A model-based resource matching system for enterprise digital transformation

Abstract

The application relates to the technical field of enterprise digital management, and provides a resource matching system for enterprise digital transformation based on a model.The system realizes efficient resource matching through the cooperative work of a server end, a background control end, a demander end and a supplier end.The server end collects multidimensional resource data through an acquisition unit, a feature matching unit filters out key features for preliminary matching, a model construction unit constructs a small model by using a learning method and a model knowledge base, a model optimization unit continuously optimizes the model by using user feedback and real-time data, a resource report generation unit trains a use scenario by establishing a supply chain simulation model, obtains a continuously updated matching small model, and finally generates a visual resource matching report.The application solves the limitation of the small model in processing complex supply chain relations, improves the matching accuracy and efficiency, and provides strong support for the digital transformation of small and medium-sized enterprises.

Description

1.1 Technical Field

[0002] This invention relates to the field of enterprise digital management technology, and in particular to a model-based resource matching system for enterprise digital transformation. 1.2 Background Technology

[0004] With the rapid development of information technology, matching enterprise supplier resources has become a key link in improving supply chain efficiency and optimizing resource allocation. Currently, the integrated application of technologies such as big data, cloud computing, and artificial intelligence has brought new opportunities to enterprise supplier management. In particular, small-model technology has received widespread attention due to its lightweight, high efficiency, and flexibility. This technology can accurately identify the optimal matching point between suppliers and enterprise needs through deep learning of large amounts of data, providing strong support for enterprise procurement decisions.

[0005] In enterprise supplier resource matching, small-model technology primarily constructs an accurate predictive model by analyzing multi-dimensional information such as historical procurement data, supplier performance, and market price dynamics. This predictive model can quickly identify high-quality suppliers that meet the enterprise's needs, while also predicting future market trends and supplier performance. Through automated algorithms, enterprises can adjust their procurement strategies in real time, optimize inventory management, reduce procurement costs, and improve the responsiveness and flexibility of the supply chain. Furthermore, small-model technology can help enterprises identify potential risk points and enhance the robustness of the supply chain.

[0006] While small-model technology has demonstrated significant advantages in matching enterprise supplier resources, it also presents several technical challenges. First, the accuracy and completeness of data are crucial to model performance. However, in practice, data is often incomplete or erroneous, affecting the model's predictive accuracy. Furthermore, as the market environment changes, small-model technology exhibits limitations in handling complex and non-linear supply chain relationships, struggling to capture all influencing factors and leading to inaccuracies in the supplier resource matching system. Non-linear supply chain relationships refer to complex, non-direct proportional relationships between elements within the supply chain. These relationships are not simple causal or linear additive relationships but encompass multiple interactions and feedback mechanisms. 1.3 Summary of the Invention

[0008] The purpose of this invention is to provide a model-based resource matching system for enterprise digital transformation, in order to solve the problem that small model technology has limitations in dealing with complex and nonlinear supply chain relationships, making it difficult to capture all influencing factors and leading to inaccuracies in the supplier resource matching system.

[0009] This invention provides a model-based resource matching system for enterprise digital transformation, comprising a server, a backend control terminal, a demand-side terminal, and a supply-side terminal. The server establishes communication connections with the backend control terminal, the demand-side terminal, and the supply-side terminal. The server constructs a small-scale enterprise digital supplier resource matching model based on data transmitted from the backend control terminal and the supply-side terminal. The server performs resource matching on the demand data from the demand-side terminal using this small-scale model. The server outputs a visualized resource matching report. The server includes:

[0010] The acquisition unit acquires resource data, which includes enterprise information, demand data, market dynamics, industry trends, and historical transaction data.

[0011] The feature matching unit receives demand data, analyzes the demand data using a feature selection algorithm, filters out the features that have the greatest impact on the matching results, and performs matching in the resource matching database based on the features that have the greatest impact on the matching results to obtain the supplier resource matching results.

[0012] The model building unit establishes a learning method and model knowledge base. The data generated during the supplier resource matching process and the resource data acquired are sequentially learned through the learning methods and models in the learning method and model knowledge base to obtain a set of learning methods and model data. Based on the set of learning methods and model data, a small digital supplier resource matching model for the enterprise is built.

[0013] The model optimization unit collects user feedback information and uses it, along with real-time resource data, as new training data. It employs a stochastic gradient descent algorithm to continuously update and optimize the small model for matching enterprise digital supplier resources. The small model for matching enterprise digital supplier resources uses reinforcement learning algorithms to learn matching decisions through interaction with the environment.

[0014] The resource report generation unit establishes a supply chain simulation model to simulate usage scenarios in the supply chain. It trains the enterprise's digital supplier resource matching mini-model using these usage scenarios to obtain usage scenario training data. This training data is then used as new training data, and the enterprise's digital supplier resource matching mini-model is continuously trained using this new data to obtain a continuously updated model. This continuously updated model is then used to process user needs data and generate a visualized resource matching report.

[0015] Furthermore, the resource matching system based on a model for enterprise digital transformation described in this invention includes an acquisition unit comprising:

[0016] Establish relational and non-relational databases, and store the acquired resource data into the corresponding database according to the data type;

[0017] Before data storage, sensitive data from multiple sources is encrypted using encryption algorithms. After data transmission is complete, the server performs integrity verification on the received data.

[0018] The server records database operation logs, including data creation, deletion, modification, and query operations. It also performs regular security audits and monitoring of the database to check for any abnormal access or data tampering.

[0019] Furthermore, the feature matching unit of the model-based resource matching system for enterprise digital transformation described in this invention includes:

[0020] Based on the characteristics of the demand data, select suitable feature selection algorithms, input the extracted features into the selected feature selection algorithms, and analyze the importance of each feature to the matching results through the feature selection algorithms.

[0021] Based on the output of the feature selection algorithm, a set of features that affect the matching results is selected;

[0022] By utilizing the effect of each feature on the matching results and the features that influence the matching results, a matching model is constructed to search for supplier resources that meet the criteria in the resource matching database;

[0023] Furthermore, the feature matching unit in the model-based resource matching system for enterprise digital transformation described in this invention further includes:

[0024] The feature values ​​in the demand data are used as input parameters, and the matching model searches and compares them in the resource matching database.

[0025] Based on the output of the matching model, supplier resources that meet the requirements are selected.

[0026] The validity of the selected supplier resources that meet the requirements is tested. After the test, the valid supplier resources that meet the requirements are obtained, and the results are returned to the demand side.

[0027] Furthermore, the model-based resource matching system for enterprise digital transformation described in this invention includes a model building unit comprising:

[0028] The server collects learning methods and models suitable for supplier resource matching, including decision trees, random forests, support vector machines, and deep learning models, and constructs a knowledge base of learning methods and models.

[0029] The learning methods and model knowledge base is used to store and manage the knowledge base architecture of learning methods and models, and the built learning methods and model knowledge base is deployed on the server;

[0030] The data generated during the supplier resource matching process, along with previously acquired resource data, are integrated and preprocessed to form a dataset for model training. The dataset used for model training is then used to train the model sequentially through learning methods and models from the model knowledge base. During training, the performance of each method and model is recorded, including metrics such as accuracy and recall, to evaluate the model's effectiveness.

[0031] Furthermore, the model building unit of the model-based resource matching system for enterprise digital transformation described in this invention further includes:

[0032] Receive specific business requirements from enterprises, including the model algorithms to be used;

[0033] Deploy enterprise-specific business requirements, including the models and algorithms to be used, onto the server and test them, continuously monitoring the performance of enterprise-specific business requirements, including the models and algorithms to be used, during the testing process;

[0034] The performance of each model includes regular updates to the learning methods and models in the model knowledge base.

[0035] Furthermore, the model-based resource matching system for enterprise digital transformation described in this invention includes a model optimization unit comprising:

[0036] Set up a user feedback system on the server to obtain user evaluations and feedback on the supplier resource allocation results through online surveys, rating systems, or user feedback forms;

[0037] The system collects and records user feedback on matching results in real time, including satisfaction, accuracy, and timeliness. It then organizes and analyzes the collected user feedback to extract key information, such as the most satisfactory or unsatisfactory matching results and suggestions for improvement.

[0038] By integrating user feedback and real-time resource data to form a new training dataset, and using the stochastic gradient descent algorithm in conjunction with the new training dataset, the parameters of the enterprise digital supplier resource matching small model are updated and optimized.

[0039] Furthermore, the model optimization unit in the model-based resource matching system for enterprise digital transformation described in this invention also includes...

[0040] A reinforcement learning environment is defined for supplier resource matching. The reinforcement learning environment includes a state space, an action space, and a reward function. A small model for enterprise digital supplier resource matching is trained using a reinforcement learning algorithm. The reinforcement learning algorithm will enable the small model for enterprise digital supplier resource matching to adjust its matching strategy based on the reward obtained after receiving different matching decisions.

[0041] A small model for matching enterprise digital supplier resources interacts with a reinforcement learning environment.

[0042] Furthermore, the resource report generation unit of the model-based resource matching system for enterprise digital transformation described in this invention includes:

[0043] Based on the real-time supply chain operation process and rules, a supply chain simulation model is built on the server. The supply chain simulation model is used to simulate the use cases in the supply chain, including procurement, production, logistics and sales.

[0044] Different usage scenario parameters are set in the simulation model, including changes in market demand, fluctuations in supplier capacity, and logistics delays.

[0045] By running simulation models, a small model for matching enterprise digital supplier resources is trained under different usage scenarios. The matching results and performance data under each scenario are recorded to form a usage scenario training dataset.

[0046] Furthermore, the resource report generation unit of the model-based resource matching system for enterprise digital transformation described in this invention further includes:

[0047] Furthermore, the resource report generation unit of the model-based resource matching system for enterprise digital transformation provided by the present invention further includes:

[0048] Determine the objectives of the simulation model and analyze the actual needs of matching digital supplier resources for enterprises. The actual needs of matching digital supplier resources for enterprises include supplier information, product requirements, and procurement history.

[0049] Choose the Simulink simulation tool to build a simulation model for supplier resource matching based on the company's actual operation, including a supplier library, a product requirement library, and a matching algorithm module.

[0050] Different usage scenarios are set in the simulation model, including changes in the number of suppliers, changes in product demand, and changes in the market environment.

[0051] The supplier data model includes basic supplier information, such as name, address, contact information, product list, price, quality, and delivery date.

[0052] The beneficial effects of this invention are as follows:

[0053] Improve matching accuracy: By comprehensively applying small model technology and multiple learning algorithms, and combining continuous training with reinforcement learning algorithms, we can accurately identify and match supplier resources that best meet the needs of enterprises.

[0054] Optimize resource allocation: The system analyzes market dynamics, industry trends and historical transaction data in real time, automatically filters and optimizes supplier resources, achieves efficient resource allocation, reduces procurement costs and improves supply chain efficiency.

[0055] Enhanced system adaptability: Through continuous updates and learning mechanisms, the system can respond to changes in the market environment and complex supply chain relationships, and adjust matching strategies in a timely manner.

[0056] Improving user experience: The introduction of a user feedback system enables users to collect feedback information on satisfaction, accuracy, and timeliness in a timely manner and use it for model optimization, thereby improving user experience.

[0057] Generate visual resource matching reports: The system can generate clear resource matching reports, which facilitates decision analysis for enterprises.

[0058] Facilitating Digital Transformation of SMEs: This invention provides SMEs with a lightweight, efficient, and flexible digital transformation solution that helps enhance their competitiveness. 1.4 Description of the attached figures

[0060] To more clearly illustrate the technical solution of the present invention, the drawings used in the embodiments will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0061] Figure 1 This is a schematic diagram of the system architecture provided for an embodiment of the present invention.

[0062] Figure 2 This is a schematic diagram of a server-side device provided in an embodiment of the present invention. 1.5 Detailed Implementation

[0064] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention. The technical solutions provided by various embodiments of this invention will be described in detail below with reference to the accompanying drawings.

[0065] Please see Figures 1 to 2 This invention provides a model-based resource matching system for enterprise digital transformation, comprising:

[0066] This invention provides a model-based resource matching system for enterprise digital transformation, comprising a server, a backend control terminal, a demand-side terminal, and a supply-side terminal. The server establishes communication connections with the backend control terminal, the demand-side terminal, and the supply-side terminal. The server constructs a small-scale enterprise digital supplier resource matching model based on data transmitted from the backend control terminal and the supply-side terminal. The server performs resource matching on the demand data from the demand-side terminal using the small-scale enterprise digital supplier resource matching model. The server outputs a visualized resource matching report. The server includes:

[0067] The acquisition unit acquires resource data, which includes enterprise information, demand data, market dynamics, industry trends, and historical transaction data.

[0068] We have established partnerships with various data providers to integrate with API interfaces, enabling us to obtain demand data, historical transaction data, and other data in real time. By integrating with various data sources through API interfaces, we ensure the accuracy and timeliness of the data.

[0069] The collected raw data is cleaned to remove duplicates, invalid data, or errors, ensuring data quality and accuracy. Data from different data sources is integrated to form a unified and complete dataset, facilitating subsequent data analysis and mining. Data mining algorithms, such as association analysis and cluster analysis, are used to extract valuable information and patterns from massive amounts of data, supporting enterprise decision-making. Data is visualized through charts, dashboards, and other methods to intuitively display enterprise information, market dynamics, industry trends, etc., helping decision-makers better understand the data and make informed decisions.

[0070] The feature matching unit receives demand data, analyzes the demand data using a feature selection algorithm, filters out the features that have the greatest impact on the matching results, and performs matching in the resource matching database based on the features that have the greatest impact on the matching results to obtain the supplier resource matching results.

[0071] The feature matching unit first analyzes the received requirement data using feature selection algorithms, evaluating the importance of each feature to the matching results, thereby selecting those features that have the greatest impact on the matching outcome. Commonly used feature selection algorithms include:

[0072] Filtering methods: These methods select features based on statistical tests, such as using correlation coefficients, mutual information, or chi-square tests to assess the correlation between features and the target variable. By setting thresholds, features that are highly correlated with the target variable can be selected.

[0073] Wrap-up method: The wrap-up method selects features by evaluating model performance. It typically uses some kind of machine learning algorithm to search for the optimal subset of features in the feature space. This method can take into account the interactions between features, but the computational cost is relatively high.

[0074] Embedding: Embedding combines the feature selection process with the model training process. For example, when training models such as decision trees or random forests, features can be selected based on their importance scores. This method allows feature selection to be completed naturally during model training.

[0075] Database matching technology involves identifying key features, and then using these features to search and match within a resource matching database. This involves the following technologies:

[0076] Indexing techniques: To improve matching efficiency, databases typically create indexes for key features. This allows the database to quickly locate data records containing the target features during a matching query, thus accelerating the matching process.

[0077] SQL Query Optimization: The feature matching unit needs to construct efficient SQL queries to retrieve matching supplier resources from the database. Optimizing the query structure and using appropriate query conditions can improve query speed and accuracy.

[0078] Fuzzy matching technology: When demand data contains a certain degree of fuzziness or uncertainty, fuzzy matching technology can be used to find supplier resources that are similar or close to the demand data. For example, a text similarity-based algorithm can be used to calculate the similarity between demand data and supplier resources and return results with higher similarity.

[0079] The model building unit establishes a learning method and model knowledge base. The data generated during the supplier resource matching process and the resource data acquired are sequentially learned through the learning methods and models in the learning method and model knowledge base to obtain a set of learning methods and model data. Based on the set of learning methods and model data, a small digital supplier resource matching model for the enterprise is built.

[0080] The model building unit utilizes machine learning and deep learning techniques to learn potential patterns and rules from large amounts of data, thereby enabling prediction and decision-making for unknown data. In supplier resource matching scenarios, machine learning and deep learning models can learn complex relationships in multi-dimensional data such as historical transaction data, enterprise information, and market dynamics, and then predict the degree of matching between new suppliers and demanders.

[0081] A knowledge base of learning methods and models is established, containing various learning algorithms and pre-trained models applicable to supplier resource matching scenarios. These algorithms and models may include, but are not limited to, linear regression, support vector machines, neural networks, and decision trees. By applying these algorithms and models to the data generated and the resource data acquired during the supplier resource matching process, a more accurate and efficient matching model can be trained.

[0082] During model training, model building units utilize techniques such as cross-validation and grid search to fine-tune the model, finding the optimal model parameters and structure. This ensures the model achieves a good fit on the training set and demonstrates good generalization ability on the test set. Furthermore, to prevent overfitting or underfitting, regularization and ensemble learning methods are employed to optimize model performance.

[0083] After obtaining multiple learning methods and model datasets through learning methods and model knowledge bases, the model building unit employs model ensemble techniques, such as voting, stacking, or fusion, to combine the prediction results of these individual models, thereby improving the overall accuracy and stability of the predictions. Finally, the trained and optimized matching mini-models are deployed into the system for use in subsequent feature matching and resource report generation processes.

[0084] The model optimization unit collects user feedback information and uses it, along with real-time resource data, as new training data. It employs a stochastic gradient descent algorithm to continuously update and optimize the small model for matching enterprise digital supplier resources. The small model for matching enterprise digital supplier resources uses reinforcement learning algorithms to learn matching decisions through interaction with the environment.

[0085] To continuously optimize the matching model, the first step is to collect user feedback. This can be achieved by designing user feedback interfaces and questionnaires, or by using technologies that track user behavior (such as clickstream data tracking). The collected user feedback may include satisfaction ratings, evaluations of the accuracy of matching results, and user opinions and suggestions regarding the matching process. This information provides a reference for model optimization.

[0086] The collected user feedback information and real-time resource data need to be effectively integrated and processed in order to be used as new training data to update and optimize the model. Data fusion technology involves integrating data from different sources and in different formats to ensure data consistency and availability. Data processing technology includes steps such as data cleaning, transformation and standardization to eliminate problems such as outliers, missing values ​​and duplicate values, making the data more suitable for model training.

[0087] Stochastic Gradient Descent (SGD) is an optimization algorithm used to adjust model parameters during training to minimize a loss function. Compared to traditional batch gradient descent, SGD updates model parameters using only one sample or a mini-batch of samples in each iteration, making the training process more efficient and capable of handling large-scale datasets. In optimizing small models for supplier resource matching, SGD helps the model converge to the optimal solution faster, improving matching accuracy.

[0088] Reinforcement learning is a machine learning method that learns optimal decision-making strategies through interaction with the environment. In supplier resource matching scenarios, reinforcement learning algorithms enable a small matching model to continuously learn and improve its matching decisions through interaction with the environment. Specifically, the model tries different matching strategies and adjusts its decision-making based on reward signals from the environment. In this way, the model can gradually learn the optimal matching strategy to adapt to constantly changing market demands and supplier resource conditions.

[0089] The resource report generation unit establishes a supply chain simulation model to simulate usage scenarios in the supply chain. It trains the enterprise's digital supplier resource matching mini-model using these usage scenarios to obtain usage scenario training data. This training data is then used as new training data, and the enterprise's digital supplier resource matching mini-model is continuously trained using this new data to obtain a continuously updated model. This continuously updated model is then used to process user needs data and generate a visualized resource matching report.

[0090] To simulate a real supply chain environment, the resource report generation unit first establishes a supply chain simulation model. This model can simulate various usage scenarios in the supply chain, such as demand changes, supplier responses, and logistics distribution. This provides training data for the enterprise's digital supplier resource matching model, which is close to real-world usage scenarios. Through the simulation model, the matching model can be tested and optimized in a relatively safe and controllable environment.

[0091] Using usage scenario training data generated by the supply chain simulation model, the matching small model is trained. In this way, new and diverse training data can be continuously provided to the model, helping it to better learn and adapt to various complex supply chain environments. This training method not only improves the model's generalization ability but also makes the model more robust and accurate when facing new scenarios.

[0092] The enterprise digital supplier resource matching small model achieves dynamic updates by continuously training with new training data. This continuous update technology ensures that the model can keep up with market changes and user needs, and always maintain its advanced nature and accuracy. Through continuous learning and optimization, the model can better adapt to new data and environments, thereby providing more accurate matching results.

[0093] The resource report generation unit also utilizes advanced data processing technology to analyze user needs, ensuring data quality and accuracy. At the same time, it uses visualization technologies such as charts, images, and animations to present complex matching results in an intuitive and easy-to-understand way. This not only improves the readability and comprehension of the report but also helps users grasp the matching results and supply chain status more quickly.

[0094] Specifically, the present invention provides a model-based resource matching system for enterprise digital transformation, comprising an acquisition unit including:

[0095] Establish relational and non-relational databases, and store the acquired resource data into the corresponding database according to the data type;

[0096] Before data storage, sensitive data from multiple sources is encrypted using encryption algorithms. After data transmission is complete, the server performs integrity verification on the received data.

[0097] The server records database operation logs, including data creation, deletion, modification, and query operations. It also performs regular security audits and monitoring of the database to check for any abnormal access or data tampering.

[0098] Specifically, the present invention provides a model-based resource matching system for enterprise digital transformation, comprising a feature matching unit including:

[0099] Based on the characteristics of the demand data, select suitable feature selection algorithms, input the extracted features into the selected feature selection algorithms, and analyze the importance of each feature to the matching results through the feature selection algorithms.

[0100] Based on the output of the feature selection algorithm, a set of features that affect the matching results is selected. Using the importance of each feature to the matching results and the set of features that affect the matching results, one or more matching models are constructed to search for supplier resources that meet the conditions in the resource matching database.

[0101] The key feature values ​​in the demand data are used as input parameters, and the matching model searches and compares them in the resource matching database.

[0102] Based on the output of the matching model, select supplier resources that meet the requirements and return the results to the demand side.

[0103] Specifically, the present invention provides a model-based resource matching system for enterprise digital transformation, wherein the model building unit includes:

[0104] The server collects and organizes various learning methods and models suitable for supplier resource matching, including decision trees, random forests, support vector machines, and deep learning models.

[0105] The learning methods and model knowledge base is used to store and manage the knowledge base architecture of these learning methods and models. The pre-built learning methods and model knowledge base is deployed on the server. The learning methods and model knowledge base includes functions for adding, deleting, modifying and querying test data, as well as the calling and updating mechanism of models.

[0106] The data generated during the supplier resource matching process, along with previously acquired resource data, are integrated and preprocessed to form a dataset suitable for model training. Using the prepared dataset, the model is trained sequentially using learning methods and models from the model knowledge base. During training, the performance of each model is recorded, including metrics such as accuracy and recall, to evaluate the model's effectiveness.

[0107] Based on the model evaluation results, the poorly performing models are optimized and adjusted, and one or more of the best-performing models are selected from among many models as the basis for building smaller models in the future.

[0108] The best selected models are integrated and combined with the specific business needs and rules of the enterprise to output a small model for matching digital supplier resources that is suitable for the enterprise.

[0109] The constructed mini-models are deployed to the server and tested in a real-world environment. During the operation of the mini-models, their performance is continuously monitored, and necessary optimizations and adjustments are made based on the actual situation. The performance of each model includes regularly updating the learning methods and models in the model knowledge base, and adjusting the parameters and algorithms of the mini-models to maintain their matching accuracy and efficiency.

[0110] Specifically, the present invention provides a model-based resource matching system for enterprise digital transformation, comprising a model optimization unit, including:

[0111] Set up a user feedback system on the server to obtain user evaluations and feedback on the supplier resource allocation results through online surveys, rating systems, or direct user feedback forms;

[0112] The system collects and records user feedback on matching results in real time, including satisfaction, accuracy, and timeliness. It then organizes and analyzes the collected user feedback to extract key information, such as the most satisfactory or unsatisfactory matching results and suggestions for improvement.

[0113] By integrating user feedback and real-time resource data to form a new training dataset, and using the stochastic gradient descent algorithm in conjunction with the new training dataset, the parameters of the enterprise digital supplier resource matching small model are updated and optimized.

[0114] Specifically, the present invention provides a model-based resource matching system for enterprise digital transformation, including a model optimization unit, and further comprising...

[0115] A reinforcement learning environment is defined for supplier resource matching. The reinforcement learning environment includes a state space, an action space, and a reward function. A small model for enterprise digital supplier resource matching is trained using a reinforcement learning algorithm. The reinforcement learning algorithm will enable the small model for enterprise digital supplier resource matching to adjust its matching strategy based on the reward obtained after receiving different matching decisions.

[0116] The enterprise digital supplier resource matching small model continuously learns and improves its matching decision-making ability through continuous interaction with the reinforcement learning environment.

[0117] Specifically, the present invention provides a model-based resource matching system for enterprise digital transformation, including a resource report generation unit, comprising:

[0118] Based on the real-time supply chain operation process and rules, a supply chain simulation model is built on the server. The supply chain simulation model is used to simulate the use cases in the supply chain, including procurement, production, logistics and sales.

[0119] Different usage scenario parameters are set in the simulation model, including changes in market demand, fluctuations in supplier capacity, and logistics delays.

[0120] By running simulation models, a small model for matching enterprise digital supplier resources is trained under different usage scenarios. The matching results and performance data under each scenario are recorded to form a usage scenario training dataset.

[0121] Specifically, the resource matching system based on a model for enterprise digital transformation provided by this invention, including a resource report generation unit, further includes:

[0122] Determine the objectives of the simulation model and analyze the actual needs of matching digital supplier resources for enterprises. The actual needs of matching digital supplier resources for enterprises include supplier information, product requirements, and procurement history.

[0123] Choose the Simulink simulation tool to build a simulation model for supplier resource matching based on the company's actual operation, including a supplier library, a product requirement library, and a matching algorithm module.

[0124] Different usage scenarios are set in the simulation model, including changes in the number of suppliers, changes in product demand, and changes in the market environment.

[0125] The supplier data model includes basic supplier information, such as name, address, contact information, product list, price, quality, and delivery date.

[0126] To address the limitations of small-model technology in handling complex and nonlinear supply chain relationships, this invention employs the following measures to improve the accuracy of the supplier resource matching system:

[0127] Multidimensional data acquisition and analysis: By acquiring data from multiple dimensions such as enterprise information, demand data, market dynamics, industry trends, and historical transaction data, the model can be provided with comprehensive data support.

[0128] Feature selection algorithm: The feature matching unit uses a feature selection algorithm to analyze the required data and filter out the features that have the greatest impact on the matching results. This step ensures that the model can focus on the most critical information points during the matching process.

[0129] Learning Methods and Model Knowledge Base: The model building unit establishes a knowledge base containing various learning methods and models (such as decision trees, random forests, support vector machines, and deep learning models). Through learning from and evaluating multiple models, the best-performing model is selected as the basis for building smaller models. This multi-model ensemble approach improves the system's ability to handle complex problems.

[0130] Continuous Model Optimization: The model optimization unit continuously updates and optimizes the enterprise's digital supplier resource matching mini-model by collecting user feedback and combining it with real-time resource data. Furthermore, reinforcement learning algorithms are introduced, enabling the model to continuously learn and improve its matching decision-making capabilities through interaction with the environment.

[0131] Supply Chain Simulation Model: The resource report generation unit establishes a supply chain simulation model to simulate usage scenarios in the supply chain (such as procurement, production, logistics, and sales). Through simulation training, the model can adapt to different supply chain scenarios and improve its ability to cope with complex and nonlinear relationships.

[0132] Data security and integrity assurance: During the data acquisition and storage process, encryption algorithms are used to encrypt sensitive data, the integrity of received data is verified, and the database security audit and monitoring are carried out regularly to ensure the accuracy of data and the robustness of the system.

[0133] Visualized resource matching report: The server outputs a visualized resource matching report, making the matching results intuitive and easy to understand, which facilitates decision analysis for the requesting party.

[0134] In summary, the technical solution of this invention effectively solves the limitations of small model technology in handling complex and nonlinear supply chain relationships by means of multi-dimensional data acquisition and analysis, feature selection algorithms, multi-model integration, continuous model optimization, supply chain simulation models, data security and integrity assurance, and visualization report generation, thereby improving the accuracy and efficiency of the supplier resource matching system.

[0135] This invention provides a model-based resource matching system for enterprise digital transformation, comprising a server, a back-end control terminal, a demand-side terminal, and a supply-side terminal. The server establishes communication connections with the back-end control terminal, the demand-side terminal, and the supply-side terminal.

[0136] Server-side functional unit

[0137] The acquisition unit acquires resource data, including enterprise information, demand data, market dynamics, industry trends, and historical transaction data.

[0138] Establish both relational and non-relational databases, storing acquired resource data into the corresponding databases based on data type. Before data storage, sensitive data from multiple sources is encrypted using encryption algorithms. After data transmission, the server performs integrity verification on the received data. The server records database operation logs, including CRUD operations, and periodically performs security audits and monitoring of the database to check for abnormal access or data tampering.

[0139] The feature matching unit receives demand data, analyzes the demand data using a feature selection algorithm, filters out the features that have the greatest impact on the matching results, and performs matching in the resource matching database to obtain supplier resource matching results.

[0140] Based on the characteristics of the demand data, suitable feature selection algorithms are selected. The extracted features are input into the selected algorithm, and the importance of each feature to the matching results is analyzed. Based on the output of the feature selection algorithm, a matching model is constructed, and searches and comparisons are performed in the resource matching database. The selected supplier resources that meet the requirements are returned to the demand side.

[0141] The model building unit establishes a learning method and model knowledge base, and uses the data generated during the supplier resource matching process and the acquired resource data to build a small digital supplier resource matching model for enterprises.

[0142] Collect and organize various learning methods and models suitable for supplier resource matching, including decision trees, random forests, support vector machines, and deep learning models. Deploy a knowledge base of learning methods and models, including functions for adding, deleting, modifying, and querying test data, as well as mechanisms for calling and updating models.

[0143] The data is integrated and preprocessed to form a dataset suitable for model training. Multiple learning methods and models are used for training, and the performance of each model is recorded and evaluated. The best-performing model is selected and integrated to output a small model for matching digital supplier resources suitable for enterprises. The constructed small model is then deployed to a server for testing and optimization in a real-world environment.

[0144] The model optimization unit collects user feedback and combines it with real-time resource data to continuously update and optimize the enterprise's digital supplier resource matching model.

[0145] A user feedback system is set up on the server to collect feedback information through online surveys, rating systems, or direct user feedback forms. User feedback on matching results, including satisfaction, accuracy, and timeliness, is collected and recorded in real time. User feedback information and real-time resource data are integrated to form a new training dataset, and the stochastic gradient descent algorithm is used to update and optimize the parameters of a small model. A reinforcement learning algorithm is introduced, defining a reinforcement learning environment (including state space, action space, and reward function), enabling the model to continuously learn and improve its matching decision-making ability through interaction with the environment.

[0146] The resource report generation unit establishes a supply chain simulation model, simulates usage scenarios in the supply chain, trains a small model for matching digital supplier resources for enterprises, and generates a visual resource matching report.

[0147] Build a supply chain simulation model on the server to simulate usage scenarios in procurement, production, logistics, sales and other aspects.

[0148] Different usage scenario parameters are set, such as changes in market demand, fluctuations in supplier capacity, and logistical delays. The simulation model is trained under different scenarios, and matching results and performance data are recorded. The continuously updated enterprise digital supplier resource matching mini-model processes user needs data to generate a visualized resource matching report.

[0149] The above describes the specific embodiments of the present invention, detailing the technical implementation steps and measures of each functional unit to address the limitations of small-model technology in handling complex supply chain relationships. Through multi-dimensional data acquisition, accurate feature matching, powerful model building, continuous model optimization, and simulation-based usage scenario training, the accuracy and efficiency of supplier resource matching are effectively improved, assisting SMEs in achieving digital transformation.

[0150] Those skilled in the art will clearly understand that the techniques in the embodiments of the present invention can be implemented using software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solutions in the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments of the present invention.

[0151] The embodiments of the present invention described above do not constitute a limitation on the scope of protection of the present invention.

Claims

1. A model-based resource matching system for enterprise digital transformation, characterized by, The system comprises a server end, a background control end, a demander end and a supplier end, the server end is in communication connection with the background control end, the demander end and the supplier end, the server end constructs an enterprise digital supplier resource matching small model according to data transmitted by the background control end and the supplier end, the server end performs resource matching on demand data of the demander end through the enterprise digital supplier resource matching small model, and the server end outputs a visual resource matching report, and the server end comprises: An acquisition unit acquires resource data, the resource data comprising enterprise information, demand data, market dynamics, industry trends and historical transaction data; A feature matching unit receives demand data, analyzes the demand data by using a feature selection algorithm, screens out features having the greatest influence on matching results, performs matching in a resource matching database according to the features having the greatest influence on the matching results, and obtains a supplier resource matching result; A model construction unit establishes a learning method and a model knowledge base, learns data generated in a supplier resource matching process and acquired resource data by using learning methods and models in the learning method and the model knowledge base in sequence, obtains a learning method and model data set, and constructs an enterprise digital supplier resource matching small model based on the learning method and model data set; A model optimization unit collects user feedback information, uses the user feedback information and real-time resource data as new training data, and continuously updates and optimizes the enterprise digital supplier resource matching small model by using a stochastic gradient descent algorithm, the enterprise digital supplier resource matching small model learns matching decisions in interaction with an environment by using a reinforcement learning algorithm; A resource report generation unit establishes a supply chain simulation model, simulates use scenarios in a supply chain, trains the enterprise digital supplier resource matching small model in the use scenarios, obtains use scenario training data, uses the use scenario training data as new training data, and continuously trains the enterprise digital supplier resource matching small model by using the new training data, obtains a continuously updated enterprise digital supplier resource matching small model, processes data of user demand by using the continuously updated enterprise digital supplier resource matching small model, and obtains a visual resource matching report; The model optimization unit comprises: A user feedback system is arranged on the server, user evaluation and feedback on a supplier resource allocation result are obtained through an online survey, a scoring system or a user feedback form; Feedback information of the users on satisfaction, accuracy and timeliness of the matching result is collected and recorded in real time, the collected user feedback information is sorted and analyzed, and information points are extracted, the information points comprising the most satisfied or unsatisfied matching result and improvement suggestions; The user feedback information and real-time resource data are integrated to form a new training data set, the enterprise digital supplier resource matching small model is updated and optimized in parameters by using a stochastic gradient descent algorithm in combination with the new training data set; The model optimization unit further comprises Defining a reinforcement learning environment for the supplier resource matching, the reinforcement learning environment including a state space, an action space, and a reward function, training the enterprise digital supplier resource matching small model using a reinforcement learning algorithm, the reinforcement learning algorithm will enable the enterprise digital supplier resource matching small model to adjust the matching strategy according to the rewards obtained after obtaining different matching decisions; The enterprise digital supplier resource matching small model interacts with the reinforcement learning environment.

2. The model-based resource matching system for enterprise digital transformation according to claim 1, the acquisition unit comprising: Establish a relational database and a non-relational database, and store the acquired resource data in the corresponding database according to the data type; Before data storage, sensitive data in multi-source data is encrypted using an encryption algorithm, and after data transmission is completed, the server performs integrity checking on the received data; The server records the operation log of the database, including data addition, deletion, modification and query operations, and periodically performs security audit and monitoring on the database to check whether there is abnormal access or data tampering.

3. The model-based resource matching system for enterprise digital transformation according to claim 1, the feature matching unit comprising: According to the characteristics of the demand data, the applicable feature selection algorithm is selected, the extracted features are input into the selected feature selection algorithm, and the importance of each feature to the matching result is analyzed through the feature selection algorithm; According to the output result of the feature selection algorithm, the feature set affecting the matching result is screened out; Using each feature and the matching result affecting feature, a matching model is constructed for searching for qualified supplier resources in the resource matching database.

4. The model-based resource matching system for enterprise digital transformation according to claim 3, the feature matching unit further comprising: Taking the feature values in the demand data as input parameters, searching and comparing in the resource matching database through the matching model; According to the output result of the matching model, the supplier resources meeting the demand are screened out, The supplier resources meeting the demand screened out are subjected to effectiveness detection, and the effective supplier resources meeting the demand are obtained after detection, and the result is returned to the demand side.

5. The model-based resource matching system for enterprise digital transformation according to claim 1, the model construction unit comprising: The server side collects learning methods and models suitable for supplier resource matching, including decision tree, random forest, support vector machine and deep learning model, and constructs a learning method and model knowledge base; The learning method and model knowledge base is used to store and manage the knowledge base architecture of learning methods and models, and the constructed learning method and model knowledge base is deployed on the server. The data generated in the supplier resource matching process and the previously obtained resource data are integrated and preprocessed to form a data set for model training. The model training data set is trained using learning methods and models in the model knowledge base in sequence. During the training process, the performance of each method and model is recorded, including accuracy and recall rate indicators, which are used to evaluate the model effect.

6. The model-based resource matching system for enterprise digital transformation according to claim 5, the model building unit further comprising: receiving enterprise-specific business requirements, including the need to use model algorithms; deploying enterprise-specific business requirements, including the need to use model algorithms, to the server and testing, and continuously monitoring the performance of enterprise-specific business requirements, including the need to use model algorithms, during testing; the performance of each model includes regularly updating the model in the learning method and model knowledge base.

7. The model-based resource matching system for enterprise digital transformation according to claim 1, the resource report generation unit comprising: based on the operation process and rules of the real-time supply chain, a supply chain simulation model is built on the server, which is used to simulate the use scenario in the supply chain, including procurement, production, logistics, and sales links; different use scenario parameters are set in the simulation model, including market demand changes, supplier capacity fluctuations, and logistics delays; by running the simulation model, train the enterprise digital supplier resource matching sub-model under different use scenarios, record the matching results and performance data under each scenario, and form a use scenario training data set.

8. The model-based resource matching system for enterprise digital transformation according to claim 7, the resource report generation unit further comprising: determine the target of the simulation model, analyze the actual demand of enterprise digital supplier resource matching, including supplier information, product demand and procurement history; selecting Simulink simulation tools to build a simulation model of supplier resource matching according to the actual operation of the enterprise, including a supplier library, a product demand library, and a matching algorithm module; set different use scenarios in the simulation model, including changes in the number of suppliers, changes in product demand, and changes in market environment; the supplier data model includes the basic information of the supplier, including name, address, contact information, product list, price, quality, and delivery period.

Images