A method and system for supplier evaluation

Abstract

The application belongs to the technical field of supplier evaluation, and discloses a supplier evaluation method and system, which comprises the following steps: obtaining target text data containing enterprise strategic guidance intention; extracting strategic keywords and corresponding effective priorities from the target text data; determining the quantitative adjustment parameters of target evaluation indexes corresponding to the strategic keywords according to the effective priorities; correcting the weights of all target evaluation indexes according to the quantitative adjustment parameters, and performing normalization processing on the weights of all preset evaluation indexes to obtain a target weight distribution scheme; obtaining original evaluation data of each supplier, combining the target weight distribution scheme, performing quantitative evaluation on each candidate supplier, and outputting an evaluation result; thus, a dynamic, accurate and highly consistent evaluation result with enterprise strategy can be provided, and the efficiency, accuracy and reliability of the evaluation are significantly improved.

Description

Technical Field

[0001] This application relates to the field of supplier evaluation technology, and more specifically, to a supplier evaluation method and system. Background Technology

[0002] In modern enterprise management, systematic supplier evaluation is a crucial step in ensuring supply chain stability and product quality. Many companies have introduced automated evaluation systems for this purpose. However, these systems often encounter difficulties in the face of rapid changes in the external environment, especially when companies need to adjust their procurement strategies. Traditional evaluation systems typically rely on pre-set, fixed evaluation indicator weights, making it difficult to flexibly adapt to new business priorities. This leads to evaluation results that are out of sync with actual strategic needs, thus affecting the accuracy and efficiency of decision-making.

[0003] Specifically, when the global economic environment and market conditions change drastically, such as due to regional tensions, global disease outbreaks, or natural disasters in key raw material producing areas, senior management must quickly adjust their procurement strategies. At this time, companies may temporarily place less emphasis on indicators such as "cost control" and instead prioritize "supply stability," "risk management capabilities," and "rapid response capabilities" to ensure continuous production line operation and uninterrupted product supply to the market.

[0004] However, the automated evaluation system initially deployed by the company, whose internal evaluation indicator weights were fixed based on previous market conditions and procurement strategies, could not automatically perceive and adapt to this new business priority. When the system evaluated and scored a new batch of suppliers according to the scheduled time, suppliers who performed well in terms of supply stability and had multiple procurement channels but might have slightly higher costs due to their greater reliability did not see an effective improvement in their overall scores. Conversely, some suppliers who offered lower prices but had fragile supply chains and were highly dependent on a particular risky region received relatively higher scores because the old weighting still favored cost factors. This evaluation result created a significant discrepancy between the company's current urgent strategic needs and actual business experience.

[0005] Auditors will immediately notice this inconsistency when reviewing the system-generated assessment reports. This discrepancy between the results and actual business judgments and the company's latest strategic direction leads to a strong sense of distrust towards the system's output. To ensure the assessment results accurately reflect the new business priorities, auditors are forced to spend significant time and effort manually adjusting the automatically generated reports. While this manual intervention temporarily compensates for the system's shortcomings, it significantly reduces the efficiency of the assessment process. Furthermore, due to differences in experience, understanding, and subjective judgment among auditors, the manually adjusted assessment results lack consistency among different auditors.

[0006] The purchasing department managers quickly noticed this widespread manual adjustment and the resulting inconsistencies in evaluation results. To standardize operations, they issued an internal guidance document attempting to provide auditors with principles for revising system scores. However, because the principles in the guidance document were not specific quantitative instructions but rather used vague qualitative expressions such as "increase as appropriate" and "decrease appropriately," different auditors still differed in their understanding and application of these principles regarding the specific magnitude and direction of score adjustments. This difference stemmed from the fact that when people translate abstract qualitative concepts into concrete quantitative operations, they are influenced by factors such as their personal cognition, experience background, and risk appetite, making it difficult to achieve a unified standard even with guidance. This further exacerbated the non-standardization problem of evaluation results, making it difficult for the company to obtain a standardized and reliable overview of suppliers and to provide consistent data support for subsequent purchasing decisions.

[0007] Ultimately, the company's management realized that neither spontaneous adjustments by auditors nor corrections based on non-quantifiable guidance documents could fundamentally solve the problem. This reliance on manual, post-hoc remediation was not only cumbersome and inefficient, but the revised data also mixed the system's original output with human judgment, making the source of the assessment data and the logic behind the adjustments complex and opaque. This severely impacted the reliability and comparability of the assessment data, preventing the company from conducting reliable horizontal comparative analysis, trend forecasting, or strategic planning based on historical data.

[0008] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0009] This application discloses a supplier evaluation method and system, which aims to solve the problems of traditional supplier evaluation systems being unable to flexibly adapt to new business priorities when facing changes in the external environment, resulting in evaluation results being out of touch with actual strategic needs, as well as the low efficiency of manual intervention and the lack of consistency and reliability of evaluation results.

[0010] Firstly, this application provides a supplier evaluation method, which includes the following steps: A1. Obtain target text data containing the company's strategic guiding intent; A2. Extract strategic keywords and their corresponding effective priorities from the target text data; A3. Based on the effective priority, determine the quantitative adjustment parameters of the target evaluation indicators corresponding to the strategic keywords; the target evaluation indicators are a part of a plurality of preset evaluation indicators; A4. Based on the quantitative adjustment parameters, correct the weights of each target evaluation indicator and normalize the weights of all preset evaluation indicators to obtain the target weight allocation scheme. A5. Obtain the original evaluation data of each supplier, combine it with the target weight allocation scheme, perform quantitative evaluation of each candidate supplier, and output the evaluation results.

[0011] Secondly, this application provides a supplier evaluation system, which includes: The text acquisition module is used to acquire target text data containing the company's strategic guiding intent; The information extraction module is used to extract strategic keywords and their corresponding effective priorities from the target text data; The parameter determination module is used to determine the quantitative adjustment parameters of the target evaluation indicators corresponding to the strategic keywords based on the effective priority; the target evaluation indicators are a part of a plurality of preset evaluation indicators; The weight adjustment module is used to correct the weights of each target evaluation indicator according to the quantitative adjustment parameters, and to normalize the weights of all preset evaluation indicators to obtain a target weight allocation scheme. The quantitative evaluation module is used to obtain the original evaluation data of each supplier, combine it with the target weight allocation scheme, perform quantitative evaluation on each candidate supplier, and output the evaluation results.

[0012] Beneficial Effects: This application provides a supplier evaluation method and system that quantifies the strategic intentions of senior management by acquiring target text data containing corporate strategic guidance and extracting strategic keywords and their corresponding effective priorities. Based on this, quantitative adjustment parameters for target evaluation indicators are determined according to the effective priorities, and the weights of each target evaluation indicator are corrected, achieving dynamic adjustment of the evaluation indicator weights to align with the company's latest strategic needs. Finally, the adjusted target weight allocation scheme is used to quantitatively evaluate suppliers and output the evaluation results. This method effectively solves the problem in existing technologies where traditional evaluation systems rely on preset fixed weights, making it difficult to flexibly adapt to new business priorities, leading to a disconnect between evaluation results and actual strategic needs. By transforming corporate strategic intentions into quantifiable evaluation indicator weight adjustment parameters, this application overcomes the drawbacks of inefficiency and inconsistency in manually adjusting evaluation reports, as well as the large differences in correction results caused by non-quantitative guidance documents. It significantly improves the accuracy, efficiency, and reliability of supplier evaluation, providing companies with a standardized and reliable panoramic view of suppliers and unified data support for subsequent procurement decisions. Attached Figure Description

[0013] Figure 1A flowchart of a supplier evaluation method provided for this application.

[0014] Figure 2 This is a schematic diagram of the structure of a supplier evaluation system provided in this application.

[0015] Labeling Explanation: 1. Text Acquisition Module; 2. Information Extraction Module; 3. Parameter Determination Module; 4. Weight Adjustment Module; 5. Quantitative Evaluation Module. Detailed Implementation

[0016] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0017] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0018] refer to Figure 1 This application proposes a supplier evaluation method, which includes the following steps: A1. Obtain target text data containing the company's strategic guiding intent; A2. Extract strategic keywords and their corresponding effective priorities from the target text data; A3. Based on the effective priority, determine the quantitative adjustment parameters of the target evaluation indicators corresponding to the strategic keywords; the target evaluation indicators are a part of a plurality of preset evaluation indicators; A4. Based on the quantitative adjustment parameters, correct the weights of each target evaluation indicator and normalize the weights of all preset evaluation indicators to obtain the target weight allocation scheme. A5. Obtain the original evaluation data of each supplier, combine it with the target weight allocation scheme, perform quantitative evaluation of each candidate supplier, and output the evaluation results.

[0019] This application achieves rapid response and precise adaptation of the supplier evaluation system to changes in corporate strategy by dynamically acquiring the company's strategic guidance and transforming it into quantifiable evaluation indicator weighting parameters. This method effectively solves the problem of the disconnect between traditional system evaluation results and actual strategic needs, improving the accuracy, efficiency, and consistency of the evaluation, and providing reliable procurement decision support for enterprises.

[0020] "Target textual data" refers to textual information that contains the company's strategic guiding intent. This information typically comes from various strategic documents issued internally by the company, such as annual reports, strategic plans, and procurement policies. This textual data is the core basis for understanding the company's current strategic direction and priorities.

[0021] "Strategic keywords" refer to words or phrases extracted from target text data that reflect a company's strategic priorities and focus, such as "cost-effectiveness," "risk mitigation," and "innovation capability." These keywords serve as a bridge connecting a company's strategic intentions with supplier evaluation metrics.

[0022] "Effective priority" refers to the importance of strategic keywords in the current corporate strategy. It not only considers the initial importance of the keywords themselves, but also dynamically adjusts them by combining their contextual semantics, timeliness and other factors, so as to more accurately reflect the real-time changes in corporate strategy.

[0023] "Target evaluation indicators" refer to pre-set evaluation indicators whose weights need to be adjusted due to the influence of strategic keywords during the supplier evaluation process. These indicators are components of the supplier evaluation system, such as "product quality," "on-time delivery rate," and "service level."

[0024] "Quantitative adjustment parameters" refer to the numerical values ​​used to adjust the weights of target evaluation indicators based on the effective priority of strategic keywords. These parameters ensure the objectivity and quantifiability of weight adjustments.

[0025] "Pre-set evaluation indicators" refer to a series of evaluation dimensions that companies set in advance when evaluating suppliers, covering all aspects of supplier performance.

[0026] The “target weight allocation scheme” refers to the final weight allocation result of all preset evaluation indicators after the strategic intent is adjusted. It reflects the degree of importance that the company attaches to each evaluation dimension under the current strategy.

[0027] "Raw assessment data" refers to the raw data collected from each supplier to evaluate their performance on various preset assessment indicators.

[0028] "Quantitative evaluation" refers to the process of calculating and analyzing the supplier's original evaluation data in conjunction with the target weight allocation scheme to obtain its comprehensive evaluation score.

[0029] The supplier evaluation method of this application achieves dynamic evaluation of suppliers through a series of steps.

[0030] In step A1, it is necessary to acquire target text data containing the company's strategic guiding intent. This step can be achieved manually, with operators collecting internally released strategic documents, such as annual reports, strategic plans, and procurement policies, and entering them into the system. Alternatively, the system can be integrated with the company's internal document management system, automatically capturing and importing the latest strategic documents through pre-defined interfaces or web crawling technology. For example, the system can periodically scan specified file directories or databases to identify and retrieve newly released strategic documents.

[0031] In step A2, strategic keywords and their corresponding effective priorities need to be extracted from the target text data. This step can be implemented manually, with domain experts or analysts reading the target text data, manually identifying the strategic keywords, and subjectively judging and assigning initial priorities based on their importance to the company's strategy. For example, experts could, based on experience, mark "cost-effectiveness" as a high priority and "innovation capability" as a medium priority. Alternatively, the system can utilize Natural Language Processing (NLP) technology to perform text analysis on the target text data. For example, the system can pre-set a strategic keyword library containing a series of words or phrases directly related to supplier evaluation indicators, such as "resilience," "stability," "cost-effectiveness," "rapid response," "risk mitigation," and "innovation capability," and pre-set an initial priority value for each keyword. The system then uses a matching algorithm to identify the strategic keywords in the text and obtain their corresponding initial priorities.

[0032] In step A3, based on the aforementioned effective priorities, it is necessary to determine the quantitative adjustment parameters of the target evaluation indicators corresponding to the aforementioned strategic keywords. These target evaluation indicators are a subset of multiple preset evaluation indicators. This step can be implemented manually, with evaluation experts determining the quantitative adjustment parameters for each target evaluation indicator based on the effective priorities of the strategic keywords, combined with their experience and understanding of the business. For example, if the effective priority of "risk mitigation" is high, the expert might manually set the quantitative adjustment parameter of the "supply chain stability" evaluation indicator to a positive value, indicating that its weight needs to be increased. Alternatively, the system can preset a rule base that defines the mapping relationship between the effective priorities of strategic keywords and the quantitative adjustment parameters of the target evaluation indicators. For example, the rule base could stipulate that when the effective priority of "risk mitigation" reaches a certain threshold, the quantitative adjustment parameter of the "supply chain stability" evaluation indicator is automatically set to a preset value.

[0033] In step A4, the weights of each target evaluation indicator need to be adjusted according to the aforementioned quantitative adjustment parameters, and the weights of all preset evaluation indicators need to be normalized to obtain the target weight allocation scheme. The system can automatically perform weight adjustment and normalization. For example, the system receives the quantitative adjustment parameters output in step A3 and automatically applies them to the current weights of the corresponding target evaluation indicators to complete the weight adjustment. After adjustment, the system automatically calculates the sum of the weights of all preset evaluation indicators and divides the weight of each preset evaluation indicator by the sum to obtain the normalized weights, thereby forming the target weight allocation scheme.

[0034] In step A5, the system needs to acquire the original evaluation data for each supplier. Combining this data with the aforementioned target weight allocation scheme, the system performs a quantitative evaluation of each candidate supplier and outputs the evaluation results. For example, the system can integrate with an enterprise's internal ERP, SCM, or other systems to automatically extract supplier performance data as the original evaluation data. Then, the system combines this original evaluation data with the target weight allocation scheme, automatically performing a weighted comprehensive calculation to obtain a comprehensive evaluation score for each supplier. Finally, the system can rank the suppliers based on their comprehensive evaluation scores and generate an evaluation report containing the ranking results and individual score values ​​for each item.

[0035] The core innovation of this application lies in providing a mechanism for translating corporate strategic guidance into quantifiable and actionable supplier evaluation indicator weighting adjustments. Compared to the closest existing technology, namely traditional automated evaluation systems that rely on preset fixed evaluation indicator weights, this application represents a significant advancement.

[0036] Traditional systems fail to adapt automatically to drastic changes in the global economic environment and market conditions, such as when companies need to prioritize "supply stability" and "risk resilience." This results in suppliers performing well on these new priority indicators without a corresponding increase in their overall score. Conversely, suppliers with lower prices but fragile supply chains may receive higher scores because the old weighting still favors cost factors. This discrepancy between the assessment results and the company's current strategic needs and actual business experience forces auditors to make extensive manual adjustments, which is not only inefficient but also leads to inconsistent assessment results due to subjective judgment differences.

[0037] This application effectively solves the above-mentioned problems by introducing steps A1 to A4. Specifically: 1. Dynamic Acquisition and Transformation of Strategic Intent: This application acquires target text data containing the enterprise's strategic guiding intent through step A1, and extracts strategic keywords and their effective priorities from it through step A2. This enables the system to proactively perceive and understand changes in enterprise strategy, something traditional systems cannot do at all. For example, when an enterprise releases a strategic document emphasizing "supply chain resilience," this application can identify the strategic keyword "resilience" and assign it a high effective priority, thus providing a basis for subsequent weight adjustments.

[0038] 2. Automatic Generation of Quantitative Adjustment Parameters: In step A3, this application automatically determines the quantitative adjustment parameters of the target evaluation indicators based on the effective priority. This replaces the traditional method of relying on manual subjective judgment for weight adjustment. For example, when the effective priority of "risk mitigation" increases, the system can automatically calculate the weight value that needs to be increased for the "supply chain stability" evaluation indicator, ensuring the objectivity and consistency of the adjustment.

[0039] 3. Dynamic Adjustment and Normalization of Evaluation Weights: In step A4, this application automatically adjusts the weights of each target evaluation indicator based on the quantitative adjustment parameters and performs normalization processing to obtain the target weight allocation scheme. This ensures that the evaluation weights are always consistent with the company's latest strategic intentions, avoiding the problem of traditional system evaluation results being out of touch with actual strategic needs. For example, when the company's strategy shifts and places greater emphasis on "rapid response," the weight of this indicator will automatically increase, allowing suppliers who perform well in rapid response to receive higher overall evaluation scores.

[0040] Through the aforementioned innovations, the supplier evaluation method proposed in this application provides a dynamic, accurate, and strategically aligned evaluation result, significantly improving the efficiency, accuracy, and reliability of the evaluation. This not only reduces the need for manual intervention and lowers operating costs but also provides enterprises with more reliable data support, enabling them to make more informed purchasing decisions and maintain a competitive edge in a rapidly changing market environment.

[0041] In some implementations, step A1 includes: A101. Obtain strategic documents published within the enterprise and identify the department that published the strategic documents, the document type, and the publication time; A102. Based on the publishing department, document type, and publishing time, and in conjunction with preset document priority rules, prioritize the strategic documents to obtain a sorted list of strategic documents; A103. Based on the sorted list of strategic documents and in conjunction with preset document filtering rules, filter out target text data that contains the company's strategic guiding intent.

[0042] In step A101, strategic documents issued internally by the company can be understood as various formal or informal texts generated during the company's daily operations and strategic planning, such as annual reports, strategic planning documents, departmental work plans, meeting minutes, high-level speeches, and policy documents. Identifying the issuing department, document type, and issuance date of these documents aims to provide basic information for subsequent priority assessment and screening. The level of the issuing department usually reflects the document's authority and strategic importance, the document type indicates its content nature and importance, and the recentity of the issuance date relates to its timeliness and current relevance.

[0043] Furthermore, in step A102, the document priority rule can comprehensively evaluate strategic documents based on factors such as the publishing department's level, the importance of the document type, and the recentity of the publication time. For example, the document priority rule can prioritize strategic documents based on the publishing department's level, the importance of the document type, and the recentity of the publication time. Strategic planning documents issued by senior management departments (such as the board of directors or the CEO's office) typically have a higher priority than daily work reports issued by lower-level departments. At the same time, documents with more recent publication times are likely to reflect the current strategic guidance intentions of the enterprise more accurately and timely. Specifically, corresponding level values ​​can be pre-configured for different departments, importance scores can be configured for different document types, and recentity values ​​can be configured for different time intervals between the publication time and the current time, forming corresponding lookup tables. In practice, these lookup tables are consulted to obtain level values, importance scores, and recentity values. The weighted sum of these three values ​​is then calculated to obtain a priority score. Finally, strategic documents are ranked according to the priority score. These rules allow for the effective sorting of massive amounts of strategic documents, resulting in a sorted list of strategic documents arranged according to importance and timeliness.

[0044] Specifically, in step A103, the document screening rules filter the sorted list of strategic documents based on their priority and content relevance. For example, the document screening rules can filter strategic documents based on their priority scores and content relevance. Content relevance refers to the degree to which the document content aligns with the company's strategic guiding intent. For instance, natural language processing technology can be used to perform semantic analysis on the document content, identifying words or phrases related to strategic elements such as the company's long-term development goals, core competitiveness, market positioning, and risk management. Combined with the document priority scores, high-priority documents with highly relevant content are selected to ensure that the selected target text data accurately and comprehensively reflects the company's strategic guiding intent.

[0045] This application's solution achieves precise capture of corporate strategic guidance intentions by refining the acquisition of target text data (step A1) into a series of steps including acquisition, identification, sorting, and filtering. First, by acquiring and identifying the metadata of strategic documents, a foundation is laid for subsequent evaluation. Second, by sorting documents using pre-defined document priority rules, priority processing is ensured for more important and timely strategic information. Finally, filtering is performed based on document priority and content relevance, effectively eliminating redundant information and non-core content, thereby ensuring that the acquired target text data is highly focused on the company's core strategic guidance intentions.

[0046] In some implementations, step A2 includes: A201. Identify strategic keywords in the target text data using a pre-defined strategic keyword library, and determine the initial priority of the association of the strategic keywords; A202. Perform contextual semantic analysis on the strategic keywords, and determine the effective priority corresponding to the strategic keywords in conjunction with the initial priority.

[0047] Strategic keywords are descriptive or biased words or phrases directly related to supplier evaluation indicators, such as resilience, stability, cost-effectiveness, rapid response, risk resistance, and innovation capability. The strategic keyword library can be understood as a structured data table storing strategic keywords, initial priority values ​​associated with those keywords, and a list of target evaluation indicators influenced by the strategic keywords. By utilizing this pre-defined strategic keyword library, strategic keywords closely related to the company's strategic intent can be efficiently identified from target text data, and an initial priority can be assigned to them.

[0048] Furthermore, when determining the effective priorities corresponding to strategic keywords, contextual semantic analysis is required for the identified strategic keywords. Contextual semantic analysis aims to understand the specific context and meaning of strategic keywords within the target text data, thereby more accurately assessing their importance. By combining the initial priority of strategic keywords with their semantic information in a specific context, a more precise and reflective effective priority can be determined.

[0049] This application's solution first utilizes a pre-set strategic keyword library for initial identification and priority assignment, ensuring the breadth and efficiency of strategic keyword extraction. Subsequently, by performing contextual semantic analysis on these keywords, it is possible to gain a deeper understanding of their specific weight and tendency within the company's strategic guiding intent, thereby avoiding misjudgments or priority biases that may arise from relying solely on a static keyword library. This ensures that the extracted strategic keywords and their effective priorities more accurately reflect the company's current strategic priorities and direction.

[0050] In practical applications, the semantics of strategic keywords can be significantly influenced by their domain-specific and ambiguous expressions. Failure to fully consider these subtle semantic differences and relying solely on general contextual semantic analysis may result in effective priorities that fail to accurately reflect the nuances and actual importance of the company's strategic guidance, thereby affecting the accuracy of subsequent evaluation indicator weight adjustments and supplier assessments. To address this, this application further proposes specific steps for determining the effective priorities corresponding to strategic keywords, aiming to improve the accuracy and robustness of priority determination through in-depth analysis of contextual information.

[0051] Specifically, step A202 includes: Obtain the contextual information of the strategic keywords in the target text data; the contextual information includes the modifiers, verbs, and sentence structure of the strategic keywords; Based on the contextual information, identify the domain-specific and ambiguous expressions of the strategic keywords; Based on the domain-specific expression and combined with preset domain semantic rules, the initial priority of the strategic keywords is adjusted in a domain-adaptive manner to obtain the domain-adjusted priority; Based on the fuzzy expression and combined with the preset fuzzy quantization rules, the priority of the domain is adjusted by fuzzy quantization to obtain an effective priority.

[0052] Among these, acquiring the contextual information of strategic keywords within the target text data refers to using natural language processing techniques to extract relevant vocabulary and grammatical structures of the strategic keywords within their respective sentences or paragraphs. Specifically, modifiers are words that directly modify or limit the strategic keywords, such as adjectives and adverbs, which reveal the degree or nature of the strategic keywords; verbs refer to actions or states related to the strategic keywords, reflecting their dynamic impact or objectives; and sentence structure refers to the grammatical position of the strategic keywords within the sentence and their relationship with other components, such as subject-verb-object structures and parallel structures. These structures help in understanding the emphasis or logical relationships of the strategic keywords. By comprehensively acquiring this contextual information, rich and accurate input can be provided for subsequent semantic analysis.

[0053] Furthermore, based on the acquired contextual information, we identify domain-specific and ambiguous expressions of strategic keywords. Domain-specific expressions refer to the unique meanings or emphases of strategic keywords within a specific industry or business area, which may differ from their meaning in a general context. For example, the keyword "efficiency" in manufacturing may emphasize production line throughput and cost control, while in R&D it may emphasize innovation cycle and resource utilization. Ambiguous expressions refer to the meaning or importance of strategic keywords not being clearly stated or having a certain degree of uncertainty in the text, such as using words like "possibly," "potentially," or "moderately" for modification. Identifying these expressions helps to more precisely understand the actual intent and scope of influence of strategic keywords.

[0054] Based on this, and according to the identified domain-specific expressions, combined with pre-defined domain semantic rules, the initial priority of strategic keywords is adjusted in a domain-adaptive manner to obtain a domain-adjusted priority. Domain semantic rules are a set of pre-defined rules based on the characteristics and strategic emphasis of different business domains, used to guide how to adjust the priority of strategic keywords according to their domain-specific meaning. For example, if the strategic keyword "innovation" is identified as a domain-specific expression in the R&D department's strategic documents, and the domain semantic rules stipulate that the R&D department should prioritize "innovation" higher than other departments, then its initial priority will be increased accordingly. This adjustment ensures that the priority of strategic keywords matches their actual importance in specific business domains.

[0055] Simultaneously, based on the identified ambiguous expressions and combined with preset fuzzy quantization rules, the domain adjustment priority is fuzzified to obtain the final effective priority. Fuzzy quantization rules are a series of computational methods used to handle uncertainty or fuzzy information, such as fuzzy logic and membership functions. When the expression of strategic keywords is ambiguous, such as "appropriately improve cost-effectiveness," the fuzzy quantization rules can refine the numerical conversion of the domain adjustment priority according to the degree of fuzziness, thereby obtaining a more discriminative effective priority. Through this processing, even if there are ambiguous expressions in the strategic text, they can be converted into quantifiable priority values, improving the accuracy and practicality of priority determination.

[0056] This application's solution, through in-depth analysis of the contextual information of strategic keywords, enables a more comprehensive and accurate understanding of their true meaning and importance. Specifically, by acquiring contextual information such as modifiers, verbs, and sentence structures, subtle semantic differences in strategic keywords can be captured. Based on this, domain-specific and ambiguous expressions are identified, allowing the system to distinguish the specific meanings of strategic keywords in different business domains and the uncertainties in their expression. It is precisely the identification of these semantic features that makes subsequent domain-adaptive adjustments and fuzzy quantification possible. Domain-adaptive adjustments ensure that the priority of strategic keywords aligns with their actual strategic emphasis in specific business domains, avoiding deviations caused by general semantic understanding. Fuzzy quantification effectively transforms the inherent ambiguity in the text into actionable quantitative values, overcoming the limitations of traditional methods in handling ambiguous information. Through this series of refined processing steps, this application's solution can closely integrate the initial priority with the actual context and semantic depth of the strategic keywords, thereby generating a more accurate and instructive effective priority.

[0057] In practical applications, corporate strategic guidance can change over time, and keywords in earlier strategic texts may gradually lose their relevance to current supplier evaluations. Failure to consider the time dimension of the effective priority of strategic keywords may result in evaluation results that fail to fully reflect the company's latest strategic direction.

[0058] In some implementations, after step A202, the following steps are also included: A203. Obtain the publication time of the target text data, and calculate the time span between the publication time and the current evaluation time. A204. Determine the attenuation factor of the strategic keywords based on the time span; A205. The effective priority is corrected using the attenuation factor.

[0059] Specifically, in step A203, the publication time of the target text data refers to the date on which the text data is officially published or takes effect. The current evaluation time point refers to the specific date when the supplier evaluation operation is carried out. The time span can be understood as the length of time between the publication time and the current evaluation time point, and its purpose is to quantify the timeliness of strategic keywords. For example, the time span can be calculated in units of days, weeks, months, or years.

[0060] In step A204, the decay factor is a multiplier used to adjust the effective priority of strategic keywords. Its value is typically between 0 and 1, and it decreases as time progresses. The decay factor can be determined based on a pre-defined decay model, such as a linear decay model, an exponential decay model, or a logarithmic decay model. Its purpose is to reflect the trend of the strategic keywords' importance gradually decreasing over time.

[0061] In practical applications, step A205, which involves correcting the effective priority using the attenuation factor, is typically achieved by multiplying the original effective priority by the attenuation factor. The corrected effective priority will more accurately reflect the actual importance of the strategic keywords at the current point in time.

[0062] This application's solution addresses the problem of strategic keyword priority losing its timeliness over time in traditional methods by introducing a time dimension to adjust the effective priority of strategic keywords. Specifically, by obtaining the publication time of the target text data and calculating the time span between it and the current evaluation time, the timeliness of strategic keywords can be quantified. Given the time span, a decay factor can be determined, reflecting the trend of the strategic keyword's importance gradually decreasing over time. Finally, this decay factor is used to adjust the previously determined effective priority, making the adjusted effective priority more accurately reflect the actual guiding significance of the strategic keywords within the current evaluation period. It is precisely because of this time-sensitive adjustment that the priority of strategic keywords can dynamically adapt to changes in corporate strategy, avoiding the risk of using outdated information for evaluation.

[0063] In some implementations, step A3 includes: A301. Obtain a list of target evaluation indicators associated with the strategic keywords, and the correlation between the strategic keywords and each target evaluation indicator in the list of target evaluation indicators; the correlation includes the direction of correlation and the strength of correlation; A302. Determine the initial adjustment parameters for each target evaluation indicator based on the effective priority and the correlation. A303. Based on the initial adjustment parameters and the current weights of each of the target evaluation indicators, and in conjunction with preset weight adjustment constraints, determine the quantitative adjustment parameters of each of the target evaluation indicators; the weight adjustment constraints include an upper limit and a lower limit for weight adjustment.

[0064] Step A301 aims to clarify which specific evaluation indicators are affected by the strategic keywords, and the specific nature of this impact. Specifically, the target evaluation indicator list refers to a subset of pre-defined evaluation indicators that are directly associated with a particular strategic keyword and require weight adjustment. The association describes the interaction between the strategic keyword and these target evaluation indicators; it can be understood as a mapping rule or function. The association direction indicates the trend of the impact of changes in the priority of the strategic keyword on the weight adjustment of the target evaluation indicators. For example, when the priority of a strategic keyword increases, does the weight of a certain target evaluation indicator increase (positive association) or decrease (negative association)? The association strength quantifies the degree of this impact; for example, it can be a numerical value representing the strength of the impact. The target evaluation indicator list can be directly extracted from the strategic keyword library, and the association can be obtained by querying a pre-defined association table, which records the associations between various strategic keywords and various pre-defined evaluation indicators.

[0065] Further, step A302 determines the initial adjustment parameters for each target evaluation indicator based on effective priority and correlation. Effective priority is a quantified value of the strategic keyword's importance, which can be obtained from step A2. The initial adjustment parameters can be determined using a weighted summation or product model. For example, for a specific target evaluation indicator, its initial adjustment parameter can be calculated as: Initial Adjustment Parameter = Effective Priority × Correlation Strength × Correlation Direction Factor. Here, the correlation direction factor is 1 for positive correlation and -1 for negative correlation. In this way, the importance of the strategic keyword and its specific impact on the evaluation indicator are comprehensively considered, thus obtaining a preliminary adjustment value.

[0066] Furthermore, in step A303, the determination of the quantitative adjustment parameters is based on the initial adjustment parameters, combined with the current weights of the target evaluation indicators and preset weight adjustment constraints. The current weight refers to the weight value of the target evaluation indicator before this strategic adjustment. Weight adjustment constraints are rules set to ensure the rationality and stability of weight adjustments. They can include upper and lower limits for weight adjustment, used to limit the maximum magnitude of weight adjustment for a single evaluation indicator or the range of its final weight value, to avoid unreasonable or extreme situations. For example, based on the initial adjustment parameters, the unconstrained adjusted weights are first calculated (e.g., by adding the current weights to the initial adjustment parameters to obtain the unconstrained adjusted weights), and then it is checked whether these weights exceed the preset upper and lower limits for weight adjustment. If they do, the weights are limited within the constraints, and the excess portion is proportionally allocated to other indicators that have not exceeded the constraints, ensuring that the total weight of all indicators remains unchanged. Thus, the increment of the constrained weight relative to the current weight serves as the quantitative adjustment parameter. By introducing current weights, we can avoid over-adjusting indicators that are already highly important; through weight adjustment constraints, we can ensure the rationality of the adjustments and prevent indicator weights from being adjusted to unrealistic ranges, thereby ensuring the stability and effectiveness of the evaluation system.

[0067] This application's solution first identifies the correlation between strategic keywords and specific target evaluation indicators, including the direction and strength of the correlation, thus providing a clear basis for subsequent weight adjustments. Subsequently, utilizing the effective priority of strategic keywords and the established correlations, initial adjustment parameters for each target evaluation indicator can be preliminarily calculated, allowing strategic intent to be quantitatively mapped onto the evaluation indicators. Finally, by considering the current weights of the target evaluation indicators and combining them with preset weight adjustment constraints, the initial adjustment parameters are corrected, resulting in more reasonable and operable quantitative adjustment parameters. This series of steps ensures the refinement and controllability of the weight adjustment process.

[0068] In some implementations, step A4 includes: A401. Use the aforementioned quantitative adjustment parameters to correct the weights of each target evaluation indicator; A402. After the correction is completed, for each preset evaluation index, the weight of the preset evaluation index is divided by the sum of the weights of all preset evaluation indexes to obtain the normalized weight. A403. The target weight allocation scheme is composed of the normalized weights of all preset evaluation indicators.

[0069] Specifically, in step A401, quantitative adjustment parameters are used to correct the weights of the target evaluation indicators. These parameters are determined based on the effective priority of strategic keywords and their correlation with the target evaluation indicators, aiming to reflect the impact of the company's strategic guidance on the importance of specific evaluation indicators. For example, if a strategic keyword (such as "resilience") is given a high priority and has a strong positive correlation with the target evaluation indicator "supply chain resilience," the quantitative adjustment parameter will increase the weight of "supply chain resilience." The correction process can employ addition, multiplication, or other preset weight adjustment algorithms to ensure that the corrected weights accurately reflect the strategic orientation.

[0070] Further, in step A402, after correcting the weights of the target evaluation indicators, the weights of all preset evaluation indicators need to be normalized. The purpose of normalization is to ensure that the sum of the weights of all evaluation indicators is 1 (or 100%), thereby ensuring that the weight allocation of each indicator is reasonable and comparable in the subsequent supplier quantitative evaluation. Specifically, for each preset evaluation indicator, its corrected weight will be divided by the sum of the weights of all preset evaluation indicators. For example, if the sum of the corrected weights of all preset evaluation indicators is S, and the corrected weight of a certain preset evaluation indicator is W_i, then its normalized weight W'_i = W_i / S.

[0071] Therefore, in step A403, the normalized weights of all preset evaluation indicators together constitute the target weight allocation scheme. This scheme is a complete set of weights, which includes the relative importance of all preset evaluation indicators under the current corporate strategic guidance. This target weight allocation scheme will be used for subsequent supplier quantitative evaluation to ensure that the evaluation results are consistent with the company's strategic objectives.

[0072] This application's solution first adjusts the weights of target evaluation indicators using quantitative adjustment parameters, allowing indicators highly relevant to the company's strategic guidance to receive higher weights and thus occupy a more important position in the evaluation system. Subsequently, by normalizing the weights of all preset evaluation indicators, the rationality of the weight allocation and the consistency of the total weight are ensured, avoiding imbalances in the overall evaluation system caused by adjustments to local weights. Therefore, this solution ensures that the final target weight allocation not only reflects the company's strategic priorities but is also mathematically rigorous and feasible, providing a solid foundation for subsequent supplier evaluation.

[0073] In some implementations, step A5 includes: A501. Obtain the original evaluation data for each supplier corresponding to each preset evaluation indicator; A502. Based on the original evaluation data, determine the individual score value of the corresponding preset evaluation indicator; A503. Using the weights of each preset evaluation indicator in the target weight allocation scheme, the individual score values ​​are weighted and calculated to obtain the comprehensive evaluation score of each supplier. A504. Sort the suppliers in descending order according to the comprehensive evaluation score; A505. The output includes the ranking results and the evaluation results of each supplier's individual score.

[0074] Step A501 aims to collect basic information for supplier evaluation. Raw evaluation data can come from various sources, such as supplier qualification documents, historical transaction records, third-party evaluation reports, and market research data. This data is typically heterogeneous, potentially including qualitative descriptions, quantitative figures, documents, images, and other formats. Obtaining this data is a prerequisite for subsequent quantitative evaluation.

[0075] Further, step A502 involves converting the acquired raw evaluation data into quantifiable individual score values. This typically requires adherence to pre-defined evaluation criteria and scoring rules. For example, for the "on-time delivery rate" metric, the raw data might be on-time delivery records of historical orders; by calculating the proportion of on-time deliveries, it can be converted into a score value of 0-100. For the "quality pass rate" metric, the raw data might be quality inspection reports; by calculating the ratio of qualified products to total production quantity, it can be converted into a score value. The purpose of this step is to unify raw data of different types and dimensions into a comparable numerical form.

[0076] Based on this, step A503 uses the target weight allocation scheme obtained in step A4 to perform a weighted comprehensive calculation of the individual scores of each preset evaluation indicator. The target weight allocation scheme reflects the degree of importance that the company's strategic guidance places on different evaluation indicators. By multiplying each individual score by its corresponding weight and then summing all the weighted scores, the comprehensive evaluation score for each supplier can be obtained. This weighted calculation ensures that the evaluation results accurately reflect the company's strategic orientation.

[0077] Subsequently, step A504 ranks the suppliers in descending order based on their overall evaluation scores. A higher overall evaluation score indicates a supplier's superior performance in meeting the company's strategic requirements. This ranking clearly identifies the best-performing suppliers, providing a clear basis for procurement decisions.

[0078] Finally, step A505 is responsible for outputting the evaluation results. The output includes not only the supplier rankings but also individual scores for each supplier on each pre-defined evaluation indicator. This detailed output helps decision-makers gain a comprehensive understanding of supplier performance across various aspects, rather than just a final overall ranking, thus supporting more refined decision analysis.

[0079] This application's solution ensures the systematic, transparent, and operational nature of the supplier quantitative evaluation process by detailing it into a series of clearly defined steps, including data acquisition, individual score conversion, weighted comprehensive calculation, ranking, and result output. Specifically, step A501 acquires comprehensive raw evaluation data, laying the foundation for subsequent quantitative analysis; step A502 unifies heterogeneous raw data into comparable individual score values, resolving the issue of inconsistent data dimensions; step A503 performs weighted comprehensive calculations using a target weight allocation scheme, ensuring that the evaluation results accurately reflect the company's strategic orientation and avoiding the subjectivity and arbitrariness of weight allocation in traditional evaluations; step A504 ranks suppliers, providing decision-makers with an intuitive comparison of strengths and weaknesses; finally, step A505 outputs evaluation results containing detailed individual scores, providing not only a final ranking but also revealing the supplier's performance on specific indicators, thereby supporting more in-depth analysis and decision-making.

[0080] In practical applications, the original evaluation data from various suppliers may originate from different systems, have diverse data formats, and may vary in update frequency. Directly using these heterogeneous raw data for scoring without proper processing may compromise the accuracy and comparability of the evaluation results, thereby reducing the overall reliability of the evaluation.

[0081] Therefore, in some implementations, step A502 includes: Obtain the source system, data format, and update frequency of the original evaluation data; Based on the source system, data format, and update frequency of the original evaluation data, and in conjunction with preset data integration rules, the original evaluation data is extracted, integrated, and cleaned to obtain evaluation data in a unified format. Based on the unified format evaluation data, and combined with the preset evaluation indicator types and scoring conversion rules, the unified format evaluation data is subjected to dimension unification and numerical conversion to obtain the individual score value of the corresponding preset evaluation indicator.

[0082] Among these, obtaining the source system, data format, and update frequency of the original assessment data refers to identifying the basic attributes of the original assessment data. This can be achieved using a data source configuration management module. This module records and manages the source system (e.g., ERP system, CRM system, supplier portal website), data format (e.g., CSV, XML, JSON, database table structure), and update frequency (e.g., daily, weekly, real-time) of all original assessment data. Identifying this information provides the necessary foundation for subsequent data processing.

[0083] Based on the source system, data format, and update frequency of the original assessment data, and in conjunction with preset data integration rules, the original assessment data is extracted, integrated, and cleaned to obtain assessment data in a unified format. This process transforms data from different sources and in different formats into a unified standard format. Specifically, this can be achieved using a data integration engine. This engine, based on preset data integration rules (e.g., defining different extraction interfaces for data from different source systems; defining corresponding conversion templates for different data formats; setting different data synchronization strategies for different update frequencies), extracts (e.g., acquiring data through API interfaces, database connectors, etc.), integrates (e.g., merging data from different systems but describing the same supplier), and cleans (e.g., removing duplicate data, filling missing values, and correcting erroneous data) the original assessment data, ultimately generating structured, standardized, and unified format assessment data. For example, data integration rules can be based on the heterogeneity of data sources (e.g., differences in data structure between SAP and Oracle systems), the diversity of data formats (e.g., differences in field definitions between Excel and JSON files), and the differences in data update frequency (e.g., financial data is updated monthly, while logistics data is updated daily), performing data extraction, data transformation, and data loading on the original assessment data.

[0084] Based on standardized evaluation data and pre-defined evaluation indicator types and scoring conversion rules, the standardized evaluation data undergoes dimensional unification and numerical transformation to obtain the individual score value for the corresponding pre-defined evaluation indicator. This involves converting standardized data into comparable score values, which can be achieved using a scoring conversion module. This module, based on pre-defined evaluation indicator types (e.g., cost indicators, quality indicators, delivery indicators, etc.) and scoring conversion rules (e.g., cost indicators may use reciprocal conversion, meaning lower costs result in higher scores; quality indicators may use percentage conversion, meaning higher pass rates result in higher scores), performs dimensional unification (e.g., unifying cost data from different units to monetary units) and numerical transformation (e.g., mapping raw values ​​to a scoring range of 0-100) on the standardized evaluation data, thereby obtaining the individual score value for the corresponding pre-defined evaluation indicator. For example, the evaluation indicator types and scoring conversion rules can be standardized based on the type of evaluation indicator (e.g., qualitative and quantitative indicators) and data distribution characteristics (e.g., normal or skewed distribution), such as using Z-score standardization or Min-Max standardization.

[0085] These steps are logically interconnected. First, identifying the attributes of the raw data is fundamental to data processing, providing direction for subsequent data integration. Second, data integration and cleaning are crucial for addressing data heterogeneity; they transform the raw data into a unified format, laying the foundation for dimensional unification and numerical conversion. Finally, dimensional unification and numerical conversion ensure the comparability of data across different indicators, providing reliable input for the final calculation of individual scores. Through the synergistic effect of these steps, the heterogeneity, diversity, and timeliness of the original assessment data can be effectively addressed, ensuring the accuracy and consistency of the assessment results.

[0086] This application's solution effectively addresses the evaluation challenges posed by the heterogeneity of raw evaluation data by introducing a preprocessing step. Specifically, by acquiring the source system, data format, and update frequency of the raw evaluation data, the system can comprehensively understand the data's characteristics. Subsequently, based on preset data integration rules, the raw data is extracted, integrated, and cleaned. This process transforms data from different channels and with varying formats into a unified and standardized format, thereby eliminating structural differences and potential errors between data. Building upon this, and combining preset evaluation indicator types and scoring conversion rules, the data in the unified format undergoes dimensional unification and numerical conversion, ensuring the numerical comparability of different evaluation indicators. It is precisely because of these preprocessing steps that raw data, which was originally difficult to directly compare and score, can be accurately and consistently converted into individual score values ​​with unified dimensions and numerical ranges, providing high-quality input for subsequent weighted comprehensive calculations.

[0087] refer to Figure 2This application provides a supplier evaluation system, which includes: Text acquisition module 1 is used to acquire target text data containing the enterprise's strategic guidance intentions (for details, please refer to step A1 above). Information extraction module 2 is used to extract strategic keywords and their corresponding effective priorities from the target text data (for details, please refer to step A2 above). The parameter determination module 3 is used to determine the quantitative adjustment parameters of the target evaluation indicators corresponding to the strategic keywords based on the effective priority; the target evaluation indicators are a part of a plurality of preset evaluation indicators (the specific process can be referred to step A3 above). The weight adjustment module 4 is used to correct the weight of each target evaluation index according to the quantitative adjustment parameters, and to normalize the weight of all preset evaluation indexes to obtain the target weight allocation scheme (the specific process can be referred to step A4 above). The quantitative evaluation module 5 is used to obtain the original evaluation data of each supplier, combine it with the target weight allocation scheme, perform quantitative evaluation on each candidate supplier, and output the evaluation results (for details, please refer to step A5 above).

[0088] 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 supplier evaluation method, characterized in that, The method includes the following steps: A1. Obtain target text data containing the company's strategic guiding intent; A2. Extract strategic keywords and their corresponding effective priorities from the target text data; A3. Based on the effective priority, determine the quantitative adjustment parameters of the target evaluation indicators corresponding to the strategic keywords; the target evaluation indicators are a part of a plurality of preset evaluation indicators; A4. Based on the quantitative adjustment parameters, correct the weights of each target evaluation indicator and normalize the weights of all preset evaluation indicators to obtain the target weight allocation scheme. A5. Obtain the original evaluation data of each supplier, combine it with the target weight allocation scheme, perform quantitative evaluation of each candidate supplier, and output the evaluation results.

2. The supplier evaluation method according to claim 1, characterized in that, Step A1 includes: A101. Obtain strategic documents published within the enterprise and identify the department that published the strategic documents, the document type, and the publication time; A102. Based on the publishing department, document type, and publishing time, and in conjunction with preset document priority rules, prioritize the strategic documents to obtain a sorted list of strategic documents; A103. Based on the sorted list of strategic documents and in conjunction with preset document filtering rules, filter out target text data that contains the company's strategic guiding intent.

3. The supplier evaluation method according to claim 1, characterized in that, Step A2 includes: A201. Identify strategic keywords in the target text data using a pre-defined strategic keyword library, and determine the initial priority of the association of the strategic keywords; A202. Perform contextual semantic analysis on the strategic keywords, and determine the effective priority corresponding to the strategic keywords in conjunction with the initial priority.

4. The supplier evaluation method according to claim 3, characterized in that, Step A202 includes: Obtain the contextual information of the strategic keywords in the target text data; the contextual information includes the modifiers, verbs, and sentence structure of the strategic keywords; Based on the contextual information, identify the domain-specific and ambiguous expressions of the strategic keywords; Based on the domain-specific expression and combined with preset domain semantic rules, the initial priority of the strategic keywords is adjusted in a domain-adaptive manner to obtain the domain-adjusted priority; Based on the fuzzy expression and combined with the preset fuzzy quantization rules, the priority of the domain is adjusted by fuzzy quantization to obtain an effective priority.

5. A supplier evaluation method according to claim 3, characterized in that, Following step A202, the following is also included: A203. Obtain the publication time of the target text data, and calculate the time span between the publication time and the current evaluation time. A204. Determine the attenuation factor of the strategic keywords based on the time span; A205. The effective priority is corrected using the attenuation factor.

6. The supplier evaluation method according to claim 1, characterized in that, Step A3 includes: A301. Obtain a list of target evaluation indicators associated with the strategic keywords, and the correlation between the strategic keywords and each target evaluation indicator in the list of target evaluation indicators; the correlation includes the direction of correlation and the strength of correlation; A302. Determine the initial adjustment parameters for each target evaluation indicator based on the effective priority and the correlation. A303. Based on the initial adjustment parameters and the current weights of each of the target evaluation indicators, and in conjunction with preset weight adjustment constraints, determine the quantitative adjustment parameters of each of the target evaluation indicators; the weight adjustment constraints include an upper limit and a lower limit for weight adjustment.

7. The supplier evaluation method according to claim 1, characterized in that, Step A4 includes: A401. Use the aforementioned quantitative adjustment parameters to correct the weights of each target evaluation indicator; A402. After the correction is completed, for each preset evaluation index, the weight of the preset evaluation index is divided by the sum of the weights of all preset evaluation indexes to obtain the normalized weight. A403. The target weight allocation scheme is composed of the normalized weights of all preset evaluation indicators.

8. The supplier evaluation method according to claim 1, characterized in that, Step A5 includes: A501. Obtain the original evaluation data for each supplier corresponding to each preset evaluation indicator; A502. Based on the original evaluation data, determine the individual score value of the corresponding preset evaluation indicator; A503. Using the weights of each preset evaluation indicator in the target weight allocation scheme, the individual score values ​​are weighted and calculated to obtain the comprehensive evaluation score of each supplier. A504. Sort the suppliers in descending order according to the comprehensive evaluation score; A505. The output includes the ranking results and the evaluation results of each supplier's individual score.

9. A supplier evaluation method according to claim 8, characterized in that, Step A502 includes: Obtain the source system, data format, and update frequency of the original evaluation data; Based on the source system, data format, and update frequency of the original evaluation data, and in conjunction with preset data integration rules, the original evaluation data is extracted, integrated, and cleaned to obtain evaluation data in a unified format. Based on the unified format evaluation data, and combined with the preset evaluation indicator types and scoring conversion rules, the unified format evaluation data is subjected to dimension unification and numerical conversion to obtain the individual score value of the corresponding preset evaluation indicator.

10. A supplier evaluation system, characterized in that, The system includes: The text acquisition module is used to acquire target text data containing the company's strategic guiding intent; The information extraction module is used to extract strategic keywords and their corresponding effective priorities from the target text data; The parameter determination module is used to determine the quantitative adjustment parameters of the target evaluation indicators corresponding to the strategic keywords based on the effective priority; the target evaluation indicators are a part of a plurality of preset evaluation indicators; The weight adjustment module is used to correct the weights of each target evaluation indicator according to the quantitative adjustment parameters, and to normalize the weights of all preset evaluation indicators to obtain a target weight allocation scheme. The quantitative evaluation module is used to obtain the original evaluation data of each supplier, combine it with the target weight allocation scheme, perform quantitative evaluation on each candidate supplier, and output the evaluation results.

Images