Enterprise diagnosis and prediction system and method based on intelligent analysis

Abstract

The invention discloses an enterprise diagnosis prediction system and method based on intelligent analysis, and the system remarkably improves the prediction accuracy of enterprise operation and the scientificity of management decision through the design of a prediction process module and a report generation module. The prediction process module performs accurate modeling on operation data of an enterprise by using an advanced rapid sparse ridge regression method to generate a prediction report of key operation indexes, so that the enterprise can insight into future operation trends and potential risks in advance and perform strategic adjustment in time; meanwhile, the report generation module performs comprehensive comparative analysis on enterprise execution capability analysis, historical data analysis and a future operation condition report and the current state of the enterprise to generate a comprehensive operation diagnosis report, so that a management layer is helped to evaluate the operation efficiency and the development prospect of the enterprise from multiple dimensions; and short boards and potential risks in operation are quickly identified and solved, so that resource allocation is optimized, and the overall competitiveness of enterprises and the capability of coping with market changes are improved.

Description

Technical Field

[0001] The present invention belongs to the field of enterprise diagnosis and prediction, and specifically, relates to an enterprise diagnosis and prediction system and method based on intelligent analysis. Background Art

[0002] In business management and operations, accurately predicting future development trends and making scientific management decisions in a timely manner are crucial to maintaining corporate competitiveness. However, existing enterprise diagnosis and forecasting systems often have problems such as insufficient prediction accuracy, single data analysis dimensions, and difficulty in providing comprehensive decision support. Traditional forecasting methods, such as simple time series analysis or linear regression models, often fail to fully capture the complex nonlinear relationships and multi-dimensional characteristics in corporate operating data, resulting in low accuracy and reliability of forecast results. In addition, existing systems also have limitations in data integration and report generation. They can often only analyze single historical data or current status, lack forward-looking analysis of future trends, and cannot provide comprehensive operational diagnostic reports. Therefore, it is difficult to support management in making effective strategic adjustments and resource allocation decisions in the face of a complex market environment.

[0003] In view of this, the present invention is proposed. Summary of the invention

[0004] The technical problem to be solved by the present invention is to overcome the deficiencies of the prior art and provide an enterprise diagnosis and prediction system and method based on intelligent analysis, which solves the problems raised in the above-mentioned background technology.

[0005] In order to solve the above technical problems, the basic concept of the technical solution adopted by the present invention is:

[0006] An enterprise diagnosis and prediction system based on intelligent analysis, comprising:

[0007] Data collection module, which is used to collect the enterprise's operational data and historical data;

[0008] Data preprocessing module, the data preprocessing module is used to normalize the collected data to remove noise and format it;

[0009] Enterprise execution capability analysis module: The enterprise execution capability analysis module is used to analyze the instruction response time, rework frequency and delayed completion time in the enterprise operation execution process, generate the enterprise execution capability analysis coefficient, and analyze whether the enterprise execution capability is qualified based on the execution capability analysis coefficient and generate an execution capability analysis report based on the results;

[0010] Enterprise historical data analysis module: The enterprise historical data analysis module is used to analyze the enterprise's historical operations and formulate the completion ratio and time difference of the task volume in the process of analyzing the enterprise's historical operations, and generate historical data analysis results;

[0011] Prediction model training module: The prediction model training module uses the fast sparse ridge regression method to model the enterprise's operating data and train a prediction model for the enterprise's future operating indicators;

[0012] Prediction process module: The prediction process module uses the trained prediction model and the latest data to predict the future operating indicators of the enterprise and generate a report on the future operating status of the enterprise;

[0013] The report generation module is used to combine the execution capability analysis report, the enterprise's future operating status report, the historical data analysis results and the current status of the enterprise to compare and analyze the report, generate an enterprise operation diagnosis report, and provide management decision support.

[0014] Optionally, when the enterprise execution capability analysis module needs to generate the enterprise execution capability analysis coefficient, it first obtains the production time and actual execution time of each instruction of the enterprise during the operation execution process from the data preprocessing module, calculates the time interval from the production to the execution of each instruction, and marks it as the response time, then counts the frequency of the enterprise reworking or re-executing instructions during the execution process, and marks it as the rework frequency, and finally records the planned completion time and actual completion time of each instruction, and calculates the delay time of the actual completion time relative to the planned completion time; after obtaining the delay time, the response time JGS, the rework frequency CPL and the delayed completion time YSC are added with the preset weight coefficient to calculate the execution capability score. The analysis coefficient X is expressed as: X=a1×JGS+a2×CPL+a3×YSC+β, where a1, a2, and a3 are preset weight coefficients, and a1>a2>a3, and β is an error correction factor; then, a threshold value of the execution capability analysis coefficient is preset as a benchmark for judging whether the execution capability of the enterprise is qualified, and the calculated execution capability analysis coefficient X is compared with the preset threshold value. If the execution capability analysis coefficient X exceeds the threshold value, it means that there are problems in the operation and execution process of the enterprise and the execution capability is unqualified. If the execution capability analysis coefficient X does not exceed the threshold value, it means that the execution capability of the enterprise is in a qualified state; finally, according to the comparison result, a corresponding execution capability analysis report is generated.

[0015] Optionally, when obtaining the historical operation report of the enterprise, first collect the planned task volume and the actual completed task volume P of each task in the historical operation process of the enterprise, and use the formula Calculate the task completion ratio, then record the planned completion time and actual completion time of each task, and calculate the completion time difference T by the formula T=DS, where W is the actual completed task amount, J is the planned completion amount, D is the actual completion time, and S is the planned completion time;

[0016] After obtaining the actual amount of completed tasks P and the completion time difference T, set the task completion ratio threshold and time difference threshold according to the enterprise management requirements or industry standards, compare the actual task completion ratio P obtained with the preset task volume ratio threshold, and compare the actual completion time difference T with the preset time difference threshold; if the task completion ratio and time difference are both qualified, generate a historical operation analysis qualified signal, indicating that the historical execution performance of the task is in line with expectations; if the task completion ratio or time difference is unqualified, generate a historical operation analysis unqualified signal, indicating that there are problems with the historical execution of the task; finally, summarize the analysis results of all tasks, form an overall evaluation of the entire enterprise's historical operations, and generate an analysis report on the enterprise's historical operation performance based on the comprehensive analysis results, pointing out areas of excellent performance and areas that need improvement.

[0017] Optionally, the steps of using the fast sparse ridge regression method to model the enterprise's operating data and train a prediction model for the enterprise's future operating indicators are:

[0018] Receive the historical operation data sent by the data preprocessing module and select features related to the company's future operation indicators from the historical operation data. Then, according to business needs, construct new feature variables, and after receiving the historical operation data sent by the preprocessing module, set aside some data as a test set; then, select the fast sparse ridge regression algorithm to build the prediction model and use the processed data to train the prediction model. During the training process, use linear regression with regularization terms to control the complexity of the prediction model; then, use the test set to evaluate the prediction accuracy and error of the prediction model and adjust the prediction model parameters according to the verification results.

[0019] Optionally, when using the trained prediction model and the latest data to predict the future operation indicators of the enterprise, first, the prediction model is deployed in the operation environment of the enterprise, and then the latest enterprise operation data is collected by the data collection module and the latest enterprise operation data is transmitted to the data preprocessing module for preprocessing, and finally the latest data preprocessed by the data preprocessing module is input into the trained prediction model;

[0020] The prediction model predicts the input data and outputs the predicted values ​​of the company's operating indicators within a specified period of time in the future. While obtaining the predicted values, the prediction model will also calculate the company's key operating indicators in the future. Then, using the prediction results output by the prediction model, analyze the volatility of the company's operating indicators in the future and evaluate the risks that the company will face in the future based on the prediction results. After obtaining the evaluation results, judge whether the company has the ability to cope with it based on the evaluation results, and generate a risk assessment report; finally, summarize all predicted key indicators to form a complete set of prediction results and generate a report on the company's future operating status based on the summarized prediction results.

[0021] Optionally, when judging whether the enterprise has the ability to respond based on the assessment results, first determine the key performance indicators that the enterprise must achieve when responding to risks and set thresholds for each key indicator. These thresholds represent the minimum level required for the enterprise to effectively respond to risks. Then, assign weights to each key indicator based on the degree of impact of the key indicator on the overall risk response capability. i , and then by the formula Calculate the comprehensive score S total Finally, according to the comprehensive score S total And compare it with the set threshold. If the comprehensive score is higher than the set threshold, it means that the company has the ability to cope with it as a whole. If the comprehensive score is lower than the threshold, the company's coping ability is insufficient.

[0022] Optionally, when generating an enterprise operation diagnosis report to provide management decision support, first, collect and summarize the execution capability analysis report, the enterprise future operation status report, the historical data analysis results, and the enterprise's current status data, and conduct a comprehensive analysis to obtain comprehensive data;

[0023] For each key indicator, analyze the difference between execution capability, historical performance, future forecast and current status, and use the formula Calculate and summarize these differences Δ k , identify significant deviation areas, then analyze whether the trend in the comprehensive data is consistent with the current actual operation, and finally compile a comprehensive enterprise operation diagnosis report based on the comprehensive analysis results, difference analysis and trend analysis results and submit the diagnosis report to the enterprise, where Δ k is the comprehensive difference value for key indicator k, C k is the actual value of the key indicator k in the current state, P k is the value of the key indicator k in the predicted value, H k is the value of the key indicator k in historical performance, σ k is the volatility of the key indicator k, w1, w2, w3 are weight factors used to adjust the impact of each difference in the comprehensive difference, and ∈ is a positive number.

[0024] Optionally, the comprehensive analysis includes comparing execution capability with the current status, comparing historical data with the current status, and comparing predicted results with the current status. When comparing execution capability with the current status, the key indicators in the execution capability analysis report are compared with the data of the company's current status, and whether the actual execution capability has reached the expected or historical level is analyzed to identify current efficiency problems. When comparing historical data with the current status, historical operating data is compared with current operating data to analyze the changing trends of the company's operations, identify continuous improvement or decline patterns, and explore the reasons for these changes. When comparing predicted results with the current status, the predicted data in the company's future operating status report is compared with the current actual data to evaluate whether the company's current operations are in line with the predicted trends and analyze the potential causes of any deviations.

[0025] An enterprise diagnosis and prediction method based on intelligent analysis includes the following steps:

[0026] Collect the enterprise's operational and historical data and normalize the collected data to remove noise and format it;

[0027] Analyze the instruction response time, rework frequency and delayed completion time during the enterprise's operation execution process, generate the enterprise's execution capability analysis coefficient, and analyze whether the enterprise's execution capability is qualified based on the execution capability analysis coefficient and generate an execution capability analysis report based on the results;

[0028] Analyze the completion ratio and time difference of the tasks set in the historical operation process of the enterprise, and generate historical data analysis results;

[0029] Use the fast sparse ridge regression method to model the enterprise's operating data and train a prediction model for the enterprise's future operating indicators;

[0030] Use the trained prediction model and the latest data to predict the company's future operating indicators and generate a report on the company's future operating status;

[0031] Combine the execution capability analysis report, the enterprise's future operating status report, and the historical data analysis results with the enterprise's current status for comparative analysis to generate an enterprise operation diagnosis report to provide management decision support.

[0032] After adopting the above technical solution, the present invention has the following beneficial effects compared with the prior art. Of course, any product implementing the present invention does not necessarily need to achieve all the advantages described below at the same time:

[0033] The present invention significantly improves the forecast accuracy of enterprise operations and the scientific nature of management decisions by designing a forecast process module and a report generation module. The forecast process module uses an advanced fast sparse ridge regression method to accurately model the enterprise's operating data and generate a forecast report on key operating indicators, so that the enterprise can gain insight into future operating trends and potential risks in advance and make timely strategic adjustments; at the same time, the report generation module conducts a comprehensive comparative analysis of the enterprise's execution capability analysis, historical data analysis, and future operating status report with the enterprise's current status to generate a comprehensive operational diagnosis report, which helps management evaluate the enterprise's operating efficiency and development prospects from multiple dimensions, quickly identify and resolve shortcomings and potential risks in operations, thereby optimizing resource allocation and enhancing the enterprise's overall competitiveness and ability to respond to market changes.

[0034] The specific implementation modes of the present invention are further described in detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The drawings described below are only some embodiments. For those skilled in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0036] Figure 1 A block diagram of a diagnosis prediction system according to an embodiment of the present invention;

[0037] Figure 2 FIG. 4 is a flowchart of a diagnosis prediction method according to an embodiment of the present invention.

[0038] It should be noted that these drawings and textual descriptions are not intended to limit the conceptual scope of the present invention in any way, but are intended to illustrate the concept of the present invention for those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0039] The present invention will now be described in further detail with reference to the accompanying drawings.

[0040] See also Figure 1 As shown, in this embodiment, a system for enterprise diagnosis and prediction based on intelligent analysis is provided, including a data acquisition module, which is used to collect the operation data and historical data of the enterprise;

[0041] Data preprocessing module, the data preprocessing module is used to normalize the collected data to remove noise and format it;

[0042] Enterprise execution capability analysis module: The enterprise execution capability analysis module is used to analyze the instruction response time, rework frequency and delayed completion time in the enterprise operation execution process, generate the enterprise execution capability analysis coefficient, and analyze whether the enterprise execution capability is qualified based on the execution capability analysis coefficient and generate an execution capability analysis report based on the results;

[0043] Enterprise historical data analysis module: The enterprise historical data analysis module is used to analyze the enterprise's historical operations and formulate the completion ratio and time difference of the task volume in the process of analyzing the enterprise's historical operations, and generate historical data analysis results;

[0044] Prediction model training module: The prediction model training module uses the fast sparse ridge regression method to model the enterprise's operating data and train a prediction model for the enterprise's future operating indicators;

[0045] Prediction process module: The prediction process module uses the trained prediction model and the latest data to predict the future operating indicators of the enterprise and generate a report on the future operating status of the enterprise. The future operating indicators include key indicators such as total profit and development stability;

[0046] The report generation module is used to combine the execution capability analysis report, the enterprise's future operating status report, the historical data analysis results and the current status of the enterprise to compare and analyze the report, generate an enterprise operation diagnosis report, and provide management decision support.

[0047] It should be noted that: by designing the forecasting process module and the report generation module, the forecasting accuracy of enterprise operations and the scientific nature of management decisions have been significantly improved. The forecasting process module uses advanced fast sparse ridge regression methods to accurately model the enterprise's operating data and generate forecast reports on key operating indicators, enabling enterprises to gain insight into future operating trends and potential risks in advance and make timely strategic adjustments; at the same time, the report generation module conducts a comprehensive comparative analysis of the enterprise's execution capability analysis, historical data analysis, and future operating status reports with the current status of the enterprise to generate a comprehensive operational diagnostic report, helping management to evaluate the enterprise's operating efficiency and development prospects from multiple dimensions, quickly identify and resolve shortcomings and potential risks in operations, thereby optimizing resource allocation and enhancing the overall competitiveness of the enterprise and its ability to respond to market changes.

[0048] In this embodiment, when the enterprise execution capability analysis module needs to generate the enterprise execution capability analysis coefficient, it first obtains the production time and actual execution time of each instruction of the enterprise during the operation execution process from the data preprocessing module, calculates the time interval from the production to the execution of each instruction, and marks it as the response time, then counts the frequency of the enterprise reworking or re-executing the instruction during the execution process, and marks it as the rework frequency, and finally records the planned completion time and actual completion time of each instruction, and calculates the delay time of the actual completion time relative to the planned completion time; after obtaining the delay time, the response time JGS, the rework frequency CPL and the delayed completion time YSC are added with the preset weight coefficient to calculate the execution capability. The analysis coefficient X is expressed as: X = a1 × JGS + a2 × CPL + a3 × YSC + β, where a1, a2, and a3 are preset weight coefficients, and a1> a2> a3, and β is an error correction factor; then, a threshold value of the execution capability analysis coefficient is preset as a benchmark for judging whether the enterprise's execution capability is qualified, and the calculated execution capability analysis coefficient X is compared with the preset threshold value. If the execution capability analysis coefficient X exceeds the threshold value, it means that there are problems in the operation and execution process of the enterprise, and the execution capability is unqualified. If the execution capability analysis coefficient X does not exceed the threshold value, it means that the enterprise's execution capability is in a qualified state; finally, according to the comparison result, a corresponding execution capability analysis report is generated. If X exceeds the threshold value, an "execution capability enhancement signal" is generated, indicating that improvement is needed; if X does not exceed the threshold value, a "qualified execution capability signal" is generated, indicating that the execution capability is normal.

[0049] In this embodiment, when obtaining the historical operation report of the enterprise, first, the planned task volume and the actual completed task volume P of each task in the historical operation process of the enterprise are collected, and the formula Calculate the task completion ratio, then record the planned completion time and actual completion time of each task, and calculate the completion time difference T by the formula T=DS, where W is the actual completed task amount, J is the planned completion amount, D is the actual completion time, and S is the planned completion time;

[0050] After obtaining the actual amount of completed tasks P and the completion time difference T, set the task completion ratio threshold and time difference threshold according to the enterprise management requirements or industry standards, compare the actual task completion ratio P obtained with the preset task volume ratio threshold, and compare the actual completion time difference T with the preset time difference threshold; if the task completion ratio and time difference are both qualified, generate a historical operation analysis qualified signal, indicating that the historical execution performance of the task is in line with expectations; if the task completion ratio or time difference is unqualified, generate a historical operation analysis unqualified signal, indicating that there are problems with the historical execution of the task; finally, summarize the analysis results of all tasks, form an overall evaluation of the entire enterprise's historical operations, and generate an analysis report on the enterprise's historical operation performance based on the comprehensive analysis results, pointing out areas of excellent performance and areas that need improvement.

[0051] In this embodiment, the steps of using the fast sparse ridge regression method to model the enterprise's operating data and train a prediction model for the enterprise's future operating indicators are as follows:

[0052] Receive the historical operation data sent by the data preprocessing module and select features related to the company's future operation indicators from the historical operation data. Then, according to business needs, construct new feature variables, and after receiving the historical operation data sent by the preprocessing module, set aside some data as a test set; then, select the fast sparse ridge regression algorithm to build the prediction model and use the processed data to train the prediction model. During the training process, use linear regression with regularization terms to control the complexity of the prediction model; then, use the test set to evaluate the prediction accuracy and error of the prediction model and adjust the prediction model parameters according to the verification results.

[0053] In this embodiment, when using the trained prediction model and the latest data to predict the future operating indicators of the enterprise, first, the prediction model is deployed in the enterprise's operating environment, then the data collection module is used to collect the latest enterprise operating data and the latest enterprise operating data is transmitted to the data preprocessing module for preprocessing, and finally the latest data preprocessed by the data preprocessing module is input into the trained prediction model;

[0054] The prediction model predicts the input data and outputs the predicted values ​​of the company's operating indicators within a specified period of time in the future. While obtaining the predicted values, the prediction model will also calculate the company's key operating indicators in the future. Then, using the prediction results output by the prediction model, analyze the volatility of the company's operating indicators in the future and evaluate the risks that the company will face in the future based on the prediction results. After obtaining the evaluation results, judge whether the company has the ability to cope with it based on the evaluation results, and generate a risk assessment report; finally, summarize all predicted key indicators to form a complete set of prediction results and generate a report on the company's future operating status based on the summarized prediction results.

[0055] In this embodiment, when judging whether an enterprise has the ability to cope with risks based on the evaluation results, first, the key performance indicators that the enterprise must achieve when coping with risks are determined, such as cash flow, debt-to-asset ratio, profit margin, and inventory turnover rate, and thresholds are set for each key indicator. These thresholds represent the minimum level required for the enterprise to effectively cope with risks. Then, a weight w is assigned to each key indicator based on the degree of influence of the key indicator on the overall risk coping ability. i , and then by the formula Calculate the comprehensive score S total Finally, according to the comprehensive score S total And compare it with the set threshold. If the comprehensive score is higher than the set threshold, it means that the company has the ability to cope with it as a whole. If the comprehensive score is lower than the threshold, the company's coping ability is insufficient.

[0056] In this embodiment, when generating an enterprise operation diagnosis report to provide management decision support, first, collect and summarize the execution capability analysis report, the enterprise future operation status report, the historical data analysis results, and the enterprise's current status data, and conduct a comprehensive analysis to obtain comprehensive data;

[0057] For each key indicator, analyze the difference between execution capability, historical performance, future forecast and current status, and use the formula Calculate and summarize these differences Δ k , identify significant deviation areas, then analyze whether the execution capabilities, historical performance and forecast trends in the comprehensive data are consistent with the current actual operations, and finally compile a comprehensive enterprise operation diagnosis report based on the comprehensive analysis results, difference analysis and trend analysis results and submit the diagnosis report to the enterprise, where Δ k is the comprehensive difference value for key indicator k, C k is the actual value of the key indicator k in the current state, P k is the value of the key indicator k in the predicted value, H k is the value of the key indicator k in historical performance, σ k is the volatility of the key indicator k, w1, w2, w3 are weight factors used to adjust the impact of each difference in the comprehensive difference, and ∈ is a positive number.

[0058] In this embodiment, the comprehensive analysis includes comparing the execution capability with the current state, comparing the historical data with the current state, and comparing the predicted results with the current state. Among them, when comparing the execution capability with the current state, the key indicators (such as execution efficiency, rework rate, etc.) in the execution capability analysis report are compared with the data of the enterprise's current state, and the actual execution capability is analyzed to see whether it has reached the expected or historical level, and the current efficiency problems are identified; when comparing the historical data with the current state, the historical operating data is compared with the current operating data, and the changing trend of the enterprise's operation is analyzed to identify the continuous improvement or decline mode, and the reasons for these changes are explored; when comparing the predicted results with the current state, the predicted data in the enterprise's future operating status report is compared with the current actual data to evaluate whether the enterprise's current operation is in line with the predicted trend and analyze the potential reasons for any deviation.

[0059] See also Figure 2 As shown, in this embodiment, a method for enterprise diagnosis and prediction based on intelligent analysis is provided, including the following steps:

[0060] Collect the enterprise's operational and historical data and normalize the collected data to remove noise and format it;

[0061] Analyze the instruction response time, rework frequency and delayed completion time during the enterprise's operation execution process, generate the enterprise's execution capability analysis coefficient, and analyze whether the enterprise's execution capability is qualified based on the execution capability analysis coefficient and generate an execution capability analysis report based on the results;

[0062] Analyze the completion ratio and time difference of the tasks set in the historical operation process of the enterprise, and generate historical data analysis results;

[0063] Use the fast sparse ridge regression method to model the enterprise's operating data and train a prediction model for the enterprise's future operating indicators;

[0064] Use the trained prediction model and the latest data to predict the company's future operating indicators and generate a report on the company's future operating status;

[0065] Combine the execution capability analysis report, the enterprise's future operating status report, and the historical data analysis results with the enterprise's current status for comparative analysis to generate an enterprise operation diagnosis report to provide management decision support.

[0066] Example: A manufacturing company hopes to improve its operational efficiency and decision-making capabilities through an intelligent analysis system. The company first deployed a data acquisition module to collect real-time operational data on its production line and historical data in the financial system. The historical data includes the generation time and execution time of each production instruction, the frequency of rework due to quality issues, and key financial indicators such as monthly revenue, expenditure, and profit for the past five years.

[0067] The data preprocessing module then processes the collected data. For example, outliers in financial data (such as missed or duplicated transactions) are identified and corrected through denoising. All data are min-max normalized to scale each data item to the [0, 1] interval for subsequent modeling and analysis;

[0068] The system extracts indicators related to execution capability from the preprocessed data. For example, the generation time of a production instruction is 10:00 on May 1, 2023, and the actual execution time is 10:30 on May 1, 2023. The response time of the instruction is 30 minutes. In addition, the company has statistics that 20 production instructions were reworked due to quality problems in the past month, and 1,000 instructions were produced. Therefore, the rework frequency is 2%. Secondly, an instruction is planned to be completed at 12:00 on May 2, 2023, and the actual completion time is 13:00 on May 2, 2023, with a delay of 1 hour. The execution capability analysis module combines these indicators and calculates the execution capability analysis coefficient X. The weight coefficients are a1=0.5, a2=0.3, and a3=0.2, respectively. The calculation formula of the execution capability analysis coefficient X is: The threshold is set to 0.25, X = 0.2708> 0.25, indicating that the company's execution capability during this period is unqualified, and an execution capability analysis report is generated, and it is recommended to improve execution efficiency. In addition, the company's goal in 2022 is to produce 10,000 products per month, and the actual average monthly completion is 9,800 pieces, so the task completion ratio P is 98%, The planned time is 30 days per month, but the actual completion time is an average of 32 days. The time difference T is, T = 32-30 = 2 days. The set task completion ratio threshold is 95%, and the time difference threshold is 5 days. Since the task completion ratio P meets the requirements and the time difference T is within an acceptable range, the system generates a "historical operation analysis qualified signal" and concludes that the operation is relatively stable in 2022. Next, the company uses the prediction model trained by the fast sparse ridge regression method to predict future operation indicators. The system models historical operation data, selects features such as production volume, rework rate, profit margin, etc., and builds a prediction model. Assume that when the model is trained, 70% of the data is used for training and 30% for testing, and the accuracy of the model reaches 95% on the test set. Then using the trained model, the system predicts the total profit and production stability in the third quarter of 2023. It is predicted that the total profit in the next three months will reach 15 million yuan and by analyzing the forecast data of the profit, it is found that the profit may drop by 10% in the third month due to market fluctuations.

[0069] The final report generation module combines the execution capability analysis report, historical data analysis results and future operation status report, and makes a comprehensive comparison with the current operation data. For example, it is found that the execution capability in May 2023 is lower than the historical average, while the profit forecast for the next three months shows an increased risk.

[0070] Through comprehensive analysis, the system identifies the main risk points of the enterprise and generates a detailed operation diagnosis report, suggesting that the enterprise adjust the production process, improve execution efficiency, and prepare for possible profit fluctuations. This report is submitted to the management and provides reliable support for decision-making.

[0071] The present invention is not limited to the above-mentioned embodiments. Anyone should be aware that any structural changes made under the enlightenment of the present invention, and any technical solutions that are the same or similar to the present invention, fall within the protection scope of the present invention. The technology, shape, and structural parts not described in detail in the present invention are all well-known technologies.

Claims

1. An enterprise diagnosis and prediction system based on intelligent analysis, characterized in that: include: Data collection module, which is used to collect the enterprise's operational data and historical data; Data preprocessing module, the data preprocessing module is used to normalize the collected data to remove noise and format it; Enterprise execution capability analysis module: The enterprise execution capability analysis module is used to analyze the instruction response time, rework frequency and delayed completion time in the enterprise operation execution process, generate the enterprise execution capability analysis coefficient, and analyze whether the enterprise execution capability is qualified based on the execution capability analysis coefficient and generate an execution capability analysis report based on the results; Enterprise historical data analysis module: The enterprise historical data analysis module is used to analyze the enterprise's historical operations and formulate the completion ratio and time difference of the task volume in the process of analyzing the enterprise's historical operations, and generate historical data analysis results; Prediction model training module: The prediction model training module uses the fast sparse ridge regression method to model the enterprise's operating data and train a prediction model for the enterprise's future operating indicators; Prediction process module: The prediction process module uses the trained prediction model and the latest data to predict the future operating indicators of the enterprise and generate a report on the future operating status of the enterprise; The report generation module is used to combine the execution capability analysis report, the enterprise's future operating status report, the historical data analysis results and the current status of the enterprise to compare and analyze the report, generate an enterprise operation diagnosis report, and provide management decision support.

2. The enterprise diagnosis and prediction system based on intelligent analysis according to claim 1 is characterized in that: When the enterprise execution capability analysis module needs to generate the enterprise execution capability analysis coefficient, it first obtains the production time and actual execution time of each instruction of the enterprise during the operation execution process from the data preprocessing module, calculates the time interval from the production to the execution of each instruction and marks it as the response time, then counts the frequency of rework or re-execution of instructions in the execution process of the enterprise, and marks it as the rework frequency, and finally records the planned completion time and actual completion time of each instruction, and calculates the delay time of the actual completion time relative to the planned completion time; after obtaining the delay time, the response time JGS, the rework frequency CPL and the delayed completion time YSC are added with the preset weight coefficient to calculate the execution capability analysis coefficient. The number X is expressed as: X=a1×JGS+a2×CPL+a3×YSC+β, where a1, a2, and a3 are preset weight coefficients, and a1>a2>a3, and β is an error correction factor; then, a threshold value of the execution capability analysis coefficient is preset as a benchmark for judging whether the execution capability of the enterprise is qualified, and the calculated execution capability analysis coefficient X is compared with the preset threshold value. If the execution capability analysis coefficient X exceeds the threshold value, it means that there are problems in the operation and execution process of the enterprise and the execution capability is unqualified. If the execution capability analysis coefficient X does not exceed the threshold value, it means that the execution capability of the enterprise is in a qualified state; finally, according to the comparison result, a corresponding execution capability analysis report is generated.

3. The enterprise diagnosis and prediction system based on intelligent analysis according to claim 1 is characterized in that: When obtaining the historical operation report of the enterprise, first collect the planned task volume and the actual completed task volume P of each task in the historical operation process of the enterprise, and use the formula Calculate the task completion ratio, then record the planned completion time and actual completion time of each task, and calculate the completion time difference T by the formula T=DS, where W is the actual completed task amount, J is the planned completion amount, D is the actual completion time, and S is the planned completion time; After obtaining the actual amount of completed tasks P and the completion time difference T, set the task completion ratio threshold and time difference threshold according to the enterprise management requirements or industry standards, compare the actual task completion ratio P obtained with the preset task volume ratio threshold, and compare the actual completion time difference T with the preset time difference threshold; if the task completion ratio and time difference are both qualified, generate a historical operation analysis qualified signal, indicating that the historical execution performance of the task is in line with expectations; if the task completion ratio or time difference is unqualified, generate a historical operation analysis unqualified signal, indicating that there are problems with the historical execution of the task; finally, summarize the analysis results of all tasks, form an overall evaluation of the entire enterprise's historical operations, and generate an analysis report on the enterprise's historical operation performance based on the comprehensive analysis results, pointing out areas of excellent performance and areas that need improvement.

4. The enterprise diagnosis and prediction system based on intelligent analysis according to claim 1 is characterized in that: The steps of using the fast sparse ridge regression method to model the enterprise's operating data and train a prediction model for the enterprise's future operating indicators are as follows: Receive the historical operation data sent by the data preprocessing module and select features related to the company's future operation indicators from the historical operation data. Then, according to business needs, construct new feature variables, and after receiving the historical operation data sent by the preprocessing module, set aside some data as a test set; then, select the fast sparse ridge regression algorithm to build the prediction model and use the processed data to train the prediction model. During the training process, use linear regression with regularization terms to control the complexity of the prediction model; then, use the test set to evaluate the prediction accuracy and error of the prediction model and adjust the prediction model parameters according to the verification results.

5. The enterprise diagnosis and prediction system based on intelligent analysis according to claim 1 is characterized in that: When using the trained prediction model and the latest data to predict the future operating indicators of the enterprise, first, the prediction model is deployed in the enterprise's operating environment, and then the data collection module is used to collect the latest enterprise operating data and the latest enterprise operating data is transmitted to the data preprocessing module for preprocessing, and finally the latest data preprocessed by the data preprocessing module is input into the trained prediction model; The prediction model predicts the input data and outputs the predicted values ​​of the enterprise's operating indicators within a specified period of time in the future. While obtaining the predicted values, the prediction model also calculates the key operating indicators of the enterprise in the future. Then, using the prediction results output by the prediction model, analyze the volatility of the enterprise's operating indicators in the future and evaluate the risks that the enterprise will face in the future based on the prediction results. After obtaining the evaluation results, it determines whether the enterprise has the ability to cope with them based on the evaluation results and generates a risk assessment report; Finally, all predicted key indicators are summarized to form a complete set of prediction results and a report on the future operating status of the enterprise is generated based on the summarized prediction results.

6. The enterprise diagnosis and prediction system based on intelligent analysis according to claim 5 is characterized in that: When judging whether an enterprise has the ability to respond based on the assessment results, first determine the key performance indicators that the enterprise must achieve when responding to risks and set thresholds for each key indicator. These thresholds represent the minimum level required for the enterprise to effectively respond to risks. Then, assign weights to each key indicator based on the degree of impact of the key indicator on the overall risk response capability. i , and then by the formula Calculate the comprehensive score S total Finally, according to the comprehensive score S total And compare it with the set threshold. If the comprehensive score is higher than the set threshold, it means that the company has the ability to cope with it as a whole. If the comprehensive score is lower than the threshold, the company's coping ability is insufficient.

7. The enterprise diagnosis and prediction system based on intelligent analysis according to claim 1 is characterized in that: When generating an enterprise operation diagnostic report to provide management decision support, first collect and summarize the execution capability analysis report, the enterprise's future operation status report, the historical data analysis results, and the enterprise's current status data, and conduct a comprehensive analysis to obtain comprehensive data.

8. The enterprise diagnosis and prediction system based on intelligent analysis according to claim 7, characterized in that: After obtaining comprehensive data, for each key indicator, analyze the difference between execution capability, historical performance, future forecast and current status, and use the formula Calculate and summarize these differences Δ k , identify significant deviation areas, then analyze whether the trend in the comprehensive data is consistent with the current actual operation, and finally compile a comprehensive enterprise operation diagnosis report based on the comprehensive analysis results, difference analysis and trend analysis results and submit the diagnosis report to the enterprise, where Δ k is the comprehensive difference value for key indicator k, C k is the actual value of the key indicator k in the current state, P k is the value of the key indicator k in the predicted value, H k is the value of the key indicator k in historical performance, σ k is the volatility of the key indicator k, w1, w2, w3 are weight factors used to adjust the impact of each difference in the comprehensive difference, and ∈ is a positive number.

9. The enterprise diagnosis and prediction system based on intelligent analysis according to claim 8 is characterized in that: Comprehensive analysis includes comparing execution capability with the current status, comparing historical data with the current status, and comparing predicted results with the current status. When comparing execution capability with the current status, the key indicators in the execution capability analysis report are compared with the data of the company's current status, and whether the actual execution capability has reached the expected or historical level is analyzed to identify current efficiency problems. When comparing historical data with the current status, historical operating data is compared with current operating data to analyze the changing trends of the company's operations, identify continuous improvement or decline patterns, and explore the reasons for these changes. When comparing predicted results with the current status, the predicted data in the company's future operating status report is compared with the current actual data to evaluate whether the company's current operations are in line with the predicted trends and analyze the potential causes of any deviations.

10. An enterprise diagnosis and prediction method based on intelligent analysis, characterized in that: The following steps are involved: Collect the enterprise's operational and historical data and normalize the collected data to remove noise and format it; Analyze the instruction response time, rework frequency and delayed completion time during the enterprise's operation execution process, generate the enterprise's execution capability analysis coefficient, and analyze whether the enterprise's execution capability is qualified based on the execution capability analysis coefficient and generate an execution capability analysis report based on the results; Analyze the completion ratio and time difference of the tasks set in the historical operation process of the enterprise, and generate historical data analysis results; Use the fast sparse ridge regression method to model the enterprise's operating data and train a prediction model for the enterprise's future operating indicators; Use the trained prediction model and the latest data to predict the company's future operating indicators and generate a report on the company's future operating status; Combine the execution capability analysis report, the enterprise's future operating status report, and the historical data analysis results with the enterprise's current status for comparative analysis to generate an enterprise operation diagnosis report to provide management decision support.

Images