Enterprise risk early warning method and system based on dynamic quantification of esg issues

Abstract

This invention belongs to the field of data analysis technology and provides a method and system for enterprise risk early warning based on dynamic quantification of ESG issues. The technical solution involves calculating the impact importance score of each issue, constructing multi-dimensional ESG scenarios, introducing cross-impact analysis to correct the probability of scenario occurrence, analyzing the changes in financial value under each scenario, and finally decomposing and standardizing the financial impact to output the financial importance score of each issue, which is then integrated to obtain a comprehensive importance score. Based on real-time text data related to the target ESG issues of the enterprise to be evaluated, public opinion analysis and issue relevance analysis are performed to generate dynamic adjustment factors and dynamically adjust the comprehensive importance score to obtain an adjusted dynamic comprehensive importance score. Based on the adjusted dynamic comprehensive importance score, all ESG issues are dynamically prioritized, and risk warnings are issued in conjunction with the set hierarchical warning rules. Corresponding risk management strategies are generated based on different risk warnings.

Description

Technical Field

[0001] This invention belongs to the field of data analysis technology, and in particular relates to a method and system for enterprise risk early warning based on dynamic quantification of ESG issues. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] As concerns about climate change, social inclusion, and governance effectiveness continue to deepen, several influential ESG disclosure standards have emerged globally. These standards differ significantly in their core principles, particularly in the definition of "materiality": some standards emphasize a company's impact on the external environment and society, i.e., "impact materiality"; others focus on the risks and opportunities that ESG issues pose to a company's own financial performance, i.e., "financial materiality"; and still others explicitly require coverage of both aspects, i.e., "dual materiality." This divergence leads to inconsistencies in the topics companies choose and the focus of their reports under different frameworks, affecting the comparability and decision-making usefulness of the information.

[0005] However, in practice, due to the relatively mature and widely applied international standards systems of the past, most companies still primarily use frameworks that focus on a single materiality dimension when conducting ESG disclosures. This has resulted in companies accumulating some experience in assessing a particular type of materiality, but lacking unified and operational methodological tools for systematically integrating both dimensions, particularly in translating ESG issues into quantifiable financial language. This uneven capability makes it difficult for companies to fully respond to increasingly stringent, multi-dimensional disclosure requirements and also limits the true integration of ESG information into strategic decision-making and value management.

[0006] For businesses, a scientific assessment of the dual importance of ESG is not only the foundation for compliant disclosure, but also a core management tool for integrating ESG principles into corporate strategy, optimizing resource allocation, managing long-term risks, and seizing transformation opportunities. In fact, ESG risks have become a key variable threatening sustainable business operations: environmental risks may lead to penalties and increased compliance costs, social risks may weaken brand reputation and customer trust, and governance risks may shake the foundation of decision-making. These non-financial risks can all be transmitted to the financial level through supply chain disruptions, increased financing costs, and loss of market share, evolving into a chain reaction of "non-financial risk → financial risk → corporate value impairment."

[0007] Current assessment methods, such as expert review, questionnaires, static financial correlation analysis, and annual assessment mechanisms, generally suffer from limitations such as subjective assessment, ambiguous financial correlations, static data, and siloed systems. These limitations make it difficult to support the dynamic, quantitative, and integrated analysis requirements inherent in the dual importance of ESG issues. Meanwhile, existing risk management systems still suffer from fragmented governance and accountability gaps. ESG risks are often treated as independent compliance tasks, not deeply integrated with core business processes, and lack unified data collection standards and technical tools, resulting in a severe deficiency in the ability to identify, quantify, and provide early warnings of non-financial risks. Summary of the Invention

[0008] To address at least one of the technical problems in the background art, this invention provides a dynamic quantitative assessment method and system for the dual importance of ESG issues. It constructs a comprehensive quantitative assessment system for the dual importance of ESG issues that can objectively reflect external consensus, accurately quantify financial impact, and has dynamic monitoring and early warning functions. This system can accurately identify high-risk issues and assist enterprises in bridging the gap between compliance disclosure and strategic management.

[0009] To achieve the above objectives, the present invention adopts the following technical solution: The first aspect of this invention provides a dynamic quantitative assessment method for the dual importance of ESG issues, comprising the following steps: The impact importance score of the issues was calculated based on the structured questionnaire data and semi-structured interview text data of the ESG issues to be evaluated. By constructing multi-dimensional ESG scenarios, cross-impact analysis is introduced to correct the probability of scenario occurrence, and a parameterized free cash flow discount model is used to analyze the changes in financial value under each scenario. Finally, the financial impact is decomposed and standardized, and the financial importance score of each issue is output. Based on the impact importance score and financial importance score of each issue, a comprehensive importance score is obtained by combining them; Based on real-time text data related to the target ESG issues of the company to be evaluated, public opinion analysis and issue relevance analysis are performed to generate dynamic adjustment factors. The comprehensive importance score is dynamically adjusted based on the dynamic adjustment factors to obtain the adjusted dynamic comprehensive importance score. All ESG issues are dynamically prioritized based on the adjusted dynamic comprehensive importance score, and risk warnings are issued in conjunction with the set tiered warning rules. Corresponding risk management strategies are generated based on different risk warnings.

[0010] Furthermore, the impact importance score of the issue is calculated based on the acquired structured questionnaire data and semi-structured interview text data of the ESG issues to be evaluated, including: The first impact importance score was obtained based on the structured questionnaire data of the ESG issues to be evaluated. Based on the semi-structured interview text data of the ESG issues to be evaluated, the second impact importance score was calculated. The first and second impact importance scores of an issue are combined to obtain the issue's overall impact importance score.

[0011] Furthermore, based on the structured questionnaire data of the ESG issues to be evaluated, a first impact importance score was obtained, including: The strength of positive and negative impacts on the issue is calculated by combining the positive impact assessment criteria table and the negative impact assessment criteria table, and the final individual comprehensive score is obtained; based on the group... k The individual composite scores of all valid respondents within the group are used to calculate the group's response to the issue. i The consensus value; based on the consensus value of each group on the issue and the value allocated to each group. k The weights are used to calculate the importance score of the questionnaire's impact; Based on the semi-structured interview text data of the ESG issues to be evaluated, a second impact importance score was calculated, including: analyzing the semi-structured interview text data to calculate the sentiment tendency value and concern intensity value; and combining the sentiment tendency value and concern intensity value to calculate the impact importance score of the interview text.

[0012] Furthermore, the construction of the multi-dimensional ESG scenarios specifically involves: determining key ESG issues and their status based on industry consensus on their importance; combining different statuses of each ESG issue to obtain a multi-dimensional ESG scenario combination. Key ESG issues include corporate governance compliance, product responsibility achievement, human capital investment, and carbon emission intensity. Corporate governance compliance includes three levels: high, medium, and low; product responsibility achievement includes two levels: excellent and average; human capital investment includes two levels: high and low; and carbon emission intensity includes three levels: low, medium, and high.

[0013] Furthermore, the introduction of cross-influence analysis to correct the probability of scenario occurrence includes: Assign prior probabilities to each state of each issue. ; A cross-influence strength matrix was constructed using expert scoring. The cross-influence strength matrix is ​​transformed according to the set transformation rules. Convert to influence coefficient matrix ; For the influence coefficient matrix After column-wise normalization, matrix N is obtained. Then, the summation of each row of N is performed and normalized again, ultimately yielding the state weight vector representing the relative importance of each state in the interconnected network. ; The prior probabilities of each state Its corresponding state weight vector Multiplying them yields the posterior probability of the state, reflecting the cross-influence between the issues. ; The global posterior probability of each state The conditions for each state are determined by assigning it to the specific scenarios in which it participates, and then the final probability of occurrence of each scenario after cross-influence correction is calculated.

[0014] Furthermore, the analysis of financial value changes under various scenarios using a parametric free cash flow discounted cash flow model includes: Based on the company's financial statements, business plans, and market data at the valuation point in time, the company value is calculated using a DCF model under the assumption of benchmark ESG performance; this value is the benchmark value. ; For each scenario We analyze the defined ESG state combinations and determine the specific adjustment direction and magnitude of each combination on the core input parameters of the DCF model. The scenario The adjusted parameter set is input into the DCF model to calculate the firm value under this scenario. ; Based on enterprise value and benchmark value The change in firm value resulting from this scenario was calculated. .

[0015] Furthermore, the dynamic adjustment factor is: ,

[0016] in, As a dynamic adjustment factor, and For the preset weighting coefficients, satisfy , The score represents the relevance of the topic. For negative emotional intensity, The higher the value, the more negative the public opinion. The average sentiment score is obtained by averaging the sentiment values ​​of each text.

[0017] A second aspect of the present invention provides a dynamic quantitative assessment system for the dual importance of ESG issues, comprising: The impact importance calculation module is used to calculate the impact importance score of an issue based on the structured questionnaire data and semi-structured interview text data of the ESG issues to be evaluated. The financial importance calculation module is used to construct multi-dimensional ESG scenarios, introduce cross-impact analysis to correct the probability of scenario occurrence, and use a parameterized free cash flow discount model to analyze the changes in financial value under each scenario. Finally, it decomposes and standardizes the financial impact and outputs the financial importance score for each issue. The comprehensive importance calculation module is used to integrate the impact importance score and financial importance score of each issue to obtain a comprehensive importance score; The dynamic adjustment module is used to conduct public opinion analysis and issue relevance analysis based on real-time text data related to the target ESG issues of the company to be evaluated, generate dynamic adjustment factors, and dynamically adjust the comprehensive importance score based on the dynamic adjustment factors to obtain the adjusted dynamic comprehensive importance score. The risk warning module is used to dynamically prioritize all ESG issues based on the adjusted dynamic comprehensive importance score, and at the same time, it provides risk warnings in combination with the set hierarchical warning rules, and generates corresponding risk management strategies according to different risk warnings.

[0018] A third aspect of the present invention provides a computer-readable storage medium.

[0019] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the enterprise risk warning method based on dynamic quantification of ESG issues as described above.

[0020] A fourth aspect of the present invention provides a computer device.

[0021] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps in the enterprise risk warning method based on dynamic quantification of ESG issues as described above.

[0022] Compared with the prior art, the beneficial effects of the present invention are: This invention breaks down information silos and static limitations by combining structured questionnaires with in-depth interview text analysis and introducing a balanced factor scoring model. This transforms the subjective opinions of multiple stakeholders into verifiable quantitative scores, achieving objectivity in the assessment of impact importance. By constructing multi-dimensional ESG scenario combinations and correcting scenario probabilities through cross-impact analysis, and relying on a parameterized free cash flow discounted cash flow model, it calculates the expected financial impact of ESG performance on corporate value under each scenario. This achieves unified language and decision-making synergy between ESG management and financial management at the strategic level, enhancing the sensitivity to identify asymmetric major risks.

[0023] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0024] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0025] Figure 1 This is a flowchart of an enterprise risk early warning method based on dynamic quantification of ESG issues provided in an embodiment of the present invention. Detailed Implementation

[0026] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0027] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0028] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0029] Example 1 like Figure 1 As shown, this embodiment provides a method for enterprise risk early warning based on dynamic quantification of ESG issues, including the following steps: Step 1: Calculate the impact importance score based on the structured questionnaire data and semi-structured interview text data of the ESG issues to be evaluated; In this embodiment, data is collected through two independent pathways: structured questionnaires and semi-structured interviews. After being processed separately, the data is weighted and fused to calculate a comprehensive impact importance score. Specifically, it includes: Step 101: Obtain the first impact importance score of the issue based on the structured questionnaire data of the ESG issues to be evaluated; Specifically, the steps include the following: Step 1011: Combine the positive impact judgment criteria table and the negative impact judgment criteria table to calculate the strength of the positive impact and the strength of the negative impact on the issue, and finally obtain the individual comprehensive score; Identify core stakeholder groups of an enterprise (e.g., employees, customers, suppliers, community, investors, etc.), for each ESG issue Structured questionnaires were distributed to representatives widely selected from each group, and the results were assessed based on the positive impact assessment criteria (Table 1) and the negative impact assessment criteria (Table 2). 5 out of 5 rating.

[0030] Table 1 Criteria for Judging Positive Impact

[0031] Table 2 Criteria for Judging Negative Impacts

[0032] For groups Chinese respondents The scores were calculated for each topic. The intensity of positive impact Intensity of negative impact And finally, an individual comprehensive score is obtained. , is represented as: , in, and For groups respectively Chinese respondents The scores were calculated for each topic. The intensity of positive and negative impacts, ,in, , , Respondents On the issue The positive impact is evaluated by multiplying its scale, scope, and likelihood scores. This comprehensive evaluation reflects the potential strength of the positive impact. ,in, , , , Respondents On the issue The scale, scope, likelihood, and irremediability of the negative impact are scored, introducing the concept of irremediability. As a severity amplifier, it ensures that irreversible damage is given higher weight, which aligns with the risk prevention principle. In this embodiment, The value range is from 1 to 125. The value ranges from 1 to 625. Respondents On the issue The overall magnitude of the impact being assessed increases with the intensity of the impact in any direction. As a balance factor, Used to measure the degree of imbalance between positive and negative impacts, when the influence of one direction is significantly dominant, this ratio approaches 1, increasing the denominator and thus making... Relative decrease.

[0033] Step 1012: Based on groups k The individual composite scores of all valid respondents within the group are used to calculate the group's response to the issue. i consensus value , is represented as: , in, Stakeholder groups The number of respondents who provided valid questionnaires.

[0034] Step 1013: Based on the consensus value of each group on the issue, calculate the questionnaire impact importance score of the issue. , is represented as: , in, The total number of stakeholder groups participating in the assessment. To be assigned to a group k The weights satisfy The weighting is open and context-dependent, allowing companies to customize it based on their industry, business model, value chain characteristics, and strategic priorities, thus making the assessment results more aligned with their unique risk exposure and management needs.

[0035] For example, energy-intensive manufacturing enterprises such as chemical and steel companies could consider giving higher weight to "environmental regulatory authorities" and "local communities" because of their large environmental footprint, significant community impact, and compliance risks; financial institutions such as banks and asset management companies could consider giving higher weight to "financial regulatory agencies" and "investors / customers" in response to their high dependence on governance risks, information disclosure, and customer trust.

[0036] Step 102: Based on the semi-structured interview text data of the ESG issues to be evaluated, calculate the second impact importance score of the issues; Specifically, the steps include the following: Step 1021: Analyze the transcribed interview text and calculate the sentiment tendency score and concern intensity score; A systematic natural language processing (NLP) analysis was performed on the transcribed interview texts: the sentiment analysis model snowNLP was used to process each interview text and calculate its overall sentiment tendency score. Its value ranges from 0 to 1. The closer it is to 0, the more negative the emotion; the closer it is to 1, the more positive the emotion.

[0037] The TF-IDF algorithm was used to automatically extract high-frequency and key words from the interview text. Combining the algorithm results with human interpretation, scattered words were summarized and clustered into several core concerns. For each identified core concern, a concern intensity score of 1 to 5 was assigned based on its frequency of mention in the interview, the intensity of the accompanying emotion, and the specificity and richness of detail in the narrative. .

[0038] Step 1022: Calculate the importance score of the interview text's impact on the topic by combining the sentiment tendency score and the concern intensity score, expressed as: , in, The average sentiment score for all interview texts. This is converted into an average negative emotional intensity, mapped to a value between 0 and 1, with a higher value indicating a more negative overall emotion. The average intensity score for all identified core concerns, ranging from 1 to 5.

[0039] Step 103: Combine the first and second impact importance scores of the issue to obtain the final impact importance score. ; In this embodiment, the breadth of information reflected in the comprehensive structured questionnaire is considered. In-depth insights unearthed through in-depth interviews The quantification results of the two independent paths are then combined to calculate the final impact importance score. , is represented as: ， in, It is an issue The final impact importance score, The score is calculated based on a structured questionnaire. The score is calculated based on in-depth interview analysis. and These are the combined weights for the questionnaire path and the interview path, respectively, and they satisfy... The specific value can be set according to the actual situation in this field. In this embodiment, the arithmetic mean method is used to fuse the two scores, that is, the questionnaire influence importance score ( ) and the importance score of the interview ( Assign equal weights = =0.5). This is the maximum reference value after the fusion of dual-source data, used to standardize the fusion result to a score range of 0-10. Its value can be set according to the actual scenario, specifically in two scenarios: when the enterprise has no historical operational data, the theoretical maximum score of 50 is taken (i.e., the fusion result reaches the theoretical maximum score of 50 for each dimension); when the enterprise has historical operational data, the actual maximum value of the dual-source fusion score over the past 3-5 years can be taken. The specific value can be set according to the actual situation in this field. In this embodiment, the theoretical maximum value is used. =50 is used for standardized calculation.

[0040] This invention abandons the traditional model that relies solely on qualitative judgments from internal experts. Instead, it designs a systematic and standardized stakeholder review process, using external consensus as the core basis for determining the materiality of an impact. Methodologically, it employs a dual-path approach of structured questionnaires and semi-structured interviews, and introduces a quantitative scoring model with balancing factors to transform subjective opinions into comparable and verifiable quantitative data. In application, this mechanism significantly enhances the objectivity and credibility of the evaluation results, effectively resisting "greenwashing" accusations from the outset and providing methodological support for companies to comply with stringent domestic and international disclosure requirements.

[0041] Step 2: By constructing multi-dimensional ESG scenarios, cross-impact analysis is introduced to correct the probability of scenario occurrence. A parametric free cash flow discounted cash flow model is then used to analyze the changes in financial value under each scenario. Finally, the monetization impact is decomposed and standardized, and a financial importance score for each issue is output. .

[0042] Specifically, the steps include the following: Step 201: Identify key ESG issues and their status; Based on industry consensus on their importance, core ESG issues are identified and defined. Each issue is divided into several discrete states, serving as the basis for subsequent quantitative analysis. In this embodiment, four key ESG issues and their states are taken as examples: Corporate governance compliance: This refers to the degree of perfection of an enterprise in terms of governance structure, internal control, information disclosure, and compliant operation, and is divided into three levels: high, medium, and low.

[0043] Product responsibility fulfillment: This reflects a company's performance level in areas such as product safety, environmental impact throughout the product lifecycle, energy-saving technology innovation, and protection of customer rights, and is divided into two levels: excellent and average.

[0044] Human capital investment: This measures the intensity of a company's resource investment in areas such as employee development, welfare benefits, innovation incentives, and team stability, and is divided into two levels: high and low.

[0045] Carbon emission intensity refers to the level of greenhouse gas emissions per unit of economic output of an enterprise. As a key quantitative indicator of environmental performance, it is divided into three levels: low, medium, and high.

[0046] Step 202: Combine the different states of each issue to construct a multi-dimensional ESG scenario combination; By combining the different states of the above issues, a complete set of scenarios describing a company's potential future ESG performance is formed. To focus on the most representative development paths, three representative typical scenarios are summarized, corresponding to leading (high scenario), benchmark (medium scenario), and lagging (low scenario) states of overall ESG performance, respectively. Examples of core state combinations for each scenario are shown in Table 3: Table 3 Examples of ESG Multidimensional Scenario Construction

[0047] Step 203: Set prior probabilities for each state of each issue; Based on industry benchmarks, strategic commitments, and historical data, prior probabilities are set for each state of each issue. Under the initial assumption that each issue is independent, the initial probability of any scenario can be obtained by multiplying the prior probabilities of each state it contains.

[0048] Specifically, industry benchmarks: refer to the industry importance issue matrix and score distribution of mainstream ESG rating agencies (such as MSCI and Sustainalytics). For example, if 60% of companies in a certain industry receive a high rating (such as A or above) on the "product responsibility" issue, this can be used as a benchmark reference value (0.60) for the prior probability of "excellent product responsibility".

[0049] Strategic resource allocation dimension: Analyze the company's publicly disclosed strategic plans and budget allocations. For example, if a company explicitly allocates 30% of its annual R&D expenditure to green product development, this can be used as financial evidence to support the probability of achieving "excellent product responsibility" and "low carbon intensity."

[0050] Historical performance trend dimension: Based on the company's ESG key performance indicators (KPIs) performance and trends in recent years. For example, if carbon emission intensity has decreased by an average of 5% annually over the past three years, and the target is to continue to decrease, then the probability setting of "low carbon emission intensity" can be supported.

[0051] The final ESG issue states and their prior probability settings are shown in Table 4: Table 4 Examples of ESG Issue States and Their Prior Probabilities

[0052] Step 204: Introduce cross-influence analysis to correct the occurrence probability of each scenario and obtain the final occurrence probability of each scenario after cross-influence correction; To characterize the real-world connections between ESG issues and revise the independence assumption, a cross-influence strength matrix was constructed using expert scoring. Convert it into an influence coefficient matrix Then, the posterior weights of each state are calculated. This is used to correct prior probabilities, resulting in probabilities that more closely resemble the actual occurrence of scenarios. .

[0053] Constructing the cross-influence intensity matrix A cross-disciplinary expert group was formed, comprising scholars and senior practitioners in asset valuation, auditing and assurance, ESG research, and industry analysis. Through the Delphi method or multiple rounds of expert interviews, the group systematically collected expert judgments on the interactions between various ESG states. To ensure consistency and comparability of the assessments, clear assessment guidelines were provided to the experts, including definitions of each ESG issue state and a unified grading standard for the degree of impact (0 = no impact, ±2 = minor impact, ±4 = strong impact, ±6 = very strong impact; positive values ​​indicate promoting effects, negative values ​​indicate inhibiting effects). Based on this standard, experts independently assessed each group of "states". →Status The influence relationships were scored. The scores from all experts were then averaged to reduce individual subjective bias, ultimately forming a quantitative cross-influence strength matrix. Matrix elements Representing state State The intensity of the impact.

[0054] Calculate the state weight vector Based on the transformation rules, the cross-influence strength matrix will be... Convert to influence coefficient matrix The conversion rule is: if ,but ;like ,but For the matrix Column-wise normalization yields matrix N. Then, the summation and normalization of each row of N are performed to obtain the state weight vector representing the relative importance of each state in the interconnected network.

[0055] Calculate the posterior probability of the state : The prior probabilities of each state Its corresponding state weight vector Multiplying them together yields the posterior probability of the state, which reflects the cross-influence between the issues.

[0056] Calculate the scenario correction probability : The global posterior probability of each state The conditions for each state are determined by assigning it to the specific scenarios in which it participates, and then the final probability of occurrence of each scenario after cross-influence correction is calculated.

[0057] Step 205: Analyze the changes in financial value under each scenario using a parametric free cash flow discounted cash flow model; To systematically and reproducibly translate ESG performance into financial value impact, this paper uses the discounted free cash flow (DCF) model from corporate valuation theory as the financial quantification basis. It establishes a dual transmission mechanism for the impact of ESG issues on DCF model parameters and adopts the DCF model as the basis for corporate value. The assessment framework influences key parameters in the DCF model through two interconnected paths, thereby altering firm value. The first path directly impacts expected future cash flows, with core parameters including revenue growth rate, operating profit margin, effective tax rate, capital expenditures, and working capital requirements. The second path indirectly affects the discount rate through the risk premium, with the weighted average cost of capital (WACC) as the core parameter. The logic behind the first path includes: 1. Revenue growth: Superior product safety and data governance enhance brand loyalty, increase market share and pricing power, thus raising the long-term revenue growth rate assumption. 2. Cost savings: Effective energy conservation, emission reduction, and resource recycling directly reduce operating costs, thereby improving operating profit margin. 3. Capital efficiency: While upfront investment in green technologies may increase short-term capital expenditures, it reduces operating costs in the long term and avoids future "brown asset" stranded losses. 4. Fines, litigation compensation, or supply chain disruptions resulting from environmental incidents will directly cause cash outflows or revenue reductions, requiring corresponding adjustments to cash flow forecasts. The logic behind the second transmission path includes: 1. Poor ESG performance significantly increases a company's regulatory, litigation, reputational, and supply chain risks. To compensate for these additional risks, investors and creditors will demand higher returns, leading to increased equity and / or debt costs, ultimately pushing up WACC. 2. Excellent and robust ESG performance is seen as a signal of a company's stronger risk management capabilities and long-term resilience, which may help reduce the cost of capital and thus increase present value.

[0058] As a further implementation method, the parametric free cash flow discounted cash flow model is used to analyze changes in financial value under various scenarios, including: Determine the benchmark enterprise value Based on the company's financial statements, business plans, and market data at the valuation point in time, the discounted free cash flow (DCF) model is used to calculate the company value under benchmark ESG performance assumptions (usually current performance or industry average). The resulting company value is the benchmark value. This value serves as a benchmark for measuring the value changes brought about by each scenario.

[0059] Contextualized financial parameter adjustment: tailored to each scenario The analysis examines the defined ESG state combinations and, based on the dual transmission mechanism, determines the specific adjustment direction and magnitude of each combination on the core input parameters of the DCF model. The adjustment magnitude can be quantitatively set based on internal financial analysis, industry benchmarking studies, academic empirical conclusions, or expert experience. For each scenario... Generate a set of adjusted DCF model input parameters, including but not limited to revenue growth rate, operating profit margin, tax rate, capital expenditure, working capital ratio, and weighted average cost of capital (WACC).

[0060] Calculate the change in firm value under each scenario : Scenario The adjusted parameter set is input into the DCF model to calculate the firm value under this scenario. Calculate the change in firm value resulting from this scenario. Therefore, when This indicates that the ESG performance represented by this scenario is expected to create additional financial value (opportunity) for the company; when This indicates that the ESG performance represented by this scenario is expected to impair the company's financial value (risk).

[0061] Step 206: Decompose the scenario financial value impact into individual ESG issues and calculate the financial value impact for each ESG issue. Expected financial impact; Using a factor decomposition method based on contribution weights, the changes in firm value are analyzed. Scientifically attribute the causes to individual ESG issues in order to assess the independent financial impact of each issue.

[0062] Calculate the contribution weight of factors within the scenario. For the scenario ESG issues in China Its contribution weight is determined by the strength of the factor’s direct financial impact and its cross-effect driving force within the scenario.

[0063] , in, This indicates the target ESG issue for which the contribution weight is to be calculated. To indicate a specific situation. Indicating ESG issues The pre-defined direct financial impact strength (e.g., a benchmark value of 1-10 points assigned by an expert scoring method) reflects its ability to individually affect the value of the enterprise. This indicates the cross-influence coefficient matrix. ESG issues extracted For the same situation Any other issue within The influence coefficient.

[0064] Allocate value and calculate the total expected financial impact of individual issues. The changes in scenario value are allocated to each issue according to their contribution weights, and a probability-weighted sum is performed on all scenarios to obtain the value of each ESG issue. Expected financial impact: , Where S represents the total number of scenarios. A positive value indicates that the issue creates financial value overall, while a negative value indicates that it poses financial risk.

[0065] Step 207: Decompose and standardize the financial impact to generate financial importance scores for individual issues. ; To facilitate cross-issue comparisons and decision ranking, the expected financial impact of monetization is standardized as a financial importance score of 0-10.

[0066] Standardization process: Selecting enterprise value as... Financial Size Benchmark Calculate the relative impact ratio of each issue. : , Nonlinear mapping: using a sigmoid function to... Mapped to a score of 0-10, the simulation reflects the non-linear characteristics of importance perception in management decisions. When the impact is small, the score increases slowly, but the sensitivity increases sharply after exceeding a certain threshold.

[0067] , in, To control the sensitivity parameter of the curve steepness, The threshold parameter is used to define the importance center.

[0068] This invention addresses the challenge of quantifying financial importance. By constructing multi-dimensional ESG scenario combinations, adjusting scenario probabilities through cross-influence analysis, and utilizing a parameterized free cash flow discounted cash flow model, it calculates the expected financial impact of ESG performance on firm value under each scenario, thereby transforming ESG risks and opportunities into tangible monetary value. The output can be directly integrated into core corporate financial processes such as return on investment analysis, capital budgeting, and valuation modeling, truly achieving unified language and collaborative decision-making between ESG management and financial management at the strategic level.

[0069] Step 3: Based on the impact importance score and financial importance score of each issue. The overall importance score is obtained by integrating the results. In this embodiment, the external impact is quantified using an impact importance score for each issue, and the internal financial impact is quantified based on a financial importance score. These two dimensions are then scientifically integrated to reflect the overall importance of the issue. Specifically, the basic comprehensive importance score is calculated using the distance-to-origin method, and the formula is as follows: , Each ESG issue can be viewed as a point on a two-dimensional plane, with its x-coordinate being... The vertical axis is , which is the Euclidean distance from the point to the origin (0,0).

[0070] This invention abandons the weighted average method, which easily masks extreme values, and innovatively introduces the distance-to-origin method for two-dimensional integration. Mathematically, this method has a stronger responsiveness to high scores in a single dimension. In application, it can effectively identify "asymmetric major risks" that excel in one dimension ("financial" or "impact") but are average in the other, such as sudden environmental accidents or potentially huge lawsuits. This ensures that such high-risk issues are not diluted in the ranking process, thus more accurately reflecting management priorities and emergency needs.

[0071] Step 4: Based on the real-time text data related to the target ESG issues of the company to be evaluated, conduct public opinion analysis and issue relevance analysis based on the real-time text data, generate dynamic adjustment factors, and dynamically adjust the comprehensive importance score based on the dynamic adjustment factors to obtain the adjusted dynamic comprehensive importance score. To enable the evaluation system to respond in real time to changes in the external environment, natural language processing (NLP)-based public opinion dynamic monitoring is introduced. By integrating web crawling, sentiment analysis, and topic modeling techniques, a dynamic adjustment factor (λ) is generated to adjust the importance of the basic comprehensive assessment in real time.

[0072] Specifically, the steps include the following: Step 401: Based on real-time text data related to the target ESG issues of the company to be evaluated, collected in real time; In this embodiment, data is acquired through legitimate means, including self-generated data, publicly available data, and data obtained through protocols. For example, keyword configuration uses the "company name" combined with specific ESG topic keywords (such as "water resources," "sewage discharge," and "carbon emissions") and their synonyms as crawling seeds. Incremental crawling is achieved through scheduled tasks to ensure continuous data updates and timeliness. The system acquires raw text data related to the enterprise and target ESG topics.

[0073] Step 402: Conduct public opinion analysis and topic relevance analysis based on real-time text data to generate dynamic adjustment factors; After preprocessing the collected texts (cleaning, word segmentation), sentiment analysis tools such as SnowNLP are used to calculate the sentiment tendency score for each text (ranging from 0 to 1, with values ​​closer to 0 indicating a more negative sentiment). Within a set time window (e.g., the past 30 days), the sentiment scores of all texts are weighted and averaged (weights can be set according to the authority of the source), and then converted into a negative sentiment intensity score. : , in, The higher the value, the more negative the public opinion. The average sentiment score is obtained by averaging the sentiment values ​​of each text.

[0074] To eliminate irrelevant noise, the LDA topic model is used to automatically identify topics from the text collection. The core discussion topics were identified. These topics were matched with target ESG issues through manual annotation or semantic similarity calculation to filter out relevant topics. The TF-IDF algorithm was then used to extract high-weighted feature words from the relevant topic texts, and the relevance score was calculated by combining the proportion of relevant topic documents with keyword overlap. (between 0 and 1).

[0075] Integrating negative emotional intensity Relevance Generate dynamic adjustment factors : , in, and For the preset weighting coefficients, satisfy (like =0.5, =0.5). When public opinion is highly negative and highly relevant, A significant increase in the value indicates a strong early warning; when public opinion is negative but low in correlation, The increase should be limited, and interference from irrelevant events should be avoided; when public opinion is positive... The value approaches 0. It also approaches 0, and the evaluation results remain stable.

[0076] Step 403: Dynamically adjust the overall importance score based on the dynamic adjustment factor to obtain the adjusted dynamic overall importance score; Dynamic adjustment factor Applied to basic comprehensive importance score This will give you a dynamically updated overall importance score. , is represented as: = (1+ ).

[0077] Step 5: Based on the adjusted dynamic comprehensive importance score, dynamically prioritize all ESG issues, and at the same time, conduct risk warnings in conjunction with the set hierarchical warning rules, and generate corresponding risk management strategies according to different risk warnings; Specifically, it includes: Step 501: Use the overall importance score Based on this, all ESG issues are ranked in real time; the ranking results are automatically adjusted according to public opinion dynamics and data updates, taking into account both the balance of "dual importance" and giving weight to asymmetric risk issues with extreme prominence in a single dimension, ensuring that high-risk issues are always at the core of management vision.

[0078] Step 502: Combine the enterprise's industry characteristics and risk tolerance to customize the threshold value, and analyze the risk level in combination with the preset three-level early warning threshold rules; In this embodiment, the custom threshold value, which combines the enterprise's industry characteristics and risk tolerance, specifically includes: This will affect the importance score. Financial Importance Score All scores are standardized to 0-10, and the basic comprehensive importance score is calculated using Euclidean distance. The maximum value is approximately Points; superimposed dynamic adjustment factors (The maximum value is 0.5, corresponding to extreme negative public opinion scenarios.) After correction, the dynamic comprehensive importance score is calculated. The theoretical maximum score is approximately 14.14 × (1 + 0.5) ≈ 21.21 points.

[0079] like A score of 1 indicates a Level 1 warning, which represents a serious risk. This risk corresponds to a situation where both aspects are of high importance or one aspect is extremely prominent, potentially leading to significant financial losses, compliance penalties, or brand crises. like This is a Level 2 warning, representing a general risk with moderate to high dual importance. It may have a phased impact and requires specific optimization of management processes. like A score of 8 indicates a Level 3 warning, which alerts to risks that have a limited scope of impact, a low level of risk, and can be addressed through routine management and control measures. Step 503: Based on the warning level and issue attributes, automatically match the preset control strategy library and output an actionable plan: If it is a Level 1 warning (severe risk): activate the emergency response mechanism, such as suspending high-risk business processes, allocating special resources for rectification, establishing a special task force for communication with stakeholders, and simultaneously developing a loss hedging plan; If it is a Level 2 warning (general risk): Implement process optimization and special control, such as revising internal management system, strengthening supplier ESG audits, conducting special training, setting phased improvement goals and tracking their implementation; If it is a Level 3 warning (risk alert): Maintain routine management and dynamic monitoring, include the issue in the annual ESG assessment list, and prevent the risk from escalating through regular data collection and trend analysis.

[0080] The strategy library allows enterprises to customize and supplement it according to industry characteristics, improving adaptability and practicality.

[0081] Example 2 This embodiment provides an enterprise risk early warning system based on dynamic quantification of ESG issues, including: The impact importance calculation module is used to calculate the impact importance score of an issue based on the structured questionnaire data and semi-structured interview text data of the ESG issues to be evaluated. The financial importance calculation module is used to construct multi-dimensional ESG scenarios, introduce cross-impact analysis to correct the probability of scenario occurrence, and use a parameterized free cash flow discount model to analyze the changes in financial value under each scenario. Finally, it decomposes and standardizes the financial impact and outputs the financial importance score for each issue. The comprehensive importance calculation module is used to integrate the impact importance score and financial importance score of each issue to obtain a comprehensive importance score; The dynamic adjustment module is used to conduct public opinion analysis and issue relevance analysis based on real-time text data related to the target ESG issues of the company to be evaluated, generate dynamic adjustment factors, and dynamically adjust the comprehensive importance score based on the dynamic adjustment factors to obtain the adjusted dynamic comprehensive importance score. The risk warning module is used to dynamically prioritize all ESG issues based on the adjusted dynamic comprehensive importance score, and at the same time, it provides risk warnings in combination with the set hierarchical warning rules, and generates corresponding risk management strategies according to different risk warnings.

[0082] It should be noted that the specific implementation of the enterprise risk warning system based on dynamic quantification of ESG issues in this embodiment of the invention is similar to the specific implementation of the enterprise risk warning method based on dynamic quantification of ESG issues in this embodiment of the invention. For details, please refer to the description in the method section. To reduce redundancy, it will not be repeated here.

[0083] Example 3 This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in the enterprise risk warning method based on dynamic quantification of ESG issues as described above.

[0084] Example 4 This embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps in the enterprise risk warning method based on dynamic quantification of ESG issues as described above.

[0085] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of hardware embodiments, software embodiments, or embodiments combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0086] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0087] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0088] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1The steps of the function specified in one or more boxes.

[0089] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0090] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for enterprise risk early warning based on dynamic quantification of ESG issues, characterized in that, Includes the following steps: The impact importance score of the issues was calculated based on the structured questionnaire data and semi-structured interview text data of the ESG issues to be evaluated. By constructing multi-dimensional ESG scenarios, cross-impact analysis is introduced to correct the probability of scenario occurrence, and a parameterized free cash flow discount model is used to analyze the changes in financial value under each scenario. Finally, the financial impact is decomposed and standardized, and the financial importance score of each issue is output. Based on the impact importance score and financial importance score of each issue, a comprehensive importance score is obtained by combining them; Based on real-time text data related to the target ESG issues of the company to be evaluated, public opinion analysis and issue relevance analysis are performed to generate dynamic adjustment factors. The comprehensive importance score is dynamically adjusted based on the dynamic adjustment factors to obtain the adjusted dynamic comprehensive importance score. All ESG issues are dynamically prioritized based on the adjusted dynamic comprehensive importance score, and risk warnings are issued in conjunction with the set tiered warning rules. Corresponding risk management strategies are generated based on different risk warnings.

2. The enterprise risk early warning method based on dynamic quantification of ESG issues as described in claim 1, characterized in that, The impact importance score of the issues is calculated based on the structured questionnaire data and semi-structured interview text data of the ESG issues to be evaluated, including: The first impact importance score was obtained based on the structured questionnaire data of the ESG issues to be evaluated. Based on the semi-structured interview text data of the ESG issues to be evaluated, the second impact importance score was calculated. The first and second impact importance scores of an issue are combined to obtain the issue's overall impact importance score.

3. The enterprise risk early warning method based on dynamic quantification of ESG issues as described in claim 2, characterized in that, The first impact importance score was obtained based on the structured questionnaire data of the ESG issues to be evaluated, including: The strength of positive and negative impacts on the issue is calculated by combining the positive impact assessment criteria table and the negative impact assessment criteria table, and the final individual comprehensive score is obtained; based on the group... k The individual composite scores of all valid respondents within the group are used to calculate the group's response to the issue. i The consensus value; based on the consensus value of each group on the issue and the value allocated to each group. k The weights are used to calculate the importance score of the questionnaire's impact; Based on the semi-structured interview text data of the ESG issues to be evaluated, a second impact importance score was calculated, including: analyzing the semi-structured interview text data to calculate the sentiment tendency value and concern intensity value; and combining the sentiment tendency value and concern intensity value to calculate the impact importance score of the interview text.

4. The enterprise risk early warning method based on dynamic quantification of ESG issues as described in claim 1, characterized in that, The construction of the multi-dimensional ESG scenarios is specifically as follows: based on the consensus on the importance of the industry, key ESG issues and their status are identified; different statuses of each ESG issue are combined to obtain a multi-dimensional ESG scenario combination. Key ESG issues include corporate governance compliance, product responsibility achievement, human capital investment, and carbon emission intensity. Corporate governance compliance is divided into three levels: high, medium, and low. Product responsibility achievement is divided into two levels: excellent and average. Human capital investment is divided into two levels: high and low. Carbon emission intensity is divided into three levels: low, medium, and high.

5. The enterprise risk early warning method based on dynamic quantification of ESG issues as described in claim 1, characterized in that, The introduction of cross-influence analysis to correct the probability of scenario occurrence includes: Assign prior probabilities to each state of each issue. ; A cross-influence strength matrix was constructed using expert scoring. The cross-influence strength matrix will be transformed according to the set transformation rules. Convert to influence coefficient matrix ; For the influence coefficient matrix After column-wise normalization, matrix N is obtained. Then, the summation of each row of N is performed and normalized again, ultimately yielding the state weight vector representing the relative importance of each state in the interconnected network. ; The prior probabilities of each state Its corresponding state weight vector Multiplying them yields the posterior probability of the state, reflecting the cross-influence between the issues. ; The global posterior probability of each state The conditions for each state are determined by assigning it to the specific scenarios in which it participates, and then the final probability of occurrence of each scenario after cross-influence correction is calculated.

6. The enterprise risk early warning method based on dynamic quantification of ESG issues as described in claim 1, characterized in that, The analysis of financial value changes under various scenarios using a parametric free cash flow discounted cash flow model includes: Based on the company's financial statements, business plans, and market data at the valuation point in time, the company value is calculated using a DCF model under the assumption of benchmark ESG performance; this value is the benchmark value. ; For each scenario We analyze the defined ESG state combinations and determine the specific adjustment direction and magnitude of each combination on the core input parameters of the DCF model. The scenario The adjusted parameter set is input into the DCF model to calculate the firm value under this scenario. ; Based on enterprise value and benchmark value The change in firm value resulting from this scenario was calculated. .

7. The enterprise risk early warning method based on dynamic quantification of ESG issues as described in claim 1, characterized in that, The dynamic adjustment factor is: ， in, As a dynamic adjustment factor, and For the preset weighting coefficients, satisfy , The score represents the relevance of the topic. For negative emotional intensity, The higher the value, the more negative the public opinion. The average sentiment score is obtained by averaging the sentiment values ​​of each text.

8. A corporate risk early warning system based on dynamic quantification of ESG issues, characterized in that: include: The impact importance calculation module is used to calculate the impact importance score of an issue based on the structured questionnaire data and semi-structured interview text data of the ESG issues to be evaluated. The financial importance calculation module is used to construct multi-dimensional ESG scenarios, introduce cross-impact analysis to correct the probability of scenario occurrence, and use a parameterized free cash flow discount model to analyze the changes in financial value under each scenario. Finally, it decomposes and standardizes the financial impact and outputs the financial importance score for each issue. The comprehensive importance calculation module is used to integrate the impact importance score and financial importance score of each issue to obtain a comprehensive importance score; The dynamic adjustment module is used to conduct public opinion analysis and issue relevance analysis based on real-time text data related to the target ESG issues of the company to be evaluated, generate dynamic adjustment factors, and dynamically adjust the comprehensive importance score based on the dynamic adjustment factors to obtain the adjusted dynamic comprehensive importance score. The risk warning module is used to dynamically prioritize all ESG issues based on the adjusted dynamic comprehensive importance score, and at the same time, it provides risk warnings in combination with the set hierarchical warning rules, and generates corresponding risk management strategies according to different risk warnings.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps in the enterprise risk warning method based on dynamic quantification of ESG issues as described in any one of claims 1-7.

10. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the enterprise risk warning method based on dynamic quantification of ESG issues as described in any one of claims 1-7.

Images