Enterprise-level dynamic decision optimization system and method based on cognitive intelligence

Abstract

The invention discloses an enterprise-level dynamic decision optimization system and method based on cognitive intelligence, and relates to the technical field of data protection. Historical enterprise supply related information is collected and integrated with a corresponding adjustment scheme to obtain a data set, and the data set is stored in a database of an enterprise management platform; historical enterprise supply related information and corresponding adjustment schemes are analyzed based on cognitive intelligence, and a production model is established; real-time enterprise supply related information is collected as input data, a plurality of real-time adjustment schemes are output through a production model, the plurality of real-time adjustment schemes are selected through a management end, and the selected real-time adjustment scheme is used as a target real-time adjustment scheme. According to the method, the safety of the data in the main database can be ensured through the sub-databases, a better data protection effect is achieved, the data in the main database can be ensured to better train a subsequent convolutional neural network model, and the accuracy of the trained production model is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of data protection, and particularly to an enterprise-level dynamic decision-making optimization system and method based on cognitive intelligence. Background Art

[0002] From the perspective of market competition, the competition among enterprises is becoming increasingly fierce, and data has become the core asset. Enterprises rely on data to gain insights into market trends, optimize production processes, and formulate precise marketing strategies. The database of the cognitive intelligence system needs to store a large amount of key business data, customer information, and the business wisdom accumulated by the enterprise over a long time. These data are the basis for enterprises to build a dynamic decision-making optimization system based on cognitive intelligence. At present, with the extensive application of information technology, network attack means are becoming increasingly complex and diverse. Hacker organizations and malware developers are constantly looking for vulnerabilities in enterprise database systems and launching unauthorized network access. The enterprise's database system may be attacked at any time, and there is a risk of data being stolen, which is impossible to guard against. Summary of the Invention

[0003] The purpose of the present invention is to provide an enterprise-level dynamic decision-making optimization system and method based on cognitive intelligence to solve the deficiencies in the background art.

[0004] To achieve the above purpose, the present invention provides the following technical solutions: An enterprise-level dynamic decision-making optimization method based on cognitive intelligence, including the following steps:

[0005] Collect historical enterprise supply-related information and the corresponding adjustment plans for integration to obtain a data set, and store the data set in the database of the enterprise management platform. Among them, the database includes a main database, multiple sub-databases, a reserve database, and a bearing chain;

[0006] Analyze the historical enterprise supply-related information and the corresponding adjustment plans based on cognitive intelligence to establish a production model;

[0007] Collect real-time enterprise supply-related information as input data and output multiple real-time adjustment plans through the production model;

[0008] Select multiple real-time adjustment plans through the management terminal, use the selected real-time adjustment plan as the target real-time adjustment plan, and store the real-time enterprise supply-related information and the target real-time adjustment plan in the database of the enterprise management platform to continuously train the production model to obtain a new production model.

[0009] In a preferred embodiment, the step of collecting historical enterprise supply-related information and the corresponding adjustment plans for integration to obtain a data set and storing the data set in the database of the enterprise management platform includes:

[0010] Clean, annotate and organize the collected historical enterprise supply-related information and corresponding adjustment plans to obtain an integrated data set;

[0011] The database in the corresponding enterprise management platform is divided into multiple sub-databases and a main database, wherein the main database is located in the middle of the multiple sub-databases, and the connection chains are respectively set in the multiple sub-databases;

[0012] A connection port is set for each receiving element in each sub-database, and a corresponding preparation database is configured for the external corresponding sub-database of the database, and the sub-database is connected to the corresponding preparation database;

[0013] The data set is stored in the main database, and the sub-database is used to protect the main database from unauthorized network access.

[0014] In a preferred embodiment, the step of respectively setting up the succession chain in the plurality of sub-databases includes:

[0015] A main database is divided in the middle space of the database, and multiple continuous sub-databases are divided around the edge space of the main database in the database;

[0016] Multiple undertaking elements are set in multiple sub-databases, and the multiple undertaking elements are arranged in sequence as an undertaking chain.

[0017] In a preferred embodiment, the step of storing the data set in the main database and protecting the main database through the sub-database when there is unauthorized network access includes:

[0018] An authorized connection port is configured corresponding to the main database, and the data set is stored in the main database based on the authorized connection port;

[0019] The network accessing the sub-database is regarded as an unauthorized network, the first receiving element is activated in sequence as the target receiving element, and the target receiving element is connected to the unauthorized network through the connection port;

[0020] The target receiving element connected to the unauthorized network is moved to the corresponding preliminary database through the sub-database transmission.

[0021] In a preferred embodiment, the step of analyzing historical enterprise supply-related information and corresponding adjustment plans based on cognitive intelligence to establish a production model includes:

[0022] The historical enterprise supply-related information and the corresponding adjustment plans are used as training data, and the training data is divided into a training set, a validation set, and a test set;

[0023] The convolutional neural network model is trained using a training set, and during the training process, the convolutional neural network model is evaluated using a validation set, and the trained convolutional neural network model is tested using a test set to obtain a trained convolutional neural network model as a production model.

[0024] In a preferred embodiment, the step of collecting real-time enterprise supply-related information as input data and outputting multiple real-time adjustment plans through the production model includes:

[0025] Formulate a collection period, and use the enterprise supply-related information within the collection period as real-time enterprise supply-related information;

[0026] Input the real-time enterprise supply-related information into the production model to output multiple real-time adjustment plans.

[0027] In a preferred embodiment, the step of storing the real-time enterprise supply-related information and the target real-time adjustment plan in the database of the enterprise management platform and continuously training the production model to obtain a new production model includes:

[0028] Determine the target real-time adjustment plan from multiple real-time adjustment plans through the management terminal;

[0029] Store the real-time enterprise supply-related information and the target real-time adjustment plan in the main database of the enterprise management platform, and integrate the data in the main database to jointly train the production model to obtain a new production model.

[0030] The present invention also provides an enterprise-level dynamic decision-making optimization system based on cognitive intelligence, including:

[0031] A preparation module for collecting historical enterprise supply-related information and corresponding adjustment plans for integration to obtain a data set, and storing the data set in the database of the enterprise management platform, where the database includes a main database, multiple sub-databases, a preparatory database, and a bearing chain;

[0032] A training module connected to the preparation module for analyzing historical enterprise supply-related information and corresponding adjustment plans based on cognitive intelligence to establish a production model;

[0033] An analysis module connected to the training module for collecting real-time enterprise supply-related information as input data and outputting multiple real-time adjustment plans through the production model;

[0034] An adjustment module connected to the analysis module for selecting multiple real-time adjustment plans through the management terminal, using the selected real-time adjustment plan as the target real-time adjustment plan, and storing the real-time enterprise supply-related information and the target real-time adjustment plan in the database of the enterprise management platform to continuously train the production model to obtain a new production model.

[0035] In the above technical solution, the technical effects and advantages provided by the present invention are as follows:

[0036] The main database of the present invention is located in the middle of multiple sub-databases. Connecting chains are respectively set in the multiple sub-databases, and the division positions of the sub-databases in the database are distributed outside the main database, which can realize the protection of the main database. Secondly, multiple connecting elements arranged in sequence are set in the sub-databases, which can play the role of data storage and connection. The purpose of having multiple connecting elements is that when there is unauthorized network access, one connecting element can be enabled. Subsequently, when there are more additional unauthorized networks and a single virtual machine cannot be docked, and to ensure the stability of the sub-database, after the connecting element is connected to the unauthorized network, it will be transferred and moved through the connection relationship between the sub-database and the standby database, and the connecting element can be moved from the inside of the sub-database to the inside of the corresponding standby database, so that the sub-database is stable. After that, when there is access to the unauthorized network after the sub-database is stable, the corresponding connecting elements can be enabled in sequence. Through the sub-database, the security of the data in the main database can be ensured, which has a good data protection effect and can ensure that the data in the main database can better train the subsequent convolutional neural network model and ensure the accuracy of the trained production model. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required to be used in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0038] Figure 1 It is a flowchart of the method of the present invention.

[0039] Figure 2 It is a system block diagram of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0041] Embodiment 1. Please refer to Figure 1 As shown, the enterprise-level dynamic decision-making optimization method based on cognitive intelligence in this embodiment includes the following steps:

[0042] S1. Collect historical enterprise supply-related information and integrate it with the corresponding adjustment plan to obtain a data set, and store the data set in the database of the enterprise management platform, wherein the database includes a main database, multiple sub-databases, a preliminary database, and a bearer chain;

[0043] S2. Analyze historical enterprise supply-related information and corresponding adjustment plans based on cognitive intelligence to establish a production model;

[0044] S3, collect real-time enterprise supply-related information as input data and output multiple real-time adjustment plans through the production model;

[0045] S4. The management terminal selects multiple real-time adjustment plans, uses the selected real-time adjustment plan as the target real-time adjustment plan, stores the real-time enterprise supply-related information and the target real-time adjustment plan in the database of the enterprise management platform, and continuously trains the production model to obtain a new production model;

[0046] As described in the above steps S1-S4, the main database is located in the middle of multiple sub-databases, and a continuation chain is set in each of the multiple sub-databases. The division position of the sub-database in the database is distributed outside the main database, which can realize the protection of the main database. Secondly, multiple continuation elements arranged in sequence are set in the sub-database, which can play the role of data storage and connection. The purpose of having multiple continuation elements is to enable a continuation element when there is unauthorized network access. Subsequently, when there are more unauthorized networks, a single virtual machine cannot be connected. In order to ensure the stability of the sub-database, after the continuation element is connected to the unauthorized network, it will be transmitted and moved through the connection relationship between the sub-database and the preparatory database, and the continuation element can be moved from the inside of the sub-database to the inside of the corresponding preparatory database. In this way, the sub-database is stable. After that, if there is unauthorized network access after the sub-database is stable, the corresponding continuation elements can be enabled in sequence. The security of the data in the main database can be guaranteed through the sub-database, which has a better data protection effect, and can ensure that the data in the main database can be used to better train the subsequent convolutional neural network model and ensure the accuracy of the trained production model. Cognitive intelligence is a development stage of artificial intelligence, which aims to enable machines to understand, interpret and reason about human language, images, sounds and other information, and have cognitive abilities similar to humans. It can process and analyze data, and emphasizes the understanding, reasoning and application of knowledge, so that machines can think and make decisions like humans.

[0047] In one embodiment, the step S1 of collecting historical enterprise supply-related information and integrating it with the corresponding adjustment plan to obtain a data set and storing the data set in a database of the enterprise management platform includes:

[0048] S11. Clean, annotate and organize the collected historical enterprise supply-related information and corresponding adjustment plans to obtain an integrated data set;

[0049] S12, dividing the database in the corresponding enterprise management platform into multiple sub-databases and a main database, wherein the main database is located in the middle of the multiple sub-databases, and the multiple sub-databases are respectively provided with a connection chain;

[0050] S13, a connection port is set for each receiving element in each sub-database, a corresponding preliminary database is configured for the external corresponding sub-database of the database, and the sub-database is connected to the corresponding preliminary database;

[0051] S14, storing the data set in the main database, and protecting the main database through the sub-database when there is unauthorized network access;

[0052] In one embodiment, the step S12 of respectively setting up the succession chains in the plurality of sub-databases includes:

[0053] S121, dividing a main database in the middle space of the database, and dividing a plurality of continuous sub-databases around the edge space of the main database in the database;

[0054] S122. Multiple undertaking elements are set in multiple sub-databases, and the multiple undertaking elements are arranged in sequence as an undertaking chain.

[0055] In one embodiment, the step S14 of storing the data set in the main database and protecting the main database through the sub-database when there is unauthorized network access includes:

[0056] S141, configuring an authorized connection port corresponding to the main database, and storing the data set in the main database based on the authorized connection port;

[0057] S142, the network accessing the sub-database is regarded as an unauthorized network, the first receiving element is activated in order as the target receiving element, and the target receiving element is connected to the unauthorized network through the connection port;

[0058] S143, moving the target receiving unit connected to the unauthorized network to the corresponding preliminary database through the sub-database transmission;

[0059] As described in the above steps S11 - S14, the historical enterprise supply - related information and the corresponding adjustment plans are as follows: Historical enterprise supply - related information: order quantity, customer demand change trend, etc.; raw material supply data, such as the inventory level of suppliers, transportation status, possible supply interruption information, etc.; and equipment operation data, such as the real - time status of equipment, fault alarm information, etc.; The adjustment plan is: indicators such as production efficiency, cost, delivery time, etc. Clean, label, and organize the historical enterprise supply - related information and the corresponding adjustment plans to obtain an integrated data set. Here, cleaning is the processing of missing values, duplicate values, outliers, and noise in the data. The labeling operation is: according to business requirements and analysis objectives, clarify the categories and standards of labeling. For example, for the supply adjustment plan, labeling categories such as "increase supply", "decrease supply", "maintain supply" can be defined; The organization operation is: data format conversion (uniformly converting data in different formats into a format suitable for analysis and storage) and data integration (integrating supply - related information and adjustment plans from different data sources, such as databases, file systems, etc.) to form a unified data set. After obtaining the data set, it is necessary to store the data set, so as to obtain a large - scale data training library for subsequent training of the convolutional neural network model. Therefore, the security and accuracy of the data are very important, which can ensure the accuracy of model training and thus ensure the accuracy of the enterprise's dynamic decision - making output. To protect the database, first, divide the database in the enterprise management platform to obtain multiple sub - databases and a main database. Among them, the main database is located in the middle of multiple sub - databases. In each of the multiple sub - databases, a connection chain is set. The division positions of the sub - databases in the database are distributed outside the main database, which can protect the main database. Secondly, multiple connection elements arranged in sequence are set in the sub - databases. Among them, the connection element is a virtual machine, which can play the role of data storage and connection. The purpose of having multiple connection elements is that when there is unauthorized network access, one connection element can be enabled. Subsequently, when there are more additional unauthorized network accesses and a single virtual machine cannot connect, and to ensure the stability of the sub - database, after the connection element is connected to the unauthorized network, it will be transferred through the connection relationship between the sub - database and the standby database, and the connection element can be moved from the inside of the sub - database to the inside of the corresponding standby database, so that the sub - database is stable. After the sub - database is stable and there is unauthorized network access again, the corresponding connection elements can be enabled in sequence. In addition, when the connection element completes the connection task with the unauthorized network in the standby database, the criterion for task completion is that the unauthorized network automatically withdraws and no longer connects to the connection element in the standby database. Then, the unconnected connection element can be transferred back to the corresponding sub - database for continued use. Through the sub - database, the security of the data in the main database can be ensured. The security of the database can be understood through the security index. Among them, the formula for the security index is: Among them, δ is the security index, n is the number of sub-databases, F i is the number of times the ith sub-database is accessed by unauthorized network, ∈ is a constant greater than zero, μ i is the weight coefficient of the i-th sub-database, α i The number of recipients used for the i-th sub-database has a better data protection effect, which can ensure that the data in the main database can be used to better train the subsequent convolutional neural network model and ensure the accuracy of the trained production model.

[0060] In one embodiment, the step S2 of analyzing historical enterprise supply-related information and corresponding adjustment plans based on cognitive intelligence to establish a production model includes:

[0061] S21, using historical enterprise supply-related information and corresponding adjustment plans as training data, and dividing the training data into a training set, a validation set, and a test set;

[0062] S22, training the convolutional neural network model through the training set, evaluating the convolutional neural network model through the validation set during the training process, testing the trained convolutional neural network model through the test set, and obtaining the trained convolutional neural network model as the production model;

[0063] As described in the above steps S21 - S22, the historical enterprise supply - related information and the corresponding adjustment plans are used as training data. The training data is divided into a training set and a test set. The historical enterprise supply - related information and the corresponding adjustment plans use the data in the main database, which has been cleaned, labeled, and sorted. First, determine the model structure. The specific operation is to select a suitable convolutional neural network structure according to the data characteristics and problem requirements. Here, LeNet is selected. The CNN model consists of a convolutional layer, a pooling layer, a fully - connected layer, etc. Then, define the model parameters: including the size, number, and stride of the convolutional kernel, the type and parameters of the pooling layer, the number of neurons in the fully - connected layer, etc. Select Adam, the cross - entropy loss function, and evaluation metrics (such as accuracy, recall, F1 - value, etc.) to compile the model. Input the training set data into the constructed convolutional neural network model. The data undergoes forward propagation in the model, through operations such as the convolutional layer and the pooling layer, to extract the features of the data, and finally, classification or regression prediction is performed through the fully - connected layer to obtain the output result of the model. During the training process, regularly use the validation set to evaluate the model, and observe the performance metrics of the model on the validation set, such as accuracy, loss value, etc. Adjust the hyperparameters of the model, such as the learning rate, regularization parameter, etc., according to the results of the validation set to prevent the model from overfitting and improve the generalization ability of the model. After training is completed, use the test set to perform a final evaluation of the model, and calculate various performance metrics of the model on the test set, such as accuracy, recall, F1 - value, mean square error, etc. The specific performance metrics can be formulated according to actual requirements.

[0064] In one embodiment, step S3 of collecting real - time enterprise supply - related information as input data and outputting multiple real - time adjustment plans through a production model includes:

[0065] S31. Set a collection period, and use the enterprise supply - related information within the collection period as real - time enterprise supply - related information;

[0066] S32. Input the real - time enterprise supply - related information into the production model to output multiple real - time adjustment plans;

[0067] In one embodiment, step S4 of storing the real - time enterprise supply - related information and the target real - time adjustment plan in the database of the enterprise management platform and continuously training the production model to obtain a new production model includes:

[0068] S41. Determine the target real - time adjustment plan from multiple real - time adjustment plans through the management terminal;

[0069] S42. Store the real - time enterprise supply - related information and the target real - time adjustment plan in the main database of the enterprise management platform, and integrate the data in the main database to jointly train the production model to obtain a new production model;

[0070] As described in the above steps S31 and S32, and S41 and S42, a collection cycle is formulated. For example, relevant enterprise supply information is collected once every half month or one month. The actual collection cycle can be changed according to requirements. The relevant enterprise supply information within the collection cycle is used as real-time relevant enterprise supply information. Through the management terminal, a target real-time adjustment plan is determined among multiple real-time adjustment plans. It is not necessarily the case that all of the obtained multiple real-time adjustment plans are suitable. Here, an operation of selection is required according to the management terminal to select the most suitable plan. The real-time relevant enterprise supply information and the target real-time adjustment plan are stored in the main database of the enterprise management platform. The data in the main database is integrated to jointly train the production model to obtain a new production model. In this way, the production model can be continuously trained according to the newly added relevant enterprise supply information and adjustment plan, ensuring the accuracy of the production model.

[0071] Example 2, please refer to Figure 2 As shown, the enterprise-level dynamic decision-making optimization system based on cognitive intelligence described in this embodiment includes:

[0072] A preparation module, which is used to collect and integrate historical relevant enterprise supply information and corresponding adjustment plans to obtain a data set, and store the data set in the database of the enterprise management platform. Among them, the database includes a main database, multiple sub-databases, a preparatory database, and a bearing chain;

[0073] A training module, connected to the preparation module, which is used to analyze the historical relevant enterprise supply information and corresponding adjustment plans based on cognitive intelligence to establish a production model;

[0074] An analysis module, connected to the training module, which is used to collect real-time relevant enterprise supply information as input data and output multiple real-time adjustment plans through the production model;

[0075] An adjustment module, connected to the analysis module, which is used to select among multiple real-time adjustment plans through the management terminal, use the selected real-time adjustment plan as the target real-time adjustment plan, store the real-time relevant enterprise supply information and the target real-time adjustment plan in the database of the enterprise management platform, and continuously train the production model to obtain a new production model;

[0076] The sub-database can ensure the security of the data in the main database, has a good data protection effect, and can ensure that the data in the main database can better train the subsequent convolutional neural network model, ensuring the accuracy of the trained production model.

[0077] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims described above.

Claims

1. An enterprise-level dynamic decision-making optimization method based on cognitive intelligence, characterized in that The following steps are involved: Collect historical enterprise supply-related information and integrate it with the corresponding adjustment plan to obtain a data set, and store the data set in the database of the enterprise management platform, where the database includes a main database, multiple sub-databases, a preliminary database, and a bearer chain; Analyze historical enterprise supply-related information and corresponding adjustment plans based on cognitive intelligence to establish a production model; Collect real-time enterprise supply-related information as input data and output multiple real-time adjustment plans through the production model; By selecting multiple real-time adjustment plans through the management end, the selected real-time adjustment plan is used as the target real-time adjustment plan, and the real-time enterprise supply-related information and the target real-time adjustment plan are stored in the database of the enterprise management platform to continuously train the production model to obtain a new production model.

2. The enterprise-level dynamic decision-making optimization method based on cognitive intelligence according to claim 1, characterized in that: The step of collecting historical enterprise supply-related information and integrating it with the corresponding adjustment plan to obtain a data set, and storing the data set in a database of the enterprise management platform includes: Clean, annotate and organize the collected historical enterprise supply-related information and corresponding adjustment plans to obtain an integrated data set; The database in the corresponding enterprise management platform is divided into multiple sub-databases and a main database, wherein the main database is located in the middle of the multiple sub-databases, and the connection chains are respectively set in the multiple sub-databases; A connection port is set for each receiving element in each sub-database, and a corresponding preparation database is configured for the external corresponding sub-database of the database, and the sub-database is connected to the corresponding preparation database; The data set is stored in the main database, and the sub-database is used to protect the main database from unauthorized network access.

3. The enterprise-level dynamic decision-making optimization method based on cognitive intelligence according to claim 2, wherein: The step of respectively setting up the continuation chains in the multiple sub-databases includes: A main database is divided in the middle space of the database, and multiple continuous sub-databases are divided around the edge space of the main database in the database; Multiple undertaking elements are set in multiple sub-databases, and the multiple undertaking elements are arranged in sequence as an undertaking chain.

4. The enterprise-level dynamic decision-making optimization method based on cognitive intelligence according to claim 3, characterized in that: The step of storing the data set in the main database and protecting the main database through the sub-database when there is unauthorized network access includes: An authorized connection port is configured corresponding to the main database, and the data set is stored in the main database based on the authorized connection port; The network accessing the sub-database is regarded as an unauthorized network, the first receiving element is activated in sequence as the target receiving element, and the target receiving element is connected to the unauthorized network through the connection port; The target receiving element connected to the unauthorized network is moved to the corresponding preliminary database through the sub-database transmission.

5. The enterprise-level dynamic decision-making optimization method based on cognitive intelligence according to claim 1, characterized in that: The step of analyzing historical enterprise supply-related information and corresponding adjustment plans based on cognitive intelligence to establish a production model includes: The historical enterprise supply-related information and the corresponding adjustment plans are used as training data, and the training data is divided into a training set, a validation set, and a test set; The convolutional neural network model is trained through the training set, and the convolutional neural network model is evaluated through the validation set during the training process. The trained convolutional neural network model is tested with the test set to obtain the trained convolutional neural network model as the production model.

6. The enterprise-level dynamic decision-making optimization method based on cognitive intelligence according to claim 1, wherein: The steps of collecting real-time enterprise supply-related information as input data and outputting multiple real-time adjustment plans through a production model include: Formulating a collection cycle and using the enterprise supply-related information within the collection cycle as real-time enterprise supply-related information; Inputting the real-time enterprise supply-related information into the production model to output multiple real-time adjustment plans.

7. The enterprise-level dynamic decision-making optimization method based on cognitive intelligence according to claim 1, characterized in that: The steps of storing the real-time enterprise supply-related information and the target real-time adjustment plan in the database of the enterprise management platform and continuously training the production model to obtain a new production model include: Determining the target real-time adjustment plan from multiple real-time adjustment plans through the management terminal; Storing the real-time enterprise supply-related information and the target real-time adjustment plan in the main database of the enterprise management platform, integrating the data in the main database, and jointly training the production model to obtain a new production model.

8. An enterprise-level dynamic decision-making optimization system based on cognitive intelligence, which is used to implement the enterprise-level dynamic decision-making optimization method based on cognitive intelligence according to any one of claims 1-7, and is characterized in that, Including: A preparation module for collecting historical enterprise supply-related information and corresponding adjustment plans, integrating them to obtain a data set, and storing the data set in the database of the enterprise management platform. The database includes a main database, multiple sub-databases, a preparatory database, and a bearing chain; A training module connected to the preparation module for analyzing the historical enterprise supply-related information and corresponding adjustment plans based on cognitive intelligence and establishing a production model; An analysis module connected to the training module for collecting real-time enterprise supply-related information as input data and outputting multiple real-time adjustment plans through the production model; An adjustment module connected to the analysis module for selecting multiple real-time adjustment plans through the management terminal, using the selected real-time adjustment plan as the target real-time adjustment plan, storing the real-time enterprise supply-related information and the target real-time adjustment plan in the database of the enterprise management platform, and continuously training the production model to obtain a new production model.

Images