Intelligent optimization system for enterprise resource planning

Abstract

The invention discloses an intelligent optimization system for enterprise resource planning, and relates to the technical field of enterprise resource management, and the system comprises a data collection and preprocessing layer, a knowledge construction and reasoning layer, a risk early warning and decision support layer, and a resource scheduling and execution layer. According to the invention, a knowledge graph covering various resources of an enterprise and association relationships thereof is constructed by using preprocessed data through a knowledge construction and reasoning layer, a causal relationship network among the data is established, and meanwhile, an intervention effect evaluation unit combines an anti-factual reasoning result of a causal reasoning unit to evaluate the intervention effect of the enterprise. Quantitative evaluation is carried out on different resource pre-allocation intervention schemes, and a detailed evaluation report is output, so that an enterprise can perceive and take corresponding measures in advance before a risk occurs, and can select an optimal scheme according to a scientific evaluation result when a resource planning decision is made, thereby effectively ensuring stable operation of enterprise resources, and improving the resource planning efficiency. And the operation risk caused by the resource problem is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of enterprise resource management, in particular to an intelligent optimization system for enterprise resource planning. BACKGROUND

[0002] In the complex system of enterprise operation management, enterprise resource planning occupies a core position. With the rapid development of information technology, the data generated in the process of enterprise operation presents an explosive growth, and the data types are increasingly diversified, including text, image, sensor data and various enterprise resource information, etc. These multi-modal data contain rich enterprise operation information, which has important value for accurately grasping the enterprise resource situation and optimizing resource allocation. At the same time, modern enterprises are facing increasingly fierce market competition and changing market environment, which requires enterprise resource planning to quickly respond to market changes and adjust resource planning strategies in real time, so as to maximize the utilization of enterprise resources and improve efficiency.

[0003] The traditional enterprise resource planning system has many shortcomings in dealing with the complex and changing operation needs of modern enterprises. In terms of data processing, the traditional system is limited to the processing of single type data and cannot effectively integrate multi-modal heterogeneous data such as text, image and sensor, making it difficult for enterprises to fully obtain various information related to operation, resulting in incomplete decision basis and affecting the accuracy and scientificity of resource planning. In terms of analysis and reasoning ability, the traditional system mainly relies on the statistical correlation of historical data and is difficult to deeply mine the internal causal relationship between data. This shallow analysis method cannot form effective resource planning strategies in the face of rapid changes in market environment, often making enterprises in a passive situation. In addition, the traditional system lacks systematic analysis of various resources and their associated relationships, cannot timely discover problems existing in resource association, and cannot provide early warning of potential risks, making enterprises face great uncertainty in the process of resource planning and increasing operation risk and cost. SUMMARY

[0004] The purpose of the present application is to make up for the shortcomings of the prior art and provide an intelligent optimization system for enterprise resource planning, which can acquire and standardize multi-modal heterogeneous data in the process of enterprise operation through the data acquisition and preprocessing layer, construct a knowledge network based on the preprocessed data in the knowledge construction and reasoning layer, and conduct deep reasoning analysis to mine the causal relationship between data. The risk warning and decision support layer utilizes the knowledge network and reasoning results to realize risk identification and intervention scheme evaluation. The resource scheduling and execution layer generates and executes resource pre-allocation schemes according to the risk warning and decision analysis results, effectively solving the problems of the traditional system such as inability to process multi-modal data, limited analysis and reasoning ability, and lack of resource association risk warning mechanism, realizing intelligent planning and management of enterprise resources, and improving the efficiency of enterprise resource planning.

[0005] The application provides the following technical scheme to solve the above technical problems: an intelligent optimization system for enterprise resource planning, which comprises a data acquisition and preprocessing layer, a knowledge construction and reasoning layer, a risk early warning and decision support layer, and a resource scheduling and execution layer. The data acquisition and preprocessing layer comprises a multi-modal data access unit and a data cleaning and conversion unit, wherein the multi-modal data access unit acquires various types of data, and the data cleaning and conversion unit processes, converts and stores the data. The knowledge construction and reasoning layer comprises a knowledge graph construction unit and a causal reasoning unit, wherein the knowledge graph construction unit constructs a resource knowledge graph, and the causal reasoning unit constructs a causal relationship network and performs counterfactual reasoning. The risk early warning and decision support layer comprises a correlation risk identification unit and an intervention effect evaluation unit, wherein the correlation risk identification unit monitors resource abnormalities and generates early warnings, and the intervention effect evaluation unit evaluates the effect of intervention schemes. The resource scheduling and execution layer comprises a pre-allocation scheme generation unit and a scheme triggering execution unit, wherein the pre-allocation scheme generation unit formulates and optimizes resource pre-allocation schemes, and the scheme triggering execution unit monitors start conditions and executes schemes to achieve dynamic scheduling and execution of resources.

[0006] Further, in the data acquisition and preprocessing layer, the multi-modal data access unit is connected to an enterprise internal market research system, a customer feedback platform and an external public information platform through an API interface, adopts a streaming data transmission protocol to acquire text data in real time, deploys an industrial camera and a monitoring camera device to collect image data, adopts an image compression technology to transmit data, simultaneously installs various sensors, transmits data to a convergence node through a wireless sensor network and then uploads the system to acquire sensor data, and is connected to multiple internal systems of the enterprise to acquire enterprise resource information in real time through a database interface.

[0007] Further, in the knowledge construction and reasoning layer, the knowledge graph construction unit defines the entity types and corresponding attributes of devices and material resources according to standardized resource attribute data, analyzes the business correlations between resource entities and defines correlation rules, adopts a graph database as a storage carrier, converts the standardized resource attribute data into a graph data structure, constructs and perfects a knowledge graph covering various types of resources and their correlation relationships of the enterprise through knowledge extraction and fusion technology, the causal reasoning unit receives various types of data output by the data acquisition and preprocessing layer, adopts a causal discovery algorithm to analyze the data to identify causal relationships, constructs a causal relationship network, filters out significant causal relationships to form stable causal chains by calculating the strength of causal relationships, and based on the causal relationship network, introduces a counterfactual reasoning model, sets different resource parameter adjustment schemes, simulates resource state changes and analyzes the impact on resource supply and demand.

[0008] Further, in the knowledge construction and reasoning layer, the causal reasoning unit receives various types of data output by the data acquisition and preprocessing layer, analyzes the data to identify causal relationships using a causal discovery algorithm, and the algorithm formula is: wherein, represents the data represents the causal relationship strength between the data is the probability of the intervention occurs, is the probability of the intervention does not occur, is the distance decay factor, is the distance between the two in the business process.

[0009] Further, in the knowledge construction and reasoning layer, the causal reasoning unit introduces a counterfactual reasoning model based on the causal relationship network, and the model formula is: wherein is the effect quantitative value of the intervention scheme , that is, the counterfactual reasoning effect quantitative result, is the effect gain weight, is the actual value of the resource target after adopting the scheme , is the baseline value when the scheme is not adopted, is a constant, and are the maximum and minimum values of the resource target, respectively, is the cost decay coefficient, is the implementation cost of the scheme .

[0010] Further, in the risk warning and decision support layer, the associated risk identification unit obtains various types of resource entity state information from the knowledge graph in real time, sets monitoring indicators and threshold values for real-time monitoring, and when the state of a certain type of resource entity exceeds the threshold value and becomes abnormal, traverses the associated path of the entity in the knowledge graph, traces the associated entities to locate the affected resource links, generates risk warning information containing the content of the warning level according to the affected links and the degree of abnormality, and transmits it to the resource scheduling and execution layer; the intervention effect evaluation unit receives different resource pre-allocation intervention schemes proposed by the resource scheduling and execution layer, combines the counterfactual reasoning results of the causal reasoning unit, and quantitatively evaluates the potential effects of each scheme in different scenarios from multiple indicators including resource gap filling degree, and generates an evaluation report on the pros and cons of each scheme according to the evaluation results.

[0011] ​​​Further, in the risk early warning and decision support layer, the potential effects of each scheme in different scenarios are quantitatively evaluated, and the evaluation formula is: Wherein, is the comprehensive effect evaluation value of the intervention scheme, , , is the evaluation index weight, satisfying , respectively corresponding to the resource target improvement degree, the cost control rate and the execution feasibility, is the actual value of the resource target after adopting the intervention scheme, is the baseline value without adopting the scheme, is the preset ideal value of the resource target, is the actual implementation cost of the intervention scheme, is the maximum allowable cost of the intervention scheme, is the scheme execution feasibility coefficient.

[0012] Further, in the resource scheduling and execution layer, the pre-allocation scheme generation unit receives the risk early warning information and intervention effect evaluation report of the risk early warning and decision support layer, and simultaneously obtains constraint condition information including enterprise device load, determines the emergency degree and gap level of resource demand according to the risk early warning information, formulates a resource pre-allocation scheme under the premise of meeting the constraint conditions in combination with the evaluation report, and clearly defines various information of the standby resource and the locking plan, stores the scheme into a database and periodically optimizes and updates according to the actual situation; the scheme triggering execution unit presets the scheme starting conditions including the cause-effect chain node triggering and the risk early warning, and through data interaction with other layers, whether the starting conditions are met is monitored in real time, when the conditions are met, the corresponding scheme is called from the database and the execution process is automatically started.

[0013] Compared with the prior art, the intelligent optimization system of enterprise resource planning has the following beneficial effects: 1. The knowledge construction and reasoning layer constructs a knowledge graph covering various resources of an enterprise and the correlation relationship thereof by using preprocessed data, and establishes a cause-effect relationship network among data, and meanwhile, the intervention effect evaluation unit combines the counterfactual reasoning result of the cause-effect reasoning unit to quantitatively evaluate different resource pre-allocation intervention schemes, and outputs a detailed evaluation report, so that the enterprise can perceive and take measures in advance before the risk occurs, and when formulating a resource planning decision, the optimal scheme can be selected according to the scientific evaluation result, thereby effectively guaranteeing the stable operation of enterprise resources and reducing the operation risk caused by resource problems.

[0014] Secondly, the application can access and process text, image, sensor and enterprise resource information and other multi-modal heterogeneous data in enterprise operation in real time through the data acquisition and preprocessing layer, the data cleaning and conversion unit pre-processes and extracts features of the data, and converts different types of data into structured information that can be directly analyzed, so that the key information in various data is fully mined, the enterprise can fully and accurately master the operation state, based on the high-quality data, the knowledge construction and reasoning layer can construct more accurate knowledge graph and causal relationship network, and provide solid data support for subsequent resource planning, so that the enterprise can realize more accurate and scientific resource planning.

[0015] Other advantages, objects, and features of the application will be set forth in part in the description which follows, and in part will become apparent to those skilled in the art upon examination of the following specification or can be learned by practice of the application. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0017] Figure 1 It is a structural block diagram of an intelligent optimization system for enterprise resource planning. Figure 2 It is a flowchart of an intelligent optimization system for enterprise resource planning. DETAILED DESCRIPTION

[0018] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined application purpose, the specific embodiments, structures, features and effects according to the present application will be described in detail below in combination with the drawings and preferred embodiments.

[0019] Embodiment one

[0020] In the daily operations of a certain manufacturing enterprise, an intelligent enterprise resource planning optimization system based on causal reasoning and multimodal perception has been put into operation. The multimodal data access unit in the data acquisition and preprocessing layer not only accesses text data such as market research reports and customer feedback, image data such as production line equipment status monitoring diagrams and warehouse inventory distribution maps, sensor data such as real-time logistics trajectories, equipment energy consumption data, production cycle data, and meteorological data, as well as information on the company's equipment, materials, personnel, suppliers, and other resources, but also labels the source of this data to ensure data traceability. The data cleaning and conversion unit then processes this data. When parsing the text data into semantic information containing changes in market demand, a semantic recognition model is used to accurately capture signals indicating a surge in demand for a certain new product type. When identifying image data into characteristic parameters such as the operating status of equipment and the storage status of inventory, subtle signs of wear on equipment and inventory stacking density are quantitatively analyzed. When converting sensor data into directly analyzable indicators such as logistics transportation timeliness and production cycle time, some abnormal data caused by temporary sensor failures is eliminated. When converting resource information into standardized resource attribute data, the resource coding format reported by different departments is unified.

[0021] After the knowledge construction and reasoning layer is started, the knowledge graph construction unit constructs a knowledge graph covering various types of resources such as enterprise equipment, materials, personnel, suppliers, and their relationships based on standardized resource attribute data. It not only defines the type and attributes of each resource entity and the association rules between entities, but also adds a weight identifier to each association relationship to distinguish the closeness of the association. The causal reasoning unit analyzes the structured semantic information, feature parameters, and indicator data output by the data acquisition and preprocessing layer, and uses the causal discovery algorithm to analyze the data to identify causal relationships. The algorithm formula is: ,in, Representation data With data The strength of the causal relationship between It is intervention When it occurs The probability of occurrence, It is intervention When it does not happen The probability of occurrence, is the distance attenuation factor, To establish a causal relationship network between the two in the business process, we comprehensively considered time series and business logic in the process of building a causal relationship network between data, identifying the causal chain between different data, such as the chain between heavy rain information in meteorological data and reduced logistics transportation timeliness, which in turn leads to insufficient raw material inventory. We also simulated the changes in resource status under different resource parameter adjustment plans such as adjusting inventory levels and changing production plans through a counterfactual reasoning model. The model formula is: wherein is the effect quantification value of the intervention scheme , i.e. the counterfactual reasoning effect quantification result, is the effect gain weight, is the actual value of the resource target after adopting the scheme , is the baseline value when the scheme is not adopted, and are the maximum and minimum values of the resource target, respectively, is the cost decay coefficient, is the implementation cost of the scheme , which details the fluctuation range of resource supply and demand under each scheme.

[0022] The risk early warning and decision support layer starts working, the associated risk identification unit monitors the state changes of each resource entity in real time based on the knowledge graph, sets a monitoring frequency of once every 10 minutes, when the operating state of a core supplier appears abnormal, the information that its capital chain is under stress is captured, by traversing the associated path of the supplier in the knowledge graph, not only the raw material supply, production plan and other related resource links affected by it are quickly located, but also the downstream finished product assembly links that may be affected are further analyzed, and risk early warning information is generated, the probability of risk occurrence and the loss estimation that may be caused are clarified, the intervention effect evaluation unit combines the counterfactual reasoning result of the causal reasoning unit to quantitatively evaluate different resource pre-allocation intervention schemes for the risk, and the evaluation formula is: wherein, is the comprehensive effect evaluation value of the intervention scheme, , , is the evaluation index weight, satisfying , respectively corresponding to the resource target improvement degree, the cost control rate and the execution feasibility, is the actual value of the resource target after adopting the intervention scheme, is the baseline value when the scheme is not adopted, is the preset ideal value of the resource target, is the actual implementation cost of the intervention scheme, is the maximum allowed cost of this type of intervention scheme, is the scheme execution feasibility coefficient, which is scored from multiple dimensions such as cost, efficiency and feasibility, and outputs a detailed effect evaluation report, which compares the advantages and disadvantages of the two schemes of temporarily finding a substitute supplier and internally allocating inventory.

[0023] The pre-allocation scheme generation unit of the resource scheduling and execution layer generates a resource pre-allocation scheme including a specific location and area of temporary rental warehouse space, a list of cross-department technical workers and skill matching degree, and a standby resource locking plan according to the risk early warning information and the effect evaluation report, in combination with the current device load, human resource scheduling flexibility and other constraint conditions of the enterprise. The scheme triggering execution unit pre-sets the scheme starting conditions. When the data acquisition and preprocessing layer monitors that the logistics delay early warning value in the cause-effect chain reaches the threshold value, it automatically sends a starting instruction to the relevant departments. Each department quickly acts according to the responsibility division in the scheme and executes the resource pre-allocation scheme, thereby ensuring the normal production and operation of the enterprise.

[0024] Embodiment Two

[0025] A retail enterprise introduces the system for resource planning optimization. The multi-modal data access unit of the data acquisition and preprocessing layer accesses real-time text data such as commodity market analysis reports and customer evaluations, image data such as store commodity display images and warehouse inventory images, sensor data such as commodity transportation trajectories and store sales rhythms, and resource information of the enterprise such as commodities, employees, and suppliers. At the same time, it interfaces with the sales system and financial system of the enterprise to realize cross-validation of data. The data cleaning and conversion unit processes these data. In the process of identifying image data as feature parameters such as commodity inventory status and display conditions, the shelf life label of the commodity is identified to remind the soon-to-expire commodity. In the process of converting sensor data into indicators such as commodity transportation time efficiency and sales rhythm, sales data in different time periods are compared and analyzed. In the process of converting resource information into standardized resource attribute data, the classification information of commodities and the performance data of employees are supplemented.

[0026] The knowledge graph construction unit of the knowledge construction and reasoning layer constructs a knowledge graph covering resources such as commodities, employees, and suppliers and their associated relationships based on these standardized resource attribute data. In the graph, information such as the sales peak and off-season of commodities and the areas of expertise of employees is also labeled. The causal reasoning unit analyzes the preprocessed data. In establishing the causal relationship network between data, historical sales data and promotion activity records are referred to, and causal chains such as market demand changes and commodity inventory, sales plans, and causal relationships between promotion activities and commodity sales surges are identified. Through the counterfactual reasoning model, the effects of resource parameter adjustment schemes such as adjusting commodity procurement quantities and changing employee scheduling are simulated, and the sales and customer satisfaction changes under different schemes are predicted.

[0027] The risk early warning and decision support layer is associated with a risk identification unit that monitors the state of resource entities in the knowledge graph in real time, sets a denser monitoring cycle for popular commodities, and when an abnormal situation of supply delay is found in a supplier of a popular commodity, locates not only the affected store sales, inventory management, etc. through the knowledge graph association path, but also tracks the possible impact on the sales of the matching goods of the commodity, generates risk early warning information, and includes the degree of influence of each link. The intervention effect evaluation unit combines the causal reasoning results to quantitatively evaluate different intervention schemes such as commodity allocation and emergency procurement, considers factors such as transportation cost, allocation time and commodity loss, and outputs an evaluation report to provide a comprehensive reference for scheme selection.

[0028] The pre-allocation scheme generation unit of the resource scheduling and execution layer generates a pre-allocation scheme including specific transportation routes and time arrangements for allocating goods from other stores, communication processes and cooperation terms for locking spare resources, etc. according to the risk early warning information and the evaluation report, combined with the constraints such as the storage capacity of the store and the work arrangement of the employees. The scheme triggering execution unit pre-sets the starting conditions, and when the risk early warning issued by the associated risk identification unit reaches a certain level, the pre-allocation scheme is automatically started and executed. In the execution process, the progress of each link is tracked in real time, and the vehicle scheduling conflict problem that occurs in the allocation process is solved in time, effectively alleviating the problem of insufficient commodity supply.

[0029] The above is only a preferred embodiment of the present application, and does not limit the present application in any form. Although the present application has been disclosed as above, it is not intended to limit the present application. Any person skilled in the art can make some changes or modifications to the above disclosed technical content without departing from the scope of the technical solution of the present application, and any brief introduction, modification, equivalent change and modification of the above embodiments made according to the technical essence of the present application are still within the scope of the technical solution of the present application.

Claims

1. An intelligent optimization system for enterprise resource planning, characterized in that: The system includes: data collection and preprocessing layer, knowledge construction and reasoning layer, risk warning and decision support layer and resource scheduling and execution layer; The data acquisition and preprocessing layer includes a multimodal data access unit and a data cleaning and conversion unit, wherein the multimodal data access unit acquires various types of data, and the data cleaning and conversion unit processes, converts and stores the data; The knowledge construction and reasoning layer includes a knowledge graph construction unit and a causal reasoning unit, wherein the knowledge graph construction unit constructs a resource knowledge graph, and the causal reasoning unit constructs a causal relationship network and performs counterfactual reasoning; The risk warning and decision support layer includes an associated risk identification unit and an intervention effect evaluation unit, wherein the associated risk identification unit monitors resource anomalies and generates warnings, and the intervention effect evaluation unit evaluates the effectiveness of intervention plans; The resource scheduling and execution layer includes a pre-allocation scheme generation unit and a scheme trigger execution unit, wherein the pre-allocation scheme generation unit formulates and optimizes the resource pre-allocation scheme, and the scheme trigger execution unit monitors the starting conditions and executes the scheme to realize dynamic scheduling and execution of resources.

2. The intelligent optimization system for enterprise resource planning according to claim 1, characterized in that: In the data collection and preprocessing layer, the multimodal data access unit connects to the company's internal market research system, customer feedback platform and external public information platform through the API interface, adopts the streaming data transmission protocol to obtain text data in real time, and deploys industrial cameras and surveillance camera equipment to collect image data, adopts image compression technology to transmit data, and installs various sensors at the same time. The data is transmitted to the aggregation node through the wireless sensor network and then uploaded to the system to obtain sensor data. At the same time, it connects to multiple internal systems of the enterprise and obtains enterprise resource information in real time through the database interface.

3. The intelligent optimization system for enterprise resource planning according to claim 1, characterized in that: In the knowledge construction and reasoning layer, the knowledge graph construction unit defines the entity types and corresponding attributes of equipment and material resources based on standardized resource attribute data, analyzes the business associations between resource entities and defines association rules, uses a graph database as a storage carrier, converts standardized resource attribute data into a graph data structure, and constructs and improves a knowledge graph covering various types of enterprise resources and their associations through knowledge extraction and fusion technology. The causal reasoning unit receives various types of data output by the data collection and preprocessing layer, uses a causal discovery algorithm to analyze the data to identify causal relationships, constructs a causal relationship network, and screens out significant causal relationships by calculating the strength of the causal relationship to form a stable causal chain. Then, based on the causal relationship network, a counterfactual reasoning model is introduced, different resource parameter adjustment plans are set, resource status changes are simulated, and the impact on resource supply and demand is analyzed.

4. The intelligent optimization system for enterprise resource planning according to claim 3, characterized in that: In the knowledge construction and reasoning layer, the causal reasoning unit receives various data output by the data acquisition and preprocessing layer and uses the causal discovery algorithm to analyze the data to identify causal relationships. The algorithm formula is: ,in, Representation data With data The strength of the causal relationship between It is intervention When it occurs The probability of occurrence, It is intervention When it does not happen The probability of occurrence, is the distance attenuation factor, The distance between the two in the business process.

5. The intelligent optimization system for enterprise resource planning according to claim 4, characterized in that: In the knowledge construction and reasoning layer, the causal reasoning unit introduces a counterfactual reasoning model based on the causal relationship network. The model formula is: ,in For intervention programs The quantitative value of the effect, that is, the quantitative result of the counterfactual reasoning effect, is the effect gain weight, To adopt the plan The actual value of the post-resource target, is a constant, is the baseline value when the scheme is not adopted. and are the maximum and minimum values ​​of the resource target, is the cost reduction coefficient, For the plan implementation costs.

6. The intelligent optimization system for enterprise resource planning according to claim 1, characterized in that: In the risk warning and decision support layer, the associated risk identification unit obtains status information of various resource entities from the knowledge graph in real time, sets monitoring indicators and thresholds for real-time monitoring, and when a certain type of resource entity status exceeds the threshold and an abnormality occurs, traverses the associated path of the entity in the knowledge graph, tracks the associated entities to locate the affected resource links, generates risk warning information containing warning level content based on the affected links and the degree of abnormality, and transmits it to the resource scheduling and execution layer; The intervention effect evaluation unit receives different resource pre-allocation intervention plans proposed by the resource scheduling and execution layer, and combines the counterfactual reasoning results of the causal reasoning unit to quantitatively evaluate the potential effects of each plan in different scenarios based on multiple indicators including the degree of filling of resource gaps. Based on the evaluation results, an evaluation report is generated to rank and analyze the advantages and disadvantages of each plan.

7. The intelligent optimization system for enterprise resource planning according to claim 6, characterized in that: In the risk warning and decision support layer, the potential effects of each plan in different scenarios are quantitatively evaluated, and the evaluation formula is as follows: ,in, is the comprehensive effect evaluation value of the intervention program, 、 、 is the evaluation index weight, satisfying , corresponding to resource target improvement, cost control rate and execution feasibility, respectively. is the actual value of the resource target after the intervention plan is adopted, is the baseline value when the scheme is not adopted. is the preset ideal value of resource target, is the actual implementation cost of the intervention program, is the maximum allowable cost of this type of intervention, It is the feasibility coefficient of the plan implementation.

8. The intelligent optimization system for enterprise resource planning according to claim 1, characterized in that: In the resource scheduling and execution layer, the pre-allocation plan generation unit receives risk warning information and intervention effect evaluation reports from the risk warning and decision support layer, and simultaneously obtains constraint information including enterprise equipment load. Based on the risk warning information, the urgency of resource demand and the magnitude of the gap are determined. Combined with the evaluation report, a resource pre-allocation plan is formulated on the premise of meeting the constraint conditions, clarifying various types of backup resource information and locking plans. The plan is stored in the database and regularly optimized and updated according to actual conditions. The scheme trigger execution unit presets scheme start-up conditions including causal chain node triggering and risk warning. It monitors whether the start-up conditions are met in real time through data interaction with other layers. When the conditions are met, the corresponding scheme is retrieved from the database and the execution process is automatically started.

Images