A method and system for integrated circuit design project risk assessment

Abstract

The present application relates to the technical field of integrated circuit design project risk assessment, and discloses an integrated circuit design project risk assessment method and system, wherein the exchange information between the design personnel and the supplier personnel is acquired, and the discussion content related to the material parameters and the potential risk indication are identified from the exchange information, so that the problem that the key information scattered in the unstructured communication record cannot be captured and understood in the prior art is effectively solved.

Description

Technical Field

[0001] This invention relates to the technical field of risk assessment for integrated circuit design projects, and more specifically, to a method and system for risk assessment of integrated circuit design projects. Background Technology

[0002] In the field of integrated circuit design, as project complexity continues to increase, various risk factors such as technical defects, supply chain fluctuations, and team collaboration barriers often intertwine, posing a severe challenge to project success.

[0003] However, existing risk assessment systems often struggle to conduct effective correlation analysis when dealing with these complex and multi-dimensional risks, leading to delays in risk identification and response measures. Traditional risk assessment methods rely heavily on empirical qualitative judgments, making it particularly difficult to quantify the transmission path of risks and their potential impact range. This limits project managers' comprehensive understanding of the true risk situation and their ability to intervene in a timely manner. Summary of the Invention

[0004] This invention provides a method and system for risk assessment of integrated circuit design projects, aiming to solve the problems that existing risk assessment methods for integrated circuit design projects are unable to conduct effective correlation analysis when dealing with complex and multi-dimensional risks, resulting in a lag in risk identification and response measures, and are unable to effectively capture and understand key information scattered in unstructured communication records, thus failing to accurately assess the true risk situation.

[0005] The technical solution of this application is as follows:

[0006] In a first aspect, this application discloses a risk assessment method for integrated circuit design projects, the method comprising:

[0007] Obtain communication information between designers and suppliers, and identify discussions related to material parameters and indications of potential risks from this communication information;

[0008] Obtain the design entity data output by the design verification tool, as well as the supplier's material parameter data;

[0009] By performing correlation analysis on the identified discussion content, design entity data, and supplier material parameter data, risk clues can be obtained.

[0010] Continuously analyze risk clues, monitor the correlation between changes in design entity data and supplier material parameter data, and propose causal hypotheses between changes in material properties and design performance issues based on the correlation.

[0011] Collect evidence supporting the causal hypothesis, including original communication records, supplier material parameter data, and design verification data, and perform parameter impact assessments to quantify the impact of material parameter variations on design performance.

[0012] The collected evidence and parameters influence the assessment results, evaluate the credibility of the causal hypothesis, and confirm the causal hypothesis as a potential risk based on the credibility.

[0013] Based on the confirmed causal assumptions and collected evidence, a risk narrative report is generated. This report includes the evolution of the risk, the chain of evidence supporting the causal assumptions, and provides index information for tracing and verifying the chain of evidence.

[0014] This technical solution effectively integrates unstructured communication information, design entity data, and supplier material parameter data. It identifies potential risk clues through correlation analysis and further quantifies the impact of material parameter changes on design performance through causal assumptions, evidence collection, and parameter impact assessment. Finally, it generates a risk narrative report that includes the risk evolution process and evidence chain, thereby solving the problem of insufficient risk identification and correlation analysis capabilities in existing technologies and improving the accuracy and traceability of risk assessment.

[0015] Furthermore, in some implementation schemes, monitoring the correlation between changes in design entity data and supplier material parameter data includes:

[0016] Obtain performance parameter values ​​from design entity data at multiple time points to form a time series sequence of performance parameters;

[0017] Obtain the material property values ​​from the supplier's material parameter data corresponding to each time point to form a time series of material properties;

[0018] When the time series of performance parameters and the time series of material properties show the same trend, and the time series of material properties changes earlier than the time series of performance parameters, a causal hypothesis is proposed that the change in the supplier's material parameters leads to the design performance problem.

[0019] This technical solution enables more accurate identification of the causal relationship between changes in material properties and design performance issues through time-series analysis, providing more reliable causal assumptions for subsequent risk assessment.

[0020] More specifically, in some implementation schemes, a parameter impact assessment is performed to quantify the impact of changes in material parameters on design performance, including:

[0021] Obtain the range of parameter variations for materials provided by the supplier across multiple batches;

[0022] The range of parameter variation is divided into multiple discrete parameter value points;

[0023] For each parameter value, the design verification tool is called to perform design performance simulation to obtain the design performance simulation value corresponding to each parameter value.

[0024] Based on the design performance simulation values ​​of all parameter points, establish a correspondence table between material parameters and design performance;

[0025] Based on the correspondence table, calculate the change in design performance corresponding to the unit change in material parameters, and use this as the influence value.

[0026] This technical solution enables accurate assessment of the specific impact of material parameter variations on design performance through simulation and quantitative analysis, providing quantitative data support for risk assessment.

[0027] Preferably, in some implementations, assessing the credibility of the causal hypothesis includes:

[0028] Obtain the number of pieces of evidence supporting the causal hypothesis, and determine the strength of the association between each piece of evidence and the causal hypothesis;

[0029] Obtain the confidence interval of the impact value obtained from the parameter impact assessment;

[0030] Calculate the confidence score of the causal hypothesis based on the amount of evidence, the level of correlation strength, and the confidence interval.

[0031] This technical solution can comprehensively consider multiple factors to quantitatively assess the credibility of causal hypotheses, thereby improving the objectivity and accuracy of risk assessment.

[0032] Building upon this, this application further proposes methods for confirming causal hypotheses as potential risks based on credibility, including:

[0033] Obtain a pre-generated risk assessment rule base, which contains multiple risk levels and their corresponding credibility score ranges;

[0034] The calculated credibility score is matched with the credibility score range in the risk assessment rule base to determine the risk level corresponding to the causal hypothesis;

[0035] When the risk level exceeds the preset risk level threshold, the causal hypothesis is confirmed as a potential risk.

[0036] This technical solution enables the automatic determination of whether causal assumptions constitute potential risks based on preset risk assessment rules, thereby improving the efficiency and consistency of risk identification.

[0037] Building upon the above, this application further proposes generating a risk narrative report, including:

[0038] Extract key events in the risk evolution process in chronological order to generate a risk evolution timeline;

[0039] The collected evidence is ordered according to its strength of association with the causal hypothesis to form a chain of evidence;

[0040] Generate a traceable storage index identifier for each piece of evidence in the chain of evidence;

[0041] Integrate the risk evolution timeline, evidence chain, and storage index identifiers into a risk narrative report.

[0042] This technical solution can generate structured and traceable risk narrative reports, clearly showing the evolution of risks and supporting evidence, making it easier for project managers to fully understand and trace risks.

[0043] Furthermore, in some implementations, index information for tracing and verifying the chain of evidence is provided, including:

[0044] Generate metadata tags for each piece of evidence in the chain of evidence, including storage path, creation timestamp, and data source identifier;

[0045] Associate metadata tags with corresponding evidence and store them in the database;

[0046] Embed link elements for metadata tags in risk narrative reports;

[0047] In response to a link element being triggered, retrieve and display the corresponding evidence content from the database.

[0048] This technical solution enables convenient tracing and verification of evidence in risk narrative reports, enhancing the transparency and credibility of risk assessment results.

[0049] In one implementation, identifying discussions related to material parameters and indications of potential risks from communication information includes:

[0050] Retrieve the text content and timestamp information from the exchanged messages;

[0051] Keyword matching is performed on the text content to identify keywords related to material parameters and the corresponding discussion time.

[0052] Based on the discussion time, the keywords are arranged in chronological order to form a timeline of the discussion of material parameters;

[0053] Identify periods of sudden increases in discussion frequency from the discussion timeline as indicators of potential risks.

[0054] This technical solution effectively identifies discussion content and potential risk signals related to material parameters from unstructured communication information through keyword matching and timeline analysis, solving the problem that traditional methods struggle to handle unstructured data.

[0055] In another implementation, correlation analysis is performed on the identified discussion content, design entity data, and supplier material parameters, including:

[0056] Extract material parameter keywords from the identified discussion content;

[0057] Extract performance parameter values ​​corresponding to material parameter keywords from the design entity data;

[0058] Extract material property values ​​corresponding to material parameter keywords from the supplier's material parameter data;

[0059] Combinations where the correlation between material property values ​​and performance parameter values ​​exceeds a preset threshold are marked as risk clues.

[0060] This technical solution enables the effective correlation of data from different sources, identifies potential risk clues, and lays the foundation for subsequent risk analysis.

[0061] Secondly, this application also discloses an integrated circuit design project risk assessment system, which includes:

[0062] The data acquisition module is used to connect to the designer's communication terminal, the design verification tool terminal, and the supplier's data interface to obtain communication information, design entity data, and supplier material parameter data.

[0063] The data storage module, connected to the data acquisition module, is used to store communication information, design entity data, and supplier material parameter data;

[0064] The risk identification module, connected to the data storage module, is used to identify discussion content and potential risk indications related to material parameters from communication information, and to perform correlation analysis between the identification results and the design entity data and the supplier's material parameters to obtain risk clues.

[0065] The correlation analysis module, connected to the risk identification module and the data storage module, is used to continuously analyze risk clues, monitor the correlation between changes in design entity data and supplier material parameter data, and propose causal hypotheses between changes in material properties and design performance issues.

[0066] The impact assessment module, connected to the correlation analysis module and the data storage module, is used to collect evidence supporting causal hypotheses and perform parameter impact assessments to quantify the impact of material parameter changes on design performance.

[0067] The risk assessment module, connected to the impact assessment module, is used to comprehensively evaluate the credibility of causal hypotheses by integrating evidence and the results of impact assessments, and to confirm causal hypotheses as potential risks based on their credibility.

[0068] The report generation module, connected to the risk assessment module and the data storage module, is used to generate risk narrative reports based on confirmed causal assumptions and collected evidence.

[0069] The display module, connected to the report generation module, is used to display the risk narrative report;

[0070] The early warning module, connected to the risk assessment module, is used to send an early warning signal to the external early warning system when the causal hypothesis is confirmed as a potential risk.

[0071] This technical solution provides a complete risk assessment system for integrated circuit design projects. Through modular design, it realizes functions such as data acquisition, risk identification, correlation analysis, impact assessment, risk evaluation, report generation, and early warning, effectively solving the problems of incomplete system functions and inability to fully cope with complex risks in existing technologies.

[0072] Beneficial effects

[0073] The risk assessment method and system for integrated circuit design projects disclosed in this application effectively solves the problem in the prior art of failing to capture and understand key information scattered in unstructured communication records by acquiring communication information between designers and suppliers and identifying discussion content and potential risk indications related to material parameters.

[0074] Simultaneously, this method performs correlation analysis on the identified discussion content, design entity data, and supplier material parameter data to obtain risk clues. It then continuously analyzes these risk clues, monitoring the correlation between changes in design entity data and supplier material parameter data. Based on these correlations, it proposes causal hypotheses between changes in material properties and design performance issues. This series of steps overcomes the limitations of existing risk assessment systems in conducting effective correlation analysis, revealing the deep-seated causal relationship and chain reactions between "uncertainty in the electrical properties of new materials" and "difficulty in design timing convergence."

[0075] Furthermore, by collecting evidence supporting the causal hypothesis and performing parameter impact assessments to quantify the impact of material parameter changes on design performance, this application can evaluate the credibility of the causal hypothesis and identify potential risks, thus solving the problem that traditional methods struggle to quantify risk transmission paths and their potential impact range. Finally, it generates a risk narrative report containing the risk evolution process, supporting evidence chains, and traceability verification index information, enabling project managers to fully understand the true risk situation and intervene in a timely manner. This avoids the disconnect between risk reports output by traditional systems and the actual experience of the project team, significantly improving the accuracy, credibility, and traceability of risk assessments, and effectively ensuring the smooth progress of integrated circuit design projects. Attached Figure Description

[0076] Figure 1 This is a flowchart of a risk assessment method for integrated circuit design projects provided by an embodiment of the present invention;

[0077] Figure 2 This is a flowchart of a method for proposing causal hypotheses between changes in material properties and design performance issues based on the correlation of changes, provided by an embodiment of the present invention.

[0078] Figure 3 This is a schematic diagram of the structure of an integrated circuit design project risk assessment system provided in an embodiment of the present invention. Detailed Implementation

[0079] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0080] Reference Figure 1 , Figure 1 This is a flowchart of a risk assessment method for integrated circuit design projects provided by an embodiment of the present invention. The method includes:

[0081] S11, Obtain communication information between designers and suppliers, and identify discussion content related to material parameters and potential risk indications from the communication information;

[0082] S12, Obtain the design entity data output by the design verification tool, as well as the supplier's material parameter data;

[0083] S13, perform correlation analysis on the identified discussion content, the design entity data and the supplier material parameter data to obtain risk clues;

[0084] S14, continuously analyze the risk clues, monitor the correlation between changes in the design entity data and the supplier's material parameter data, and propose causal hypotheses between changes in material properties and design performance problems based on the correlation between changes;

[0085] S15, collect evidence supporting the causal hypothesis, including original communication records, supplier material parameter data, and design verification data, and perform a parameter impact assessment to quantify the impact of material parameter changes on design performance;

[0086] S16, Based on the collected evidence and the results of the parameter impact assessment, evaluate the credibility of the causal hypothesis, and confirm the causal hypothesis as a potential risk based on the credibility.

[0087] S17. Based on the confirmed causal hypothesis and the collected evidence, a risk narrative report is generated. The report includes the evolution of the risk, the chain of evidence supporting the causal hypothesis, and provides index information for tracing and verifying the chain of evidence.

[0088] This application, through in-depth correlation analysis of unstructured communication information, design entity data, and supplier material parameter data, can effectively identify and quantify potential risks in integrated circuit design projects, particularly design performance issues caused by variations in material properties. By constructing causal hypotheses, collecting evidence, and conducting parameter impact assessments, this application can provide credible and quantifiable risk assessment results and generate risk narrative reports that include the risk evolution process and a traceable chain of evidence. This significantly enhances project managers' understanding of the true risk situation and their ability to intervene in a timely manner, effectively compensating for the shortcomings of existing technologies in handling complex, multi-dimensional risks.

[0089] The "communication information" mentioned in this application refers to various communication records generated between designers and suppliers during project collaboration. These records may take forms including, but are not limited to, instant messaging messages, emails, meeting minutes, and transcripts of recorded telephone conversations. This information is typically unstructured but may contain key discussions regarding material parameters, design requirements, manufacturing processes, and potential risk signals. "Design entity data" refers to various data output by design verification tools (such as EDA tools) during the integrated circuit design process, including but not limited to circuit netlists, layout data, simulation waveforms, and performance reports (such as timing reports and power consumption reports). This data reflects the specific implementation and performance of the design. "Supplier material parameter data" refers to the technical specifications and characteristics of the materials used in their products provided by material suppliers, such as dielectric constant, resistivity, coefficient of thermal expansion, and mechanical strength. These parameters directly affect the performance and reliability of the integrated circuit.

[0090] The risk assessment method for integrated circuit design projects proposed in this application aims to identify, assess, and report potential risks in integrated circuit design projects through systematic data analysis and risk modeling.

[0091] In practical implementation, the first step is to obtain communication information between designers and suppliers. This information can be collected in various ways, such as automatically capturing chat logs through the API of integrated communication tools or retrieving email content through email server interfaces. Once the communication information is obtained, it's necessary to identify discussions related to material parameters and potential risk indicators. For example, Natural Language Processing (NLP) technology can be used to analyze the text of the communication information, identifying material parameter-related discussions through keyword matching (such as "dielectric constant," "resistivity," "batch variation," etc.). Simultaneously, by analyzing the frequency and sentiment of the discussions, potential risk indicators can be identified. For instance, a significant increase in the frequency of discussions about a particular material parameter within a short period may indicate a potential problem with that parameter.

[0092] Secondly, it is necessary to obtain the design entity data output by the design verification tool, as well as the supplier's material parameter data. Design entity data can be automatically extracted from the output logs or database of the design verification tool; for example, by periodically exporting simulation result reports. Supplier material parameter data can be obtained from the supplier's specifications, test reports, or their data management system; for example, by periodically synchronizing the latest batch data of material parameters through a data interface.

[0093] Subsequently, the identified discussion content, design entity data, and supplier material parameter data are correlated to identify risk clues. For example, a material parameter mentioned in the communication can be correlated with the performance indicators affected by that parameter in the design entity data, and then compared with the actual value of that material parameter provided by the supplier. If it is found that there is high attention paid to a certain material parameter in the communication, and the actual value of that parameter deviates from the design expectation, while the design performance data also shows anomalies, then it can be marked as a risk clue.

[0094] Next, the aforementioned risk clues are continuously analyzed to monitor the correlation between changes in design entity data and supplier material parameter data. Based on this correlation, causal hypotheses about the relationship between changes in material properties and design performance issues are proposed. For example, key performance parameters (such as timing margin and power consumption) from the design entity data and key material property values ​​(such as dielectric constant and resistivity) from the supplier material parameter data can be collected periodically. By analyzing the trends of these data over time, if it is found that the change in a certain material property is consistent with the change trend of a certain design performance parameter, and the change in the material property precedes the change in the design performance parameter, a preliminary causal hypothesis can be proposed that "the change in this material property caused the design performance problem."

[0095] To verify the aforementioned causal hypothesis, evidence supporting it needs to be collected. This evidence includes original communication records, supplier material parameter data, and design verification data. For example, this could involve tracing back to the initial communication records discussing the material parameter, obtaining historical batch data for that parameter, and design verification reports corresponding to the time points of change in the material parameter. Simultaneously, a parameter impact assessment should be performed to quantify the impact of material parameter changes on design performance. For instance, parameter scanning simulations can be conducted using design verification tools to simulate changes in design performance under different material parameter values, thereby quantifying the degree of impact of a unit change in material parameters on design performance.

[0096] Based on this, the credibility of the causal hypothesis is assessed by comprehensively considering the collected evidence and the results of the parameter impact assessment, and the causal hypothesis is confirmed as a potential risk based on the credibility score. For example, a comprehensive credibility score can be calculated based on factors such as the quantity of evidence, the strength of the correlation between the evidence and the causal hypothesis, and the confidence interval of the parameter impact assessment results. If the score reaches a preset threshold, the causal hypothesis is confirmed as a potential risk.

[0097] Finally, based on the confirmed causal assumptions and collected evidence, a risk narrative report is generated. This report includes the evolution of the risk, the chain of evidence supporting the causal assumptions, and provides indexing information for tracing and verifying the chain of evidence. For example, the report can present the entire process from initial discussions to final risk confirmation in a timeline format, listing all evidence supporting the risk, each with a traceable link or identifier so that project managers can access the original data for verification at any time.

[0098] The risk assessment method for integrated circuit design projects presented in this application integrates unstructured communication information, design entity data, and supplier material parameter data to construct a complete risk assessment process from information identification to risk confirmation and report generation. Compared with existing technologies, the core innovation of this application lies in its ability to identify potential risk signals from multi-source heterogeneous data and quantify the impact of material parameter variations on design performance through correlation analysis and causal hypothesis verification, thereby providing more credible and traceable risk assessment results.

[0099] For details, please refer to Figure 2 , Figure 2 This is a flowchart of a method for proposing causal hypotheses between changes in material properties and design performance issues based on the correlation of changes, provided by an embodiment of the present invention. The method may include the following steps:

[0100] S141, Obtain the performance parameter values ​​from the design entity data at multiple time points to form a time series sequence of performance parameters;

[0101] S142, Obtain the material characteristic values ​​from the supplier's material parameter data corresponding to each time point, and form a material characteristic time sequence;

[0102] S143, when the time series of performance parameters is consistent with the time series of material properties, and the time series of material properties changes earlier than the time series of performance parameters, a causal hypothesis is proposed that the change in the supplier's material parameters leads to the design performance problem.

[0103] The performance parameter time series refers to the set of values ​​obtained through design verification tools at different stages or test batches of an integrated circuit design project, reflecting the changes of key parameters (such as power consumption, latency, yield, etc.) over time that reflect design performance. These values ​​are arranged in chronological order to reveal their dynamic changes. The material property time series refers to the set of values ​​obtained from the supplier over time for key material properties (such as resistivity, dielectric constant, doping concentration, etc.) within the time period corresponding to the time points when the performance parameters were obtained. These material property values ​​are also arranged in chronological order to reflect the stability or changes in the material itself.

[0104] Furthermore, when analyzing the time series of performance parameters and material properties, if a consistency is found in their changing trends—for example, both rising or falling simultaneously—and the time point of change in the material property time series precedes that of the performance parameter time series, a preliminary causal hypothesis can be proposed. This temporal sequence is key to establishing a causal relationship; that is, changes in material properties are considered a precondition or trigger for design performance problems.

[0105] This application's approach constructs time-series sequences of performance parameters and material properties, and rigorously compares their changing trends and temporal order to systematically identify potential causal relationships. Because changes in material properties typically precede their observable impact on design performance, this temporal lag allows for early warning and assumption of potential design performance issues by monitoring early changes in the material property time-series sequences. In this way, risk clues with strong causal indices can be sifted from large amounts of data, providing a solid foundation for subsequent risk assessment and verification.

[0106] The aforementioned technical solution provides a more accurate and scientific method to monitor the correlation between changes in design entity data and supplier material parameter data, and to propose causal hypotheses. This mechanism, based on time-series analysis and trend consistency judgment, significantly improves the accuracy of identifying design performance problems caused by changes in material properties, avoiding potential misjudgments that may arise from static data correlation analysis alone. Therefore, potential design risks can be identified earlier and more accurately, providing strong support for risk management in integrated circuit design projects and facilitating timely intervention to reduce the risk of project failure.

[0107] In some embodiments described above, this application proposes quantifying the impact of material parameter variations on design performance by performing parameter impact assessments. However, in practice, without a systematic and accurate assessment method, the quantified impact values ​​may contain uncertainties or biases, thus affecting the accurate assessment of the credibility of causal assumptions. Failure to address these issues may reduce the reliability of risk assessment results or even misjudge the severity of potential risks. Therefore, this application further proposes a more refined parameter impact assessment method, aiming to accurately quantify the specific impact of material parameter variations on design performance through structured simulation and analysis.

[0108] In this regard, the aforementioned method, specifically the performance parameter impact assessment, quantifies the impact of material parameter variations on design performance, including:

[0109] Obtain the range of parameter variations for materials provided by the supplier across multiple batches;

[0110] The range of parameter variation is divided into multiple discrete parameter value points;

[0111] For each parameter value point, the design verification tool is invoked to perform design performance simulation to obtain the design performance simulation value corresponding to each parameter value point;

[0112] Based on the simulated design performance values ​​at all parameter points, establish a table showing the correspondence between material parameters and design performance;

[0113] Based on the corresponding relationship table, the change in design performance corresponding to the unit change in material parameters is calculated and used as the influence value.

[0114] Specifically, obtaining the variation range of material parameters provided by the supplier across multiple batches refers to obtaining the actual fluctuation range of a specific material parameter in different production batches or supply cycles from the supplier. This can be achieved, for example, by reviewing the supplier's product specifications, quality control reports, or historical data records. The purpose is to clarify the actual boundaries of material parameter variation, providing a reliable input basis for subsequent simulation analysis.

[0115] Dividing the parameter variation range into multiple discrete parameter value points can be understood as selecting a series of representative, evenly spaced, or distributed numerical points within a defined material parameter variation range. For example, if the variation range of a certain resistance parameter is 100Ω ± 5%, then discrete points such as 95Ω, 96Ω, ..., 105Ω can be selected between 95Ω and 105Ω. The aim is to comprehensively cover the parameter variation range as much as possible with a limited number of simulations, so as to accurately capture the nonlinear relationship between material parameters and design performance.

[0116] In practical applications, for each parameter value, the design verification tool is invoked to perform design performance simulation, obtaining the simulated design performance value corresponding to each parameter value. Specifically, each discrete parameter value is used as input, and circuit or system-level simulations are performed using commonly used design verification tools in integrated circuit design (such as SPICE simulators, Verilog simulators, or physical verification tools). Through simulation, the specific values ​​of various performance indicators (such as power consumption, speed, timing, noise margin, etc.) of the integrated circuit design under that specific material parameter value can be obtained. The purpose is to simulate the actual impact of material parameter changes on design performance through virtual experiments.

[0117] Furthermore, based on the simulated design performance values ​​for all parameter points, a correspondence table between material parameters and design performance is established. This involves mapping each parameter point to its corresponding simulated design performance value, forming a data table or functional relationship. For example, it could be a two-dimensional table, with one column representing the material parameter value and the other column representing the corresponding design performance value. The purpose is to intuitively demonstrate the quantitative relationship between material parameters and design performance, providing a data foundation for subsequent influence value calculations.

[0118] Therefore, based on the aforementioned correspondence table, the change in design performance corresponding to a unit change in material parameters is calculated as the influence value. Specifically, this involves analyzing the correspondence table, for example using mathematical methods such as differencing, regression analysis, or curve fitting, to calculate the change in design performance that occurs when the material parameters change by one unit. This change is defined as the influence value, which quantifies the sensitivity of material parameters to design performance. Its purpose is to provide a standardized and comparable indicator for assessing the severity of the impact of different material parameters on design performance.

[0119] This application's solution systematically acquires the actual variation range of material parameters and discretizes them into multiple value points, thereby comprehensively covering potential fluctuations in material parameters. This technical solution enables precise quantification of the impact of material parameter variations on design performance, avoiding biases from subjective judgment or experience-based estimations. This systematic evaluation method makes the risk assessment process more scientific and rigorous, improving the accuracy of causal hypothesis credibility assessment. Furthermore, by establishing a correspondence table between material parameters and design performance, designers are provided with intuitive data, helping to anticipate and mitigate potential material-related risks during the design phase, thus enhancing the robustness and reliability of integrated circuit design.

[0120] Specifically, in some implementations of the aforementioned risk assessment method for integrated circuit design projects, the credibility of causal assumptions can be assessed in the following ways.

[0121] The assessment of the credibility of the causal hypothesis includes:

[0122] Obtain the number of pieces of evidence supporting the causal hypothesis, and determine the strength of the association between each piece of evidence and the causal hypothesis;

[0123] Obtain the confidence interval of the impact value obtained from the parameter impact assessment;

[0124] The confidence score of the causal hypothesis is calculated based on the amount of evidence, the level of correlation strength, and the confidence interval.

[0125] The quantity of evidence supporting the causal hypothesis refers to the total amount of evidence collected after gathering supporting evidence. This evidence may include original communication records, supplier material parameter data, and design verification data. Simultaneously, it is necessary to determine the correlation strength level between each piece of evidence and the causal hypothesis. The correlation strength level can be divided based on factors such as the directness, reliability, and completeness of the evidence; for example, it can be divided into three levels: high, medium, and low, or a more detailed numerical scoring system can be used. The higher the correlation strength level, the stronger the support the evidence provides for the causal hypothesis.

[0126] Furthermore, obtaining the confidence interval for the impact value obtained from the parameter impact assessment refers to performing a statistical analysis on the impact value after performing the parameter impact assessment and quantifying the impact value of material parameter changes on design performance, in order to determine its possible range. The confidence interval reflects the accuracy and reliability of the impact value estimation. For example, a 95% confidence interval can be calculated, indicating that the actual impact value will fall within this interval with a 95% probability.

[0127] Therefore, based on the amount of evidence obtained, the strength of the association between each piece of evidence and the causal hypothesis, and the confidence interval of the impact value obtained from the parameter impact assessment, the credibility score of the causal hypothesis can be calculated. This credibility score is a comprehensive indicator used to quantify the reliability of the causal hypothesis. For example, a weighted algorithm can be designed to take into account factors such as the amount of evidence, the strength of the association, and the confidence interval. The more evidence, the higher the strength of the association, and the narrower the confidence interval (i.e., the more precise the impact value), the higher the credibility score.

[0128] This application's solution, by comprehensively considering the quantity of evidence supporting the causal hypothesis, the strength of the correlation between the evidence and the causal hypothesis, and the confidence interval of the parameter influence assessment results, can quantitatively evaluate the reliability of the causal hypothesis from multiple dimensions. Through this technical solution, a more scientific and quantitative method can be provided to evaluate the causal hypothesis of material property changes causing design performance problems in integrated circuit design projects. This multi-dimensional and comprehensive evaluation method not only improves the accuracy and reliability of risk assessment but also provides a solid data foundation for subsequent risk confirmation and report generation, helping design teams and management make more informed decisions and effectively reduce project risks.

[0129] While the aforementioned schemes for assessing the credibility of causal hypotheses can calculate a credibility score based on the amount of evidence, the strength of association, and the confidence interval of the impact value, there is still a lack of specific judgment criteria and operational procedures for translating this credibility score into a clear identification of potential risks. Without a standardized mechanism, risk identification may rely on the subjective judgment of assessors, leading to inconsistencies in assessment results or the overlooking of potential risks.

[0130] In response, this application further proposes a specific method for confirming causal assumptions as potential risks based on credibility. By introducing a risk assessment rule base and risk level thresholds, the risk confirmation process is made more standardized and automated, thereby improving the accuracy and efficiency of risk assessment.

[0131] The above-mentioned confirmation of causal assumptions as potential risks based on credibility includes:

[0132] Obtain a pre-generated risk assessment rule base, which contains multiple risk levels and their corresponding confidence score ranges;

[0133] The calculated credibility score is matched with the credibility score range in the risk assessment rule base to determine the risk level corresponding to the causal hypothesis;

[0134] When the risk level exceeds a preset risk level threshold, the causal hypothesis is confirmed as a potential risk.

[0135] Specifically, the risk assessment rule base can be understood as a predefined set of standards designed to map continuous credibility scores to discrete risk levels. This rule base is typically built and maintained by domain experts based on historical data, industry experience, and risk preferences. The risk levels may include, but are not limited to, "low risk," "medium risk," and "high risk," each corresponding to a specific credibility score range. For example, a credibility score of 0-30 might be defined as "low risk," 31-70 as "medium risk," and 71-100 as "high risk."

[0136] After obtaining the risk assessment rule base, the previously calculated confidence scores of causal hypotheses are matched with the various confidence score ranges defined in the rule base. This matching process aims to determine which risk level range the current causal hypothesis's confidence score falls into, thereby clarifying the risk level of the causal hypothesis.

[0137] Furthermore, to definitively confirm whether a causal hypothesis is a potential risk, a preset risk level threshold needs to be introduced. This threshold is a pre-determined risk level boundary, such as "medium risk" or "high risk." When the risk level of the causal hypothesis obtained through matching is higher than or equal to this preset risk level threshold, the causal hypothesis is considered to have a sufficient degree of risk and is thus formally confirmed or marked as a potential risk.

[0138] This application's solution effectively solves the standardized problem of transforming the credibility scores of causal assumptions into clear potential risk identification by introducing a risk assessment rule base and preset risk level thresholds. Through the above technical solution, this application provides a more objective, standardized, and automated potential risk identification mechanism. This mechanism significantly reduces reliance on subjective human judgment in the risk identification process, thereby improving the accuracy and consistency of risk assessment. Furthermore, through clear risk level classification and threshold setting, early identification and warning of potential risks can be achieved, enabling project teams to take preventative and mitigation measures more promptly, effectively reducing the risks faced by integrated circuit design projects, and thus improving project success rates and product quality.

[0139] In some embodiments described above in this application, a risk narrative report is generated, which includes the evolution of the risk, a chain of evidence supporting the causal hypothesis, and provides index information for retrospectively verifying the chain of evidence. However, in its implementation, if the report generation method lacks structure and systematicity, the evolution of the risk may not be clear enough, the logic of the chain of evidence may be weak, and the efficiency of retrospective verification may be low, thereby affecting the effectiveness of risk assessment and the practicality of the report.

[0140] In this regard, this application further proposes steps for generating a risk narrative report, including:

[0141] Extract key events in the evolution of the risk in chronological order to generate a risk evolution timeline;

[0142] The collected evidence is sorted according to its strength of association with the causal hypothesis to form a chain of evidence;

[0143] Generate a traceable storage index identifier for each piece of evidence in the chain of evidence;

[0144] The risk evolution timeline, the evidence chain, and the storage index identifier are integrated into the risk narrative report.

[0145] Specifically, generating a risk evolution timeline involves identifying all significant milestones, decision points, and key data changes throughout the entire process from the occurrence and development of a risk to its confirmation, and arranging and presenting these events in chronological order. The aim is to provide a clear and intuitive understanding of risk development, aiding in the comprehension of the dynamic evolution of risks.

[0146] Forming a chain of evidence involves prioritizing all supporting evidence, such as original communication records, supplier material parameter data, and design verification data, according to their relevance or strength of support for the causal hypothesis. Evidence with strong relevance is placed at the beginning of the chain to highlight its importance. The goal is to construct a logically rigorous and persuasive evidence system that facilitates review and verification.

[0147] In practical applications, generating a traceable storage index identifier for each piece of evidence means assigning a unique identifier to each piece of evidence in the chain of evidence, which can be used to locate its original storage location. This identifier can be a file path, database record ID, hash value, or any information that can point to the original data source. Its purpose is to ensure that all evidence in the report can be easily and quickly traced back to its original source, thereby enhancing the report's transparency and credibility. Furthermore, integrating the risk evolution timeline, evidence chain, and storage index identifier into a risk narrative report means organically combining the above structured information to form a complete, coherent, and easily understandable risk narrative report. This report not only presents "what" and "why" the risk exists, but also clearly demonstrates "how" it occurred and "how" it can be verified.

[0148] This application's solution presents the risk evolution process in a structured timeline, making the occurrence and development of risks readily apparent and resolving the issues of disorganized event sequences and unclear logic that may exist in traditional reports. The resulting risk narrative report, generated through this technical solution, is not only comprehensive in content but also clearly structured, logically rigorous, and highly traceable. The risk evolution timeline visually demonstrates the dynamic changes in risks, facilitating a rapid grasp of the overall risk picture; the construction of the evidence chain strengthens the arguments supporting causal hypotheses, improving the accuracy of risk assessment; and traceable storage indexes ensure the authenticity and reliability of all information in the report, significantly reducing the difficulty and time cost of information verification. These improvements collectively enhance the practical value and decision support capabilities of risk narrative reports, enabling potential risks in integrated circuit design projects to be more effectively identified, understood, and managed.

[0149] In some of the embodiments described above in this application, although a traceable storage index identifier is generated for each piece of evidence in the chain of evidence, and index information is provided for tracing and verifying the chain of evidence, in practical applications, simply providing an identifier or index information may not be sufficient to ensure that the original evidence can be traced efficiently, accurately, and comprehensively when needed. For example, when the sources of evidence are diverse, the storage locations are scattered, or the data volume is large, a simple index may not provide sufficient contextual information, resulting in a time-consuming and error-prone tracing process, and even making it difficult to verify the authenticity and integrity of the evidence.

[0150] In this regard, this application further proposes the aforementioned index information for tracing and verifying the chain of evidence, including:

[0151] Generate metadata tags for each piece of evidence in the chain of evidence, including storage path, creation timestamp, and data source identifier;

[0152] The metadata tags are associated with the corresponding evidence and stored in the database;

[0153] The link element for embedding the metadata tag is placed in the risk narrative report;

[0154] In response to the triggering of the linked element, the corresponding evidence content is retrieved from the database and displayed.

[0155] Specifically, metadata tags are data attached to evidence that describes its own attributes and characteristics. These include: a storage path, indicating the evidence's specific location within the storage system (e.g., a file server directory path or cloud storage URI); a creation timestamp, recording the precise time the evidence was generated or captured by the system to ensure traceability and the timeliness of evidence verification; and a data source identifier, clarifying the original origin of the evidence (e.g., from the designer's email system, a design verification tool's log file, or a supplier's parameter reporting system). This information collectively constitutes the evidence's unique identifier and contextual information, aiming to provide detailed foundational data for subsequent traceability and verification.

[0156] This involves associating metadata tags with corresponding evidence and storing them in a database. This can be understood as creating a record for each piece of evidence in a dedicated evidence management database. This record contains not only the evidence's metadata tags but also references to the actual evidence content or directly stores the binary data of the evidence. The purpose of this associative storage is to achieve centralized evidence management, rapid retrieval, and data consistency, ensuring that the one-to-one correspondence between metadata tags and original evidence is not disrupted.

[0157] In practical applications, embedding metadata tags in risk narrative reports refers to presenting metadata tags as clickable hyperlinks, buttons, or icons within the generated risk narrative report (such as a PDF document, web report, or interactive dashboard). For example, when a report mentions a key piece of evidence, a clickable link will appear next to it, containing the corresponding evidence's metadata tag information. The purpose is to provide an intuitive and convenient way for users to directly access the original evidence from the report.

[0158] Furthermore, responding to the triggering of a linked element, retrieving and displaying the corresponding evidence content from the database means that when a user clicks on a link embedded in the report, the system parses the metadata tag information contained in the link and uses this information to locate and retrieve the corresponding original evidence data from the database. Subsequently, this evidence content is presented on the user interface in an appropriate format (such as a text viewer, image browser, or document reader), with the aim of enabling users to instantly view the detailed content of the original evidence, thereby completing the tracing and verification of the evidence chain.

[0159] This application's solution generates metadata tags for each piece of evidence in the evidence chain, including storage path, creation timestamp, and data source identifier, giving each piece of evidence detailed and verifiable identity information. Through this technical solution, this application significantly improves the traceability and verification capability of the evidence chain during the risk assessment process of integrated circuit design projects. Specifically, the detailed metadata tags make the source, time, and storage location of each piece of evidence immediately clear, greatly simplifying the evidence location and verification process and effectively avoiding traceability difficulties caused by missing or ambiguous information. Furthermore, by storing the metadata tags in a database in association with the evidence and embedding them in the report as link elements, centralized management and convenient access to evidence are achieved, allowing users to retrieve and view original evidence with a single click, thereby significantly improving the transparency and credibility of the risk assessment report. This mechanism not only ensures the authenticity and integrity of the evidence but also significantly improves the efficiency and accuracy of risk assessment, providing decision-makers with a more solid and reliable basis.

[0160] Specifically, the identification of material parameter-related discussion content and potential risk indications from the aforementioned communication information can be carried out in the following manner.

[0161] The process of identifying discussion content related to material parameters and indications of potential risks from the communication information includes:

[0162] Obtain the text content and timestamp information from the communication information;

[0163] The text content is subjected to keyword matching to identify keywords related to material parameters and the corresponding discussion time.

[0164] Based on the discussion time, the keywords are arranged in chronological order to form a discussion timeline of the material parameters;

[0165] Identify periods of sudden increases in discussion frequency from the discussion timeline as indicators of potential risks.

[0166] Specifically, the communication information can take various forms, including but not limited to emails, instant messaging records, meeting minutes, and project documents, and includes communication content between designers and suppliers regarding integrated circuit design projects. When acquiring this communication information, it is necessary to obtain both the text content and the corresponding timestamp information for subsequent time-series analysis.

[0167] The keyword matching process for the text content aims to accurately filter out discussions related to material parameters from a massive amount of communication information. These keywords can be predefined, such as "resistivity," "capacitance," "dielectric constant," "threshold voltage," "process node," "material supplier," and "batch," or they can be automatically extracted from historical data using machine learning or natural language processing techniques. By matching these keywords, specific material parameters and the times they were mentioned in the communication can be identified.

[0168] Furthermore, based on the identified discussion times, the keywords are arranged chronologically to construct a discussion timeline for the material parameters. This timeline visually demonstrates the changes in attention given to specific material parameters throughout the project lifecycle. Analyzing this timeline identifies periods of sudden increases in discussion frequency. A sudden increase in discussion frequency often indicates potential anomalies, controversies, or issues requiring special attention for a particular material parameter; therefore, these periods of increased discussion frequency are considered indicators of potential risks. For example, if a material parameter is frequently mentioned within a certain timeframe, and the tone or content of the discussion suggests uncertainty or problems, that period may foreshadow potential risks.

[0169] This application's solution systematically acquires and analyzes communication information between designers and suppliers, enabling the identification of potential material parameter-related risks at the source. Through this technical solution, this application achieves early warning and accurate identification of potential risks in integrated circuit design projects. Compared to relying solely on design verification data or material parameter data for risk assessment, this solution, through in-depth analysis of personnel communication information, can capture risk signals hidden in unstructured data, especially in the early stages of material parameter anomalies or changes. This method not only improves the sensitivity of risk identification but also provides richer and more timely information for subsequent risk correlation analysis, thereby effectively reducing the risk of design rework or performance failures due to material issues in the later stages of the project, significantly improving the success rate and efficiency of integrated circuit design projects.

[0170] In some embodiments of this application, the identified discussion content, design entity data, and supplier material parameter data are correlated to obtain risk clues. Specifically, the correlation analysis of the identified discussion content, design entity data, and supplier material parameter data includes the following steps:

[0171] Extract material parameter keywords from the identified discussion content;

[0172] Extract the performance parameter values ​​corresponding to the material parameter keywords from the design entity data;

[0173] Extract material property values ​​corresponding to the material parameter keywords from the supplier's material parameter data;

[0174] Combinations where the correlation between the material property value and the performance parameter value exceeds a preset threshold are marked as risk clues.

[0175] Extracting material parameter keywords from the identified discussion content can be understood as using natural language processing techniques, text mining algorithms, or predefined keyword dictionaries to perform in-depth analysis of the text content in the communication between designers and suppliers, in order to identify terms or phrases related to integrated circuit material parameters, such as "dielectric constant," "resistivity," and "coefficient of thermal expansion." The extraction of these keywords aims to transform unstructured communication information into structured information that can be used for subsequent data correlation analysis, thereby providing clear focus points for risk identification.

[0176] Furthermore, extracting performance parameter values ​​corresponding to the material parameter keywords from the design entity data refers to searching and extracting design performance indicators that may be related to these material parameters from the design entity data output by the design verification tool after identifying specific material parameter keywords. For example, if the identified keyword is "dielectric constant," it may be necessary to extract performance parameter values ​​related to circuit signal integrity, delay, power consumption, or crosstalk. These performance parameter values ​​can be specific values ​​from simulation results, test reports, or design specifications, reflecting the design's behavior under specific conditions.

[0177] Furthermore, extracting material property values ​​corresponding to the material parameter keywords from the supplier's material parameter data refers to obtaining the actual property values ​​or their range of variation of the corresponding materials from the material parameter data provided by the supplier based on the identified material parameter keywords. For example, if the keyword is "resistivity," then the resistivity values ​​of a specific batch of material are obtained from the material specifications, test reports, or databases provided by the supplier. These material property values ​​are the direct basis for evaluating the impact of materials on design performance, and their accuracy is crucial for subsequent correlation analysis.

[0178] Finally, combinations where the correlation between the material property value and the performance parameter value exceeds a preset threshold are marked as risk clues. The correlation can be calculated using various statistical methods or machine learning models, such as Pearson correlation coefficient, Spearman rank correlation coefficient, regression analysis, or causal inference models. The preset threshold is a configurable parameter used to define what level of correlation is considered significant or noteworthy. When the correlation between a material property value and a design performance parameter value reaches or exceeds this threshold, it indicates that changes in the material parameter may have a significant impact on the design performance, thus marking such strongly correlated combinations as potential risk clues.

[0179] This application's solution effectively identifies potential risk clues by conducting detailed correlation analysis of communication information between designers and suppliers, design entity data, and supplier material parameter data. Through the above technical solution, this application provides a more refined and data-driven risk clue identification mechanism. Compared to only performing coarse correlation analysis, this solution, by explicitly extracting material parameter keywords, corresponding performance parameter values, and material property values, and quantifying the correlation between them, can more accurately pinpoint which specific material parameter changes may lead to which design performance problems. This helps avoid misjudging irrelevant discussions or data as risks, thereby improving the efficiency and accuracy of risk identification and laying a solid foundation for subsequent risk analysis and assessment.

[0180] refer to Figure 3 , Figure 3 This is a schematic diagram of the structure of an integrated circuit design project risk assessment system provided in an embodiment of the present invention. The system includes:

[0181] The data acquisition module is used to connect to the designer's communication terminal, the design verification tool terminal, and the supplier's data interface to obtain communication information, design entity data, and supplier material parameter data.

[0182] A data storage module, connected to the data acquisition module, is used to store the communication information, the design entity data, and the supplier material parameter data;

[0183] The risk identification module, connected to the data storage module, is used to identify discussion content and potential risk indications related to material parameters from the communication information, and to perform correlation analysis between the identification results and the design entity data and the supplier's material parameters to obtain risk clues.

[0184] The correlation analysis module, connected to the risk identification module and the data storage module, is used to continuously analyze the risk clues, monitor the correlation between changes in the design entity data and the supplier's material parameter data, and propose causal hypotheses between changes in material properties and design performance issues.

[0185] The impact assessment module, connected to the correlation analysis module and the data storage module, is used to collect evidence supporting the causal hypothesis and perform parameter impact assessment to quantify the impact of material parameter changes on design performance.

[0186] A risk assessment module, connected to the impact assessment module, is used to assess the credibility of the causal hypothesis by integrating the evidence and the results of the impact assessment, and to confirm the causal hypothesis as a potential risk based on the credibility.

[0187] The report generation module, connected to the risk assessment module and the data storage module, is used to generate a risk narrative report based on the confirmed causal assumptions and the collected evidence.

[0188] The display module, connected to the report generation module, is used to display the risk narrative report;

[0189] The early warning module, connected to the risk assessment module, is used to send an early warning signal to an external early warning system when the causal hypothesis is confirmed as a potential risk.

[0190] The integrated circuit design project risk assessment system proposed in this application aims to automate and systematically assess potential risks in integrated circuit design projects through modular design. By integrating multi-source heterogeneous data, including unstructured communication information, structured design entity data, and supplier material parameter data, the system can effectively identify, quantify, and report design performance problems caused by changes in material properties, thereby significantly improving project managers' understanding of the true risk situation and their ability to intervene in a timely manner.

[0191] Specifically, the data acquisition module is configured to connect to the designer's communication terminal, the design verification tool terminal, and the supplier's data interface. The connection to the designer's communication terminal can be via an integrated instant messaging tool's application programming interface (API) or mail server interface to obtain communication information between the designer and supplier personnel in real time or periodically. The connection to the design verification tool terminal can be via a file system interface, database connection, or specific EDA tool API to automatically extract design entity data, such as simulation reports, netlist files, or layout data. The connection to the supplier's data interface can be via standardized data exchange protocols (such as RESTful API, SOAP) or file transfer protocols (such as FTP) to obtain the latest supplier material parameter data. In this way, the data acquisition module can comprehensively and efficiently collect various types of raw data required for the project.

[0192] The data storage module is connected to the data acquisition module and is used to store the acquired communication information, design entity data, and supplier material parameter data. This module can employ various storage technologies, such as relational databases, non-relational databases (e.g., document databases, graph databases), or distributed file systems, to accommodate different types and scales of data. The stored data can be indexed and categorized for efficient retrieval and analysis by subsequent modules.

[0193] The risk identification module is connected to the data storage module. Its function is to identify discussion content related to material parameters and potential risk indicators from the stored communication information, and to perform correlation analysis between the identification results and the design entity data and supplier material parameters to ultimately obtain risk clues. The specific methods for identifying discussion content and potential risk indicators from communication information, and how to perform correlation analysis between the discussion content, design entity data, and supplier material parameters to obtain risk clues, have already been described in the above embodiments, and will not be repeated here. It is important to emphasize that the risk identification module can integrate Natural Language Processing (NLP) engines, keyword matching algorithms, and text mining techniques to automatically extract key information from unstructured text and use preset rules or machine learning models to identify potential risk signals.

[0194] The correlation analysis module is connected to the risk identification module and the data storage module. It is used to continuously analyze risk clues, monitor the correlation between changes in design entity data and supplier material parameter data, and propose causal hypotheses between changes in material properties and design performance issues. The specific methods for monitoring data changes and proposing causal hypotheses have already been described in the above embodiments and will not be repeated here. It is important to emphasize that the correlation analysis module can employ techniques such as time series analysis, statistical regression analysis, and causal inference models in machine learning to identify potential causal relationships between data.

[0195] The impact assessment module is connected to the correlation analysis module and the data storage module. It is used to collect evidence supporting the causal hypothesis and perform parameter impact assessments to quantify the impact of material parameter changes on design performance. The specific methods for collecting evidence and performing parameter impact assessments have been described in the above embodiments and will not be repeated here. It is important to emphasize that the impact assessment module can be deeply integrated with design verification tools. For example, by calling simulation scripts or APIs, it can automatically perform parameter scan simulations, thereby accurately quantifying the sensitivity of material parameter changes to design performance.

[0196] The risk assessment module is connected to the impact assessment module and is used to comprehensively assess the results of the impact assessment based on the collected evidence and parameters, evaluate the credibility of the causal hypothesis, and confirm the causal hypothesis as a potential risk based on the credibility. The specific methods for assessing credibility and confirming potential risks have been described in the above embodiments and will not be repeated here. It is important to emphasize that the risk assessment module can have a built-in risk assessment rule base, which can contain risk scoring models trained based on expert experience or historical data, used to calculate the credibility score of the causal hypothesis and compare it with a preset risk level threshold.

[0197] The report generation module is connected to the risk assessment module and the data storage module, and is used to generate a risk narrative report based on the confirmed causal assumptions and collected evidence. The specific content and method of generating the risk narrative report have been described in the above embodiments and will not be repeated here. It should be emphasized that the report generation module can use template engine technology to automatically integrate the risk evolution timeline, evidence chain, and tracing index information into a structured and easy-to-read report.

[0198] The display module is connected to the report generation module and is used to display the risk narrative report. This module can be a graphical user interface (GUI), a web interface, or a dashboard integrated into the project management platform, so that project managers can intuitively view the risk assessment results.

[0199] The early warning module is connected to the risk assessment module and is used to send an early warning signal to an external early warning system when the causal hypothesis is confirmed as a potential risk. This early warning signal may include information such as the risk level, affected design performance, and suggested areas of concern, and can be sent via email, SMS, instant messaging, or integration into an enterprise early warning platform to ensure that the project team receives timely risk notifications and takes appropriate countermeasures.

[0200] The integrated circuit design project risk assessment system proposed in this application, through its modular architecture and close collaboration between modules, effectively addresses the problem of existing technologies struggling to conduct effective correlation analysis when dealing with complex, multi-dimensional risks. Compared to traditional risk assessment systems, this system can identify key risk clues from unstructured communication information and perform in-depth correlation analysis on multi-source heterogeneous data through automated tools and algorithms, thereby revealing the deep causal relationship between changes in material properties and design performance issues.

[0201] This system overcomes the limitations of existing systems that primarily rely on structured data input by leveraging the extensive data access capabilities of its data acquisition module. It can capture "hidden" risk signals scattered throughout unstructured communication records. The collaborative work of the risk identification and correlation analysis modules enables the system to identify and model complex relationships between technical issues, supply chain problems, and team communication issues, avoiding the treatment of related risks as isolated events. Furthermore, the quantitative impact values ​​provided by the impact assessment module offer project managers a more accurate basis for risk quantification, rather than relying solely on empirical qualitative judgments. Finally, the risk narrative reports provided by the report generation and display modules, including the risk evolution process and a traceable chain of evidence, greatly enhance the transparency and credibility of risk assessment results. This allows project managers to trust the system's warnings and take timely and effective risk response measures, thereby significantly reducing the risk of major losses such as project tape-out failures.

[0202] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for risk assessment of integrated circuit design projects, characterized in that, The method includes: Acquire communication information between designers and suppliers, and identify discussions related to material parameters and indications of potential risks from the communication information; Obtain the design entity data output by the design verification tool, as well as the supplier's material parameter data; By performing correlation analysis on the identified discussion content, the design entity data, and the supplier material parameter data, risk clues can be obtained. The risk clues are continuously analyzed to monitor the correlation between changes in the design entity data and the supplier's material parameter data. Based on the correlation, causal hypotheses between changes in material properties and design performance problems are proposed. Collect evidence supporting the causal hypothesis, including original communication records, supplier material parameter data, and design verification data, and perform a parameter impact assessment to quantify the impact of material parameter changes on design performance; Based on the collected evidence and the results of the parameter impact assessment, the credibility of the causal hypothesis is evaluated, and the causal hypothesis is confirmed as a potential risk based on the credibility. Based on the confirmed causal hypothesis and the collected evidence, a risk narrative report is generated, which includes the evolution of the risk, the chain of evidence supporting the causal hypothesis, and provides index information for tracing and verifying the chain of evidence.

2. The method according to claim 1, characterized in that, The monitoring of the correlation between changes in the design entity data and the supplier's material parameter data includes: The performance parameter values ​​from the design entity data at multiple time points are obtained to form a time series sequence of performance parameters. Obtain the material property values ​​from the supplier's material parameter data corresponding to each time point, and form a material property time series; When the time series of performance parameters and the time series of material properties show the same trend, and the time series of material properties changes earlier than the time series of performance parameters, a causal hypothesis is proposed that the change in the supplier's material parameters leads to the design performance problem.

3. The method according to claim 1, characterized in that, The performance parameter impact assessment quantifies the impact of material parameter changes on design performance, including: Obtain the range of parameter variations for materials provided by the supplier across multiple batches; The range of parameter variation is divided into multiple discrete parameter value points; For each parameter value point, the design verification tool is invoked to perform design performance simulation to obtain the design performance simulation value corresponding to each parameter value point; Based on the simulated design performance values ​​at all parameter points, establish a table showing the correspondence between material parameters and design performance; Based on the corresponding relationship table, the change in design performance corresponding to the unit change in material parameters is calculated and used as the influence value.

4. The method according to claim 1, characterized in that, The assessment of the credibility of the causal hypothesis includes: Obtain the number of pieces of evidence supporting the causal hypothesis, and determine the strength of the association between each piece of evidence and the causal hypothesis; Obtain the confidence interval of the impact value obtained from the parameter impact assessment; The confidence score of the causal hypothesis is calculated based on the amount of evidence, the level of correlation strength, and the confidence interval.

5. The method according to claim 4, characterized in that, The step of confirming the causal hypothesis as a potential risk based on the credibility level includes: Obtain a pre-generated risk assessment rule base, which contains multiple risk levels and their corresponding confidence score ranges; The calculated credibility score is matched with the credibility score range in the risk assessment rule base to determine the risk level corresponding to the causal hypothesis; When the risk level exceeds a preset risk level threshold, the causal hypothesis is confirmed as a potential risk.

6. The method according to claim 1, characterized in that, The generation of the risk narrative report includes: Extract key events in the evolution of the risk in chronological order to generate a risk evolution timeline; The collected evidence is sorted according to its strength of association with the causal hypothesis to form a chain of evidence; Generate a traceable storage index identifier for each piece of evidence in the chain of evidence; The risk evolution timeline, the evidence chain, and the storage index identifier are integrated into the risk narrative report.

7. The method according to claim 6, characterized in that, The index information provided for tracing and verifying the chain of evidence includes: Generate metadata tags for each piece of evidence in the chain of evidence, including storage path, creation timestamp, and data source identifier; The metadata tags are associated with the corresponding evidence and stored in the database; The link element for embedding the metadata tag is placed in the risk narrative report; In response to the triggering of the linked element, the corresponding evidence content is retrieved from the database and displayed.

8. The method according to claim 1, characterized in that, The process of identifying discussion content related to material parameters and indications of potential risks from the communication information includes: Obtain the text content and timestamp information from the communication information; The text content is subjected to keyword matching to identify keywords related to material parameters and the corresponding discussion time. Based on the discussion time, the keywords are arranged in chronological order to form a discussion timeline of the material parameters; Identify periods of sudden increases in discussion frequency from the discussion timeline as indicators of potential risks.

9. The method according to claim 1, characterized in that, The process of performing correlation analysis on the identified discussion content, the design entity data, and the supplier material parameters includes: Extract material parameter keywords from the identified discussion content; Extract the performance parameter values ​​corresponding to the material parameter keywords from the design entity data; Extract material property values ​​corresponding to the material parameter keywords from the supplier's material parameter data; Combinations where the correlation between the material property value and the performance parameter value exceeds a preset threshold are marked as risk clues.

10. A risk assessment system for integrated circuit design projects, characterized in that, The system includes: The data acquisition module is used to connect to the designer's communication terminal, the design verification tool terminal, and the supplier's data interface to obtain communication information, design entity data, and supplier material parameter data. A data storage module, connected to the data acquisition module, is used to store the communication information, the design entity data, and the supplier material parameter data; The risk identification module, connected to the data storage module, is used to identify discussion content and potential risk indications related to material parameters from the communication information, and to perform correlation analysis between the identification results and the design entity data and the supplier's material parameters to obtain risk clues. The correlation analysis module, connected to the risk identification module and the data storage module, is used to continuously analyze the risk clues, monitor the correlation between changes in the design entity data and the supplier's material parameter data, and propose causal hypotheses between changes in material properties and design performance issues. The impact assessment module, connected to the correlation analysis module and the data storage module, is used to collect evidence supporting the causal hypothesis and perform parameter impact assessment to quantify the impact of material parameter changes on design performance. A risk assessment module, connected to the impact assessment module, is used to assess the credibility of the causal hypothesis by integrating the evidence and the results of the impact assessment, and to confirm the causal hypothesis as a potential risk based on the credibility. The report generation module, connected to the risk assessment module and the data storage module, is used to generate a risk narrative report based on the confirmed causal assumptions and the collected evidence. The display module, connected to the report generation module, is used to display the risk narrative report; The early warning module, connected to the risk assessment module, is used to send an early warning signal to an external early warning system when the causal hypothesis is confirmed as a potential risk.

Images