them; a carrier matching module for matching the optimal carrier from a carrier database; a cargo tracking module for real-time cargo tracking and determining if there are any transportation anomalies; and an early warning module for responding accordingly based on the judgment results. This invention selects transportation routes by quantifying positive and negative correlation information, simultaneously considering advantages such as transportation efficiency and cost, as well as risks such as congestion and weather, making the selected transportation routes more consistent with actual transportation needs. This avoids the one-sidedness of traditional route selection and improves transportation efficiency.

Description

Technical Field

[0001] This application relates to the field of supply chain management technology, specifically to intelligent supply chain control methods and systems based on process nodes. Background Technology

[0002] With the acceleration of global trade integration and the diversification and upgrading of consumer market demands, supply chain management has become a core element for enterprises to improve operational efficiency and reduce cost risks. Currently, most enterprises' supply chain management still relies on traditional manual scheduling and decentralized system collaboration models. While these can support basic operations, their intelligent, precise, and real-time capabilities in complex market environments are no longer adequate to meet actual needs. Specifically, these shortcomings are manifested in: 1. Existing supply chain systems mostly use a single authentication method for login and access control, which lacks multi-dimensional verification of user identity information. This can easily lead to security risks such as overstepping of permissions and information leakage, and cannot meet the differentiated security protection needs of multiple roles such as suppliers, logistics providers, and enterprise administrators in the supply chain. 2. Traditional route planning often relies solely on single factors such as distance and transportation costs, without fully integrating related information such as road conditions, weather, and warehouse node load. This results in insufficient flexibility in the planned routes and is prone to problems such as transportation delays and resource waste. 3. Existing carrier selection relies heavily on manual experience or fixed partnerships, and an intelligent matching mechanism based on multiple dimensions such as capacity, timeliness, and cost has not been established, making it difficult to quickly select the optimal carrier and affecting the efficiency of transportation resource allocation. Summary of the Invention

[0003] The purpose of this invention is to provide a supply chain intelligent management and control method and system based on process nodes, in order to solve the problems faced in the above-mentioned background art.

[0004] The objective of this invention can be achieved through the following technical solutions: A supply chain intelligent management and control system based on process nodes, the system comprising: The login management module is used to obtain relevant information of logged-in users and to protect user security based on the relevant information. The logistics route planning module is used to determine the candidate route based on the origin and destination of the order, and to determine the transportation route of the logistics based on the analysis of the correlation information of the candidate route. A carrier matching module is used to match the optimal carrier from the system's carrier database for transportation. The cargo tracking module is used to track cargo in real time, thereby determining whether there are any transportation abnormalities. The early warning module responds accordingly based on the judgment result.

[0005] Furthermore, the method for protecting user security based on relevant information is as follows: the relevant information includes the user's login information and access information; Based on the obtained login information, feature tags are extracted to obtain multiple real-time feature tags, and a real-time feature tag set is constructed based on the real-time feature tags; the user's historical login information is obtained, and corresponding feature tags are extracted from the historical login information to obtain multiple historical feature tags; a standard feature tag set is constructed using the historical feature tags as standard feature tags; the Jaccard similarity between the real-time feature tag set and the standard feature tag set is calculated and denoted as the login security coefficient. The login security coefficient is compared with a preset login security coefficient threshold. If the login security coefficient is less than the login security coefficient threshold, the user is prevented from logging in; otherwise, the user is allowed to log in normally.

[0006] Furthermore, methods for protecting user security based on relevant information also include: When a user is allowed to log in normally, their access information is obtained, including browsing traffic Q and the number of times access was exceeded (r), in a time series. Below, through the formula Calculate the access risk coefficient The access risk coefficient is compared with a preset access risk coefficient threshold. If the access risk coefficient is greater than the preset access risk coefficient threshold, the user is prevented from accessing the system further; otherwise, the user is allowed to continue accessing the system. in, Time series The proposed access and browsing traffic variation function is as follows. This is a function that sets the historical browsing traffic variation based on historical browsing traffic. Time series Down The points, Time series Down The points.

[0007] Furthermore, the method for determining the transportation route of materials is as follows: Based on the origin and destination of the order, multiple candidate transportation routes are automatically generated from historical transportation routes; positive and negative correlation information of each candidate transportation route is obtained, and the positive and negative correlation information is quantified to obtain multiple positive correlation values ​​and multiple negative correlation values ​​respectively; the positive correlation values ​​are weighted and accumulated to obtain the total positive correlation value, the negative correlation values ​​are weighted and accumulated to obtain the total negative correlation value, and the total positive correlation value is divided by the total negative correlation value to obtain the optimal value of each candidate transportation route; Based on the preference value, the candidate transportation routes are sorted in descending order, and the top three candidate transportation routes are recommended to the user. The user can then independently determine the transportation route for the materials from the three recommended candidate transportation routes.

[0008] Furthermore, the carrier matching module works as follows: Obtain order demand information, extract features from the demand information to obtain feature values ​​of multiple demand features of the order, and construct an order demand matrix based on the feature values ​​of each demand feature of the order. ; Demand information for each carrier is obtained from the carrier database. Feature extraction is performed on this demand information to obtain feature values ​​for multiple demand features of each carrier. Based on these feature values, a carrier demand matrix is ​​constructed. Each demand feature of the carrier in matrix B corresponds one-to-one with each demand feature in matrix A. Order demand matrix Carrier demand matrix By taking the difference, we obtain the difference matrix. ; Difference matrix The deviation coefficient is obtained by summing up the individual elements. , The carrier with the smallest deviation coefficient is selected as the optimal carrier for transportation. in, Let i be the feature value of the i-th demand feature of the order. Let n be the total number of demand features. Let be the feature value of the i-th demand feature of the carrier.

[0009] Furthermore, the cargo tracking module operates as follows: Transportation anomalies include cargo location anomalies and cargo status anomalies. GPS positioning is installed on transportation equipment to obtain cargo location in real time and determine whether the transportation time from the previous transportation node to the next transportation node exceeds the transportation time threshold. When the transportation time threshold is exceeded, it is determined that there is an anomaly in the cargo location, and the transportation time needs to be corrected in combination with environmental impact factors. Sensors are installed inside the cargo transport warehouse to acquire real-time data values ​​detected by the sensors. These real-time data values ​​are then compared with a preset standard data threshold range. If the real-time data value is outside the standard data threshold range, the cargo status is determined to be abnormal.

[0010] Furthermore, the method for obtaining environmental impact factors is as follows: Environmental impact information from the previous transportation node to the next transportation node is obtained, features are extracted, and parameter values ​​for multiple impact items are obtained. These values ​​are then calculated using formulas. The environmental impact factors were determined. in, Let j be the parameter value of the j-th influencing term. Let j be the standard parameter values ​​of the influencing terms. Let j be the deviation comparison value of the j-th influencing term. To the total number of items affected, .

[0011] The method for intelligent supply chain management based on process nodes is implemented through the intelligent supply chain management system based on process nodes as described in the above claims.

[0012] The beneficial effects of this invention are: This invention selects transportation routes by quantifying positive and negative correlation information. It can simultaneously take into account the advantages of transportation efficiency and cost, as well as the risks of congestion and weather, and conduct a comprehensive analysis to make the selected transportation routes more in line with actual transportation needs. This avoids the one-sidedness of traditional route selection and improves transportation efficiency.

[0013] This invention replaces subjective human judgment with a demand matrix combined with numerical deviation calculation, avoiding one-sided selections such as only looking at cost and ignoring capacity, or only looking at timeliness and ignoring damage rate. It can simultaneously cover core characteristics such as timeliness, cost, capacity, and damage rate, ensuring that multiple supply chain transportation needs can be met with a single match. It eliminates the need to screen carriers from different dimensions multiple times, making the matching results more accurate and objective, thereby improving matching efficiency. Attached Figure Description

[0014] Figure 1 This is a system module block diagram of the present invention. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] In one embodiment, a supply chain intelligent management and control system based on process nodes is disclosed, such as... Figure 1 As shown, the system mainly includes: The login management module is used to obtain relevant information about logged-in users and to protect user security based on that information. The logistics route planning module is used to determine the candidate routes based on the origin and destination of the order, and to determine the transportation route of the logistics based on the correlation information of the candidate routes. The carrier matching module is used to match the best carrier from the system's carrier database for transportation. The cargo tracking module is used to track cargo in real time, thereby determining whether there are any transportation abnormalities. The early warning module responds accordingly based on the judgment results.

[0017] Through the above technical solution, this application first obtains relevant information of logged-in users through the login management module, including user login information and access information, and protects user security based on the relevant information to ensure the security of logged-in users and the system; then, the logistics route planning module selects a suitable logistics transportation route for the current order. When selecting a transportation route, it can simultaneously consider the advantages of transportation efficiency and cost, as well as the risks of congestion and weather, and conduct a comprehensive analysis to make the selected transportation route more in line with actual transportation, avoiding the one-sidedness of traditional route selection and improving transportation efficiency; then, the carrier matching module matches the optimal carrier from the system's carrier database for transportation; finally, the cargo tracking module tracks the cargo in real time to determine whether there are any transportation anomalies, including judgment of cargo location anomalies and cargo status anomalies. Once an anomaly is detected, the early warning module can respond accordingly; this achieves intelligent management and control of the supply chain.

[0018] The method for user security protection based on relevant information is as follows: Relevant information includes the user's login information and access information; based on the acquired login information, feature tags are extracted to obtain multiple real-time feature tags, and a real-time feature tag set is constructed based on these tags; the user's historical login information is acquired, and corresponding feature tags are extracted from the historical login information to obtain multiple historical feature tags; a standard feature tag set is constructed using the historical feature tags as standard feature tags; the Jaccard similarity between the real-time feature tag set and the standard feature tag set is calculated and denoted as the login security coefficient; the login security coefficient is compared with a preset login security coefficient threshold; if the login security coefficient is less than the threshold, the user is prevented from logging in; otherwise, the user is allowed to log in normally. When a user is allowed to log in normally, their access information is obtained, including browsing traffic Q and the number of times access was exceeded (r), in a time series. Below, through the formula Calculate the access risk coefficient The access risk coefficient is compared with a preset access risk coefficient threshold. If the access risk coefficient is greater than the preset access risk coefficient threshold, the user is prevented from accessing the system further; otherwise, the user is allowed to continue accessing the system. in, Time series The proposed access and browsing traffic variation function is as follows. This is a function that sets the historical browsing traffic variation based on historical browsing traffic. Time series Down The points, Time series Down The points.

[0019] The above technical solution provides specific methods for user security protection. Firstly, relevant information includes the user's login information and access information. Login information includes login time period, login method, login IP, etc., while access information includes browsing traffic and number of access violations. Based on the obtained login information, feature tags are extracted to obtain multiple real-time feature tags. A real-time feature tag set is then constructed based on these tags. For example, after extracting login information features, feature tags such as login time period, login method, and login IP for this login are obtained and recorded as real-time feature tags, and a real-time feature tag set is constructed. Next, based on the user's historical login information, corresponding feature tags are extracted from historical login information to obtain multiple historical feature tags. Commonly used feature tags from these historical feature tags are used as the user's standard feature tags, constructing a user standard feature tag set. The user's standard feature tag set represents the user's security protection based on their login history. A trust set is generated based on user login information. The Jaccard similarity between the real-time feature tag set and the standard feature tag set is calculated and denoted as the login security coefficient. Obviously, the higher the login security coefficient, the higher the similarity between the real-time feature tag set and the standard feature tag set, indicating that the user's login is highly secure. A login security coefficient threshold is set based on the experience and professional knowledge of those in the field. The login security coefficient is compared with the preset login security coefficient threshold. When the login security coefficient is less than the login security coefficient threshold, it indicates that the user's login is significantly different from the usual login, indicating a high probability of anomalies, and the user's login is blocked; otherwise, the user is allowed to log in normally. By using a multi-dimensional feature tag Jaccard verification method to judge the user's login security, compared with traditional single password authentication, it can effectively identify unauthorized logins in uncommon scenarios, thereby improving account security.

[0020] When a user is allowed to log in normally, their access information is obtained, including browsing traffic Q and the number of times access was exceeded (r), in a time series. Below, through the formula Calculate the access risk coefficient The access risk coefficient is compared with a preset access risk coefficient threshold. If the access risk coefficient is greater than the preset threshold, further access is blocked; otherwise, the user is allowed to continue accessing the system. Time series The proposed access and browsing traffic variation function is as follows. This is a function that sets the historical browsing traffic variation based on historical browsing traffic. Time series Down The points, Time series Down The integral; This represents the ratio between changes in visitor traffic over a time series and changes in standard visitor traffic. A larger ratio indicates a higher likelihood of abnormal user activity. This indicates whether a user has exceeded their privileges; the higher the value, the greater the likelihood of abnormal access. Therefore, a comprehensive analysis of both factors yields an access risk coefficient. Based on the experience and expertise of those in the field, an access risk coefficient threshold is set. The access risk coefficient is compared with the preset threshold. When the access risk coefficient exceeds the preset threshold, it indicates a high risk of user access, and the access is deemed abnormal, preventing further access. Otherwise, the user is allowed to continue. This method, combined with dynamic calculation of access traffic and unauthorized behavior, can immediately intercept high-frequency access to core data and dangerous behaviors such as unauthorized access, avoiding the lag of manual inspection. Furthermore, dual authentication of login and access information further enhances the system's security performance.

[0021] The method for determining the transportation route of materials is as follows: Based on the origin and destination of the order, multiple candidate transportation routes are automatically generated from historical transportation routes; positive and negative correlation information of each candidate transportation route is obtained, and the positive and negative correlation information is quantified to obtain multiple positive correlation values ​​and multiple negative correlation values ​​respectively; the positive correlation values ​​are weighted and accumulated to obtain the total positive correlation value, the negative correlation values ​​are weighted and accumulated to obtain the total negative correlation value, and the total positive correlation value is divided by the total negative correlation value to obtain the optimal value of each candidate transportation route; Based on the preference value, the candidate transportation routes are sorted in descending order, and the top three candidate transportation routes are recommended to the user. The user can then independently determine the transportation route for the materials from the three recommended candidate transportation routes.

[0022] The above technical solution provides a specific method for the logistics route planning module to determine the material transportation route. First, based on the order's origin and destination, multiple candidate transportation routes are automatically generated from historical transportation routes. Then, positive and negative correlation information is obtained for each candidate route. Positive correlation information represents advantageous factors for the route, with higher values ​​being better, such as transportation timeliness (faster is better), transportation cost (lower is better), and road condition stability (more stable is better). Negative correlation information represents disadvantageous factors for the route, with lower values ​​being better, such as road congestion rate (lower is better) and weather impact probability (lower is better). This positive and negative correlation information is then quantified to obtain multiple positive and multiple negative correlation values. The quantitative processing method combines the actual performance of various indicators in historical transportation data, and maps the indicators corresponding to positive / negative correlation information to obtain correlation values. This quantitative method is an existing technology widely used in the field of logistics route planning, and will not be described in detail here. Then, the positive correlation values ​​are weighted and accumulated to obtain the total positive correlation value, and the negative correlation values ​​are weighted and accumulated to obtain the total negative correlation value. Dividing the total positive correlation value by the total negative correlation value yields the optimal value for each candidate transportation route. It can be seen that the larger the optimal value, the better the corresponding transportation route. Therefore, based on the size of the optimal value, the candidate transportation routes are sorted in descending order, and the top three candidate transportation routes are recommended to the user. The user can then independently determine the transportation route for the materials from the three recommended candidate transportation routes. By using the quantitative method of positive and negative correlation information to select transportation routes, the advantages of transportation efficiency and cost can be comprehensively analyzed along with the risks of congestion and weather, making the selected transportation routes more in line with actual transportation, avoiding the one-sidedness of traditional route selection, and improving transportation efficiency.

[0023] The carrier matching module works as follows: it obtains the order's demand information, extracts features from the demand information to obtain feature values ​​of multiple demand features of the order, and constructs an order demand matrix based on the feature values ​​of each demand feature of the order. ; Demand information for each carrier is obtained from the carrier database. Feature extraction is performed on this demand information to obtain feature values ​​for multiple demand features of each carrier. Based on these feature values, a carrier demand matrix is ​​constructed. Each demand feature of the carrier in matrix B corresponds one-to-one with each demand feature in matrix A. Order demand matrix Carrier demand matrix By taking the difference, we obtain the difference matrix. ; Difference matrix The deviation coefficient is obtained by summing up the individual elements. , The carrier with the smallest deviation coefficient is selected as the optimal carrier for transportation. in, Let i be the feature value of the i-th demand feature of the order. Let n be the total number of demand features. Let be the feature value of the i-th demand feature of the carrier.

[0024] The above technical solution provides a specific method for selecting the optimal carrier for transportation. First, it obtains the order's demand information, which may include transportation timeliness, transportation cost, capacity, and cargo damage rate. Feature extraction is then performed on this demand information to obtain feature values ​​for multiple demand features of the order. The feature values ​​of each demand feature can be obtained by numerical mapping based on historical data from relevant fields. This method of acquisition is a conventional technical means of order and carrier information processing in the logistics supply chain field, and will not be elaborated further. Based on the feature values ​​of each demand feature of the order, an order demand matrix is ​​constructed. Then, demand information for each carrier is retrieved from the carrier database. Feature extraction is performed on the demand information of each carrier to obtain feature values ​​of multiple demand features of the carrier. Based on the feature values ​​of each demand feature of the carrier, a carrier demand matrix is ​​constructed. Each demand characteristic of the carrier in matrix B corresponds one-to-one with each demand characteristic in matrix A; the order demand matrix is ​​then... Carrier demand matrix By taking the difference, we obtain the difference matrix. Finally, the difference matrix The deviation coefficient is obtained by summing up the individual elements. , The carrier with the smallest deviation coefficient is selected as the optimal carrier for transportation; difference matrix In this model, smaller element values ​​indicate that the corresponding carrier better meets the order's requirements. Therefore, by accumulating the values ​​of each element, a deviation coefficient is obtained. It is known that a smaller deviation coefficient indicates a better fit for the order's requirements. Thus, the carrier with the smallest deviation coefficient is selected as the optimal carrier for transportation. This method uses a demand matrix combined with numerical deviation calculations to replace subjective human judgment, avoiding one-sided selections that only consider cost while ignoring capacity or timeliness while ignoring damage rate. It simultaneously covers core characteristics such as timeliness, cost, capacity, and damage rate, ensuring that a single match meets the multiple demands of the supply chain transportation. This eliminates the need for multiple screenings of carriers across different dimensions, making the matching results more accurate and objective, thereby improving matching efficiency.

[0025] The cargo tracking module works as follows: Transportation anomalies include cargo location anomalies and cargo status anomalies. GPS positioning is installed on the transportation equipment to obtain the cargo location in real time and to determine whether the transportation time from the previous transportation node to the next transportation node exceeds the transportation time threshold. When the transportation time threshold is exceeded, it is determined that there is an anomaly in the cargo location, and the transportation time needs to be corrected in combination with environmental impact factors. Install corresponding sensors in the cargo transportation warehouse to obtain real-time data values ​​detected by the sensors. Compare the real-time data values ​​with the preset standard data threshold range. If the real-time data value is not within the standard data threshold range, it is determined that there is an abnormality in the cargo status. The method for obtaining environmental impact factors is as follows: Environmental impact information from the previous transportation node to the next transportation node is obtained, features are extracted to obtain parameter values ​​for multiple impact items, and then calculated using formulas. The environmental impact factors were determined. in, Let j be the parameter value of the j-th influencing term. Let j be the standard parameter values ​​of the influencing terms. Let j be the deviation comparison value of the j-th influencing term. To the total number of items affected, .

[0026] The above technical solution provides a specific method for determining whether there are transportation anomalies in goods, including two types of judgments: abnormal cargo location and abnormal cargo status. Specifically, when judging abnormal cargo location, GPS positioning is installed on the transportation equipment to obtain the cargo location in real time, and it is determined whether the transportation time from the previous transportation node to the next transportation node exceeds a transportation time threshold. The transportation time threshold can be determined based on historical data of the corresponding transportation node. When the transportation time threshold is exceeded, it is judged that there is an abnormality in the cargo location. However, transportation time is often affected by environmental factors, such as weather, road conditions, and construction, which can all affect the normal transportation time. Therefore, it is necessary to make corrections in conjunction with environmental impact factors. The method for obtaining environmental impact factors is as follows: obtain environmental impact information within the time from the previous transportation node to the next transportation node, perform feature extraction, obtain parameter values ​​for multiple impact items, and then use formulas... The environmental impact factors were derived, among which, Let j be the parameter value of the j-th influencing term. The standard parameter values ​​for j influencing items are set based on the historical data of each influencing item. The deviation comparison value for the j-th influencing term is manually set and used for dimensionless measurement and weighting operations. To the total number of items affected, It can be seen that the larger the environmental impact factor, the greater the difference between the parameter values ​​of each environmental impact item and the corresponding standard parameter values ​​during actual transportation, so more corrections are needed for transportation time. When the corrected transportation time exceeds the transportation time threshold, it is judged that the cargo location is abnormal, and an early warning response is issued. For example, after judging that the cargo location is abnormal, the system automatically sends an SMS and a background pop-up to the logistics dispatcher, and displays the current GPS location of the cargo. At the same time, it pushes the real-time road conditions of the section to the dispatch backend to assist in adjusting the subsequent transportation route. This can generate transportation time more accurately based on the actual situation, making the judgment more accurate. When judging the abnormal condition of goods, corresponding sensors are installed in the cargo transportation warehouse. For example, when transporting frozen goods, temperature sensors are installed to monitor the temperature and obtain the real-time data values ​​detected by the sensors. The real-time data values ​​are compared with preset standard data threshold ranges. Each preset standard data threshold range is determined based on the historical data of the corresponding goods and professional knowledge in the field. When the real-time data value is not within the standard data threshold range, for example, when transporting frozen goods, if the temperature is found to be outside the set temperature threshold range, it is judged that the condition of the goods is abnormal and an appropriate early warning response is issued. For example, when an abnormal condition of the goods is judged, the system automatically sends a voice broadcast to the driver and sends an early warning notification to the quality inspector of the nearest transit warehouse to prioritize the arrival inspection; thereby reducing the loss of goods.

[0027] The supply chain intelligent management and control method based on process nodes is implemented through the aforementioned supply chain intelligent management and control system based on process nodes.

[0028] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0029] All formulas in this specification are calculated by removing dimensions and obtaining numerical values. The method of removing dimensions is determined based on existing technology, and the formulas are derived from software simulations using a large amount of collected data to obtain the most recent real-world formulas. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0030] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A process node based intelligent management and control system for supply chain, characterized in that, The system includes: The login management module is used to obtain relevant information of the logged-in user, including the user's login information and access information. Based on the real-time login information, a real-time feature tag set is constructed, and based on the historical login information, a standard feature tag set is constructed. The Jaccard similarity between the two is calculated, and login security protection is performed on the user based on the Jaccard similarity. The access information includes access browsing traffic Q and access super-privilege times r in a time sequence The access risk coefficient is calculated by formula The user is protected based on the access risk coefficient The user is protected based on the access risk coefficient; wherein, is a time sequence The access browsing traffic change function is determined, The historical access browsing traffic change function is set according to historical access browsing traffic, is a time sequence The integral of is a time sequence The integral of​​​ The logistics route planning module automatically generates multiple candidate transportation routes from historical transportation routes based on the origin and destination of the order. It then acquires positive and negative correlation information for each candidate route, quantifies this information to obtain multiple positive and negative correlation values, assigns weights to the positive correlation values ​​and sums them to obtain a total positive correlation value, assigns weights to the negative correlation values ​​and sums them to obtain a total negative correlation value, and divides the total positive correlation value by the total negative correlation value to obtain the optimal value for each candidate transportation route. Finally, it autonomously determines the material transportation route based on the optimal value. The carrier matching module generates an order demand matrix based on the order demand information, obtains the demand information of each carrier from the carrier database, generates a carrier demand matrix based on the carrier demand information, calculates the difference between the order demand matrix and the carrier demand matrix to obtain a difference matrix, sums the elements in the difference matrix to obtain the deviation coefficient, and selects the carrier with the smallest deviation coefficient as the optimal carrier for transportation. The cargo tracking module is used to track the location and status of cargo in real time and determine whether the location and status of cargo are abnormal. The early warning module responds accordingly when it detects an anomaly. 2.The process node based supply chain intelligent management and control system according to claim 1, wherein, The methods for protecting user login and access security are as follows: The system extracts feature tags from the acquired login information to obtain multiple real-time feature tags, and constructs a real-time feature tag set based on these tags. It also acquires the user's historical login information, extracts corresponding feature tags from this information to obtain multiple historical feature tags, and constructs a standard feature tag set using these historical tags as standard tags. The Jaccard similarity between the real-time feature tag set and the standard feature tag set is recorded as the login security coefficient. The login security coefficient is compared with a preset login security coefficient threshold. If the login security coefficient is less than the threshold, the user is prevented from logging in; otherwise, the user is allowed to log in normally. When the user is allowed to log in normally, the access risk coefficient is accessed compared with a preset access risk coefficient threshold, when the access risk coefficient is greater than the preset access risk coefficient threshold, the user is prevented from further accessing, otherwise the user is allowed to continue accessing. 3.The process node based supply chain intelligent management and control system according to claim 1, wherein, The method for autonomously determining the transportation path of materials based on the preferred value is as follows: Based on the preference value, the candidate transportation routes are sorted in descending order, and the top three candidate transportation routes are recommended to the user. The user can then choose the transportation route for the materials from the three recommended candidate transportation routes. 4.The process node based supply chain intelligent management and control system according to claim 1, wherein, The method for selecting the carrier with the smallest deviation coefficient as the optimal carrier for transportation is as follows: The demand information of the order is subjected to feature extraction to obtain feature values of multiple demand features of the order, and an order demand matrix is established based on the feature values of the demand features of the order ; The demand information of each carrier is subjected to feature extraction to obtain feature values of multiple demand features of the carrier, and a carrier demand matrix is established based on the feature values of the multiple demand features of the carrier Each demand feature of the carrier in the matrix B corresponds to each demand feature in the matrix A. The difference matrix is represented as: ; the bias coefficient is represented as: ; The carrier with the smallest deviation coefficient is selected as the optimal carrier for transportation; wherein, is a feature value for the i-th demand feature of the order, is, n is the total number of demand features, is a feature value for the i-th demand feature of the carrier. 5.The process node based supply chain intelligent management and control system according to claim 1, wherein, The cargo tracking module works as follows: GPS positioning is installed on transportation equipment to obtain the real-time location of goods and determine whether the transportation time from the previous transportation node to the next transportation node exceeds the transportation time threshold. If the transportation time threshold is exceeded, it is determined that the location of the goods is abnormal and the transportation time needs to be corrected in combination with environmental impact factors. Sensors are installed inside the cargo transport warehouse to acquire real-time data values ​​detected by the sensors. These real-time data values ​​are then compared with a preset standard data threshold range. If the real-time data value is outside the standard data threshold range, the cargo status is determined to be abnormal. 6.The process node based supply chain intelligent management and control system according to claim 5, wherein, The method for obtaining environmental impact factors is as follows: Obtain the environmental impact information in the last transport node to the next transport node, carry out feature extraction, obtain the parameter values of multiple impact items, and obtain the environmental impact factor through the formula ​ wherein is a parameter value for the jth influence term, is a standard parameter value for the j influence terms, is a bias ratio value for the jth influence term, is the total number of influence terms, .

7. A process node based intelligent management and control method for supply chain, characterized in that, The method is implemented by the supply chain intelligent management and control system based on process nodes as described in any one of claims 1-6.

Images