Intelligent bid invitation decision-making method and device

Abstract

The embodiment of the invention provides an intelligent bid invitation decision-making method and device, and relates to the technical field of intelligent bid invitation, and the method specifically comprises the steps: obtaining a target latest environmental protection policy feature set and a target supply chain feature set corresponding to a bidding scheme; obtaining scores corresponding to a plurality of environmental protection indexes including a carbon emission index of the bidding scheme, so as to generate an environmental protection score corresponding to the bidding scheme; obtaining a technical score corresponding to the bidding scheme, and determining initial weight coefficients corresponding to the environmental protection score and the technical score of the bidding scheme based on a double-weight game model; adjusting the initial weight coefficients corresponding to the environmental protection score and the technical score based on the historical bidding behavior of a bidder corresponding to the bidding scheme, and generating target weight coefficients corresponding to the environmental protection score and the technical score; and weighting calculation is carried out to obtain a comprehensive score, so that a target bid winning scheme is screened out from the plurality of bidding schemes. According to the invention, the decision accuracy in the bid invitation process can be improved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of intelligent bidding, and in particular to an intelligent bidding decision-making method and device. Background Art

[0002] In modern business, bidding, as a widely used method for procurement and partner selection, plays a key role in various fields. Companies use bidding to select suitable partners from a large number of bidding companies. However, the current bidding process faces many challenges.

[0003] With growing environmental awareness, environmental assessments in bidding are crucial. However, existing environmental assessment methods are often static and formal, which can easily lead to inaccuracies and affect bidding results.

[0004] Therefore, how to improve the accuracy of decision-making in the bidding process has become an urgent problem to be solved. Summary of the Invention

[0005] In view of this, the embodiments of the present application provide an intelligent bidding decision-making method and device, which can improve the decision-making accuracy in the bidding process.

[0006] In a first aspect, an embodiment of the present application provides an intelligent bidding decision-making method, comprising:

[0007] For any one of the multiple bidding schemes, obtaining a target latest environmental protection policy feature set and a target supply chain feature set corresponding to the bidding scheme;

[0008] Combining the target latest environmental policy feature set and the target supply chain feature set, obtaining scores corresponding to multiple environmental indicators of the bidding scheme to generate an environmental score corresponding to the bidding scheme; the multiple environmental indicators include a carbon emission indicator;

[0009] Obtaining the technical score corresponding to the bidding scheme, and determining the initial weight coefficients corresponding to the environmental score and the technical score of the bidding scheme respectively based on a double-weighted game model;

[0010] Adjusting the initial weight coefficients corresponding to the environmental protection score and the technical score based on the historical bidding behavior of the bidder corresponding to the bidding scheme to generate target weight coefficients corresponding to the environmental protection score and the technical score;

[0011] The environmental protection score and the technical score are weighted according to the target weight coefficient to obtain a comprehensive score, and based on the comprehensive scores corresponding to the various bidders, a target winning bid is screened out from multiple bidding schemes.

[0012] As an optional implementation of the embodiment of the present application, for any one of the multiple bidding schemes, obtaining the target latest environmental protection policy feature set and target supply chain feature set corresponding to the bidding scheme includes:

[0013] Obtaining the latest industry policy text corresponding to the bidding scheme, and extracting key features from the latest industry policy text to generate the target latest environmental protection policy feature set;

[0014] The supply chain data of the bidder corresponding to the bidding scheme is obtained, and features are extracted from the supply chain data to generate the target supply chain feature set.

[0015] As an optional implementation of the embodiment of the present application, the combination of the target latest environmental policy feature set and the target supply chain feature set to obtain scores corresponding to multiple environmental indicators of the bidding scheme to generate an environmental score corresponding to the bidding scheme includes:

[0016] Based on the target latest environmental policy feature set and the target supply chain feature set, score the multiple environmental indicators of the bidding scheme respectively to obtain the initial score corresponding to each environmental indicator;

[0017] According to the weight coefficient corresponding to each environmental protection indicator, the initial score corresponding to each environmental protection indicator is weightedly calculated to obtain the environmental protection score corresponding to the bidding scheme.

[0018] As an optional implementation of the embodiment of the present application, before performing weighted calculation on the initial scores corresponding to the respective environmental indicators according to the weight coefficients corresponding to the respective environmental indicators to obtain the environmental score corresponding to the bidding scheme, the method further includes:

[0019] Based on the latest environmental protection policy data characteristics of the target, the weight coefficients corresponding to the various environmental protection indicators are adjusted and updated.

[0020] As an optional implementation of the embodiment of the present application, the multiple environmental protection indicators include a carbon emission indicator;

[0021] The carbon emission index is obtained based on the target carbon emission factor corresponding to the bidding scheme.

[0022] As an optional implementation of the embodiment of the present application, the method further includes:

[0023] Obtain the bill of materials and process flow chart in the bidding proposal;

[0024] Calculating a first carbon emission factor based on the bill of materials, and calculating a second carbon emission factor based on the energy consumption nodes in the process flow chart;

[0025] The first carbon emission factor and the second carbon emission factor are added to obtain a target carbon emission factor.

[0026] As an optional implementation of the embodiment of the present application, after performing weighted calculation on the environmental protection score and the technical score according to the target weight coefficient to obtain a comprehensive score, and screening a target winning bid from multiple bidding schemes based on the comprehensive scores corresponding to the bidders, the method further includes:

[0027] Based on the state-action space algorithm, the target weight coefficients of the technical score and the environmental score are used as the state space, the weight adjustment direction is used as the action space, and the dual-weight game model is optimized and trained according to the reward value; the reward value is set according to the impact of the target bidder on the long-term environmental protection goals after executing the corresponding bidding plan.

[0028] In a second aspect, an embodiment of the present application provides an intelligent bidding decision-making device, comprising:

[0029] A first acquisition unit is configured to acquire, for any one of the multiple bidding schemes, a target latest environmental protection policy feature set and a target supply chain feature set corresponding to the bidding scheme;

[0030] a generating unit, configured to obtain scores corresponding to a plurality of environmental protection indicators of the bidding scheme, respectively, by combining the target latest environmental protection policy feature set and the target supply chain feature set, so as to generate an environmental protection score corresponding to the bidding scheme; the plurality of environmental protection indicators including a carbon emission indicator;

[0031] A second acquisition unit is configured to acquire a technical score corresponding to the bidding scheme, and determine initial weight coefficients corresponding to the environmental score and the technical score of the bidding scheme respectively based on a double-weighted game model;

[0032] an adjusting unit, configured to adjust the initial weight coefficients corresponding to the environmental protection score and the technical score respectively based on the historical bidding behavior of the bidder corresponding to the bidding scheme, and generate target weight coefficients corresponding to the environmental protection score and the technical score respectively;

[0033] A screening unit is used to perform weighted calculation on the environmental protection score and the technical score according to the target weight coefficient to obtain a comprehensive score, and screen out a target winning bid from multiple bidding schemes based on the comprehensive scores corresponding to the respective bidders.

[0034] As an optional implementation of an embodiment of the present application, the first acquisition unit is specifically used to obtain the latest industry policy text corresponding to the bidding scheme, and perform key feature extraction on the latest industry policy text to generate the target latest environmental protection policy feature set; obtain the supply chain data of the bidder corresponding to the bidding scheme, and perform feature extraction on the supply chain data to generate the target supply chain feature set.

[0035] As an optional implementation of an embodiment of the present application, the generation unit is specifically used to score the multiple environmental indicators of the bidding scheme in combination with the target latest environmental policy feature set and the target supply chain feature set, and obtain the initial score corresponding to each environmental indicator; according to the weight coefficient corresponding to each environmental indicator, the initial score corresponding to each environmental indicator is weightedly calculated to obtain the environmental score corresponding to the bidding scheme.

[0036] As an optional implementation of the embodiment of the present application, the intelligent bidding decision-making device also includes an updating unit, which is specifically used to adjust and update the weight coefficients corresponding to the various environmental protection indicators based on the latest environmental protection policy data characteristics of the target.

[0037] As an optional implementation of the embodiment of the present application, the multiple environmental protection indicators include a carbon emission indicator; the carbon emission indicator is obtained based on the target carbon emission factor corresponding to the bidding scheme.

[0038] As an optional implementation of an embodiment of the present application, the generation unit is also used to obtain the bill of materials and process flow chart in the bidding proposal; calculate the first carbon emission factor based on the bill of materials, and calculate the second carbon emission factor based on the energy consumption nodes in the process flow chart; add the first carbon emission factor and the second carbon emission factor to obtain the target carbon emission factor.

[0039] As an optional implementation of an embodiment of the present application, the intelligent bidding decision-making device also includes an optimization unit, which is specifically used to use the target weight coefficients of the technical score and the environmental score as the state space, the weight adjustment direction as the action space, based on the state-action space algorithm, and optimize the training of the dual-weight game model according to the reward value; the reward value is set according to the impact of the target bidder on the long-term environmental protection goal after executing the corresponding bidding plan.

[0040] In a third aspect, an embodiment of the present application provides an electronic device comprising: a memory and a processor, wherein the memory is used to store a computer program; and the processor is used to enable the electronic device to implement the intelligent bidding decision-making method described in any one of the above embodiments when executing the computer program.

[0041] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a computing device, the computing device implements the intelligent bidding decision-making method described in any of the above embodiments.

[0042] The intelligent bidding decision-making method provided in the embodiment of the present application is specifically as follows: obtaining the environmental protection score corresponding to each bidding scheme; obtaining the technical score corresponding to the bidding scheme based on the bidding behavior data of the bidder corresponding to each bidding scheme, and the bidding scheme; the bidding behavior data includes the historical bidding behavior data corresponding to each bidder; based on the double-weighted game model, combined with the Nash equilibrium theory, the target weight coefficients corresponding to the environmental protection score and the technical score of the bidding scheme are determined, and the environmental protection score and the technical score are weightedly calculated according to the target weight coefficient to obtain a comprehensive score; based on the comprehensive score corresponding to each bidder, the target winning scheme is screened out from multiple bidding schemes.

[0043] This application targets any one of the multiple bidding schemes, obtains the target latest environmental policy feature set and target supply chain feature set corresponding to the bidding scheme; then, in combination with the target latest environmental policy feature set and target supply chain feature set, obtains the scores corresponding to the multiple environmental indicators of the bidding scheme to generate the environmental score corresponding to the bidding scheme; the multiple environmental indicators include carbon emission indicators; and then, by obtaining the target latest environmental policy feature set and target supply chain feature set corresponding to the bidding scheme, closely combining the latest environmental policy and actual supply chain data to evaluate the bidding scheme, it is possible to accurately evaluate the environmental protection level of the bidding scheme, avoid the limitations of the traditional static evaluation method, make the environmental evaluation more in line with reality, and provide a basis for screening the winning scheme. Then, based on the double-weighted game model, the initial weight coefficients of the environmental score and the technical score are determined, and the weights are adjusted in combination with the bidder's historical bidding behavior. Then, in combination with the actual capabilities of the bidder, the initial weight coefficients corresponding to the environmental score and the technical score of the bidding scheme are adjusted to improve the accuracy of the bidding decision. Therefore, this application obtains a comprehensive score by weighted calculation of the environmental protection score and the technical score, which can comprehensively consider the environmental protection and technical levels of the bidding scheme, help to screen out bidding schemes that better meet the requirements in terms of environmental protection and technology, and improve the decision-making accuracy in the bidding process. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0045] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0046] Figure 1 One of the step flow charts of the intelligent bidding decision-making method provided in an embodiment of the present application;

[0047] Figure 2 This is the second step flow chart of the intelligent bidding decision-making method provided in the embodiment of the present application;

[0048] Figure 3 A schematic diagram of the structure of the intelligent bidding decision-making device provided in an embodiment of the present application;

[0049] Figure 4 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0050] In order to more clearly understand the above-mentioned objectives, features and advantages of the present disclosure, the scheme of the present disclosure will be further described below. It should be noted that the embodiments of the present disclosure and the features therein can be combined with each other in the absence of conflict.

[0051] In the following description, many specific details are set forth to facilitate a full understanding of the present disclosure, but the present disclosure may also be implemented in other ways different from those described herein; it is obvious that the embodiments in the specification are only part of the embodiments of the present disclosure, rather than all of the embodiments.

[0052] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete way. In addition, in the description of the embodiments of the present application, unless otherwise specified, the meaning of "multiple" refers to two or more.

[0053] It should be noted that, in this document, the term "comprises" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article, or apparatus comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus comprising the element.

[0054] The present application embodiment provides an intelligent bidding decision method, referring to Figure 1 As shown, the intelligent bidding decision method includes the following steps S101-S105:

[0055] S101. For any one of a plurality of bidding schemes, obtain a target latest environmental protection policy feature set and a target supply chain feature set corresponding to the bidding scheme.

[0056] In the scenario of intelligent bidding decision-making, after the tendering party releases the bidding information, it can receive bid proposals submitted by multiple bidders. At this time, in order to comprehensively evaluate each bid proposal, the embodiment of the present application sets up a corresponding intelligent bidding decision-making platform through the intelligent bidding decision-making method. After receiving the bid proposals of each bidder, each bid proposal is input into the intelligent bidding decision-making platform. The intelligent bidding decision-making platform gradually screens the target winning bidder and the target winning bid proposal corresponding to the target winning bidder through the intelligent bidding decision-making method provided by the embodiment of the present application.

[0057] Specifically, each bidding proposal needs to be evaluated and its corresponding comprehensive score is obtained to determine the bidding proposal with the highest comprehensive score as the target winning proposal.

[0058] Therefore, when calculating the comprehensive score corresponding to the bidding scheme, the embodiment of the present application needs to first obtain the target latest environmental protection policy feature set and target supply chain feature set corresponding to any one of the multiple bidding schemes.

[0059] The target latest environmental policy feature set includes multiple features that reflect the target latest environmental policy corresponding to the bidding proposal, such as policy effective date, environmental protection technical requirements, carbon emission limits, and the environmental friendliness of various materials. These features are derived from various environmental protection-related policy documents, regulations, industry standards, and market dynamics. Furthermore, by obtaining the features of relevant industry policies, it is possible to keep up with updates and changes in environmental protection-related market policies and obtain environmental guidance on market policies, laying the foundation for the subsequent evaluation of the bidding proposal's environmental friendliness score.

[0060] For example, in a bidding process for a city rail transit project, target market policies might include the latest national "Environmental Impact Assessment Standards for Urban Rail Transit Construction Projects," the industry's mandatory "Rail Transit Engineering Construction Safety Risk Management Specifications," and the specific project requirements of the regional carbon emissions peak action plan. Information on these policies, including environmental protection policies, industry regulatory policies, and market access policies, can be obtained from official government websites, policy and regulatory databases, industry association releases, and news platforms.

[0061] The target supply chain feature set includes various characteristic data in the supply chain links related to the bidding scheme, such as carbon emission data in the raw material procurement link, energy consumption data in the logistics and transportation process, resource utilization efficiency data in the production and manufacturing process, etc. Through these characteristic data, the data that affects the environmental protection of the current bidding scheme when the bidding scheme completes each link in the supply chain is measured.

[0062] Specifically, in step S101, for any one of the multiple bidding schemes, obtaining the target latest environmental protection policy feature set and the target supply chain feature set corresponding to the bidding scheme may be implemented as follows:

[0063] Step 1: Obtain the latest industry policy text corresponding to the bidding scheme, and extract key features of the latest industry policy text to generate the target latest environmental protection policy feature set.

[0064] In order to improve the efficiency of obtaining the latest industry policy texts, in an embodiment of the present application, a third-party market policy data platform is accessed through a RESTful API (representational state transfer application program interface), and the OAuth 2.0 protocol is used for security authentication to safely and efficiently collect the relevant latest industry policy texts from the third-party market data platform.

[0065] The RESTful API is a lightweight Web API design style based on the HTTP protocol. It offers excellent readability and scalability, enabling easy interaction with different platforms. For example, it can obtain information on raw material price fluctuations from a commodity price index platform or the latest policy updates from an industry policy database. OAuth 2.0 is an authorization framework that allows third-party applications to access protected resources with user authorization. This effectively ensures the security of user information and data during data transmission, preventing unauthorized access and data leakage.

[0066] It should be noted that when performing feature extraction, for text data such as policy documents, the TF-IDF (Term Frequency-Inverse Document Frequency) algorithm can be used to extract keywords to highlight important and frequently occurring words in the text; then, combined with a word vector model (such as Word2Vec, which can map words into vectors in the vector space so that semantically similar words are closer in the vector space) to generate semantic feature vectors, the text data is converted into a numerical vector form that can be understood and processed by computers, facilitating advanced analysis operations such as text classification and similarity calculation.

[0067] Step 2: Obtain the supply chain data of the bidder corresponding to the bidding scheme, and perform feature extraction on the supply chain data to generate the target supply chain feature set.

[0068] Specifically, the target supply chain feature set is closely related to the bidder's resource allocation capabilities. Therefore, it can be connected to the bidder's corresponding enterprise resource planning (ERP) system to collect the bidder's supply chain data, such as production energy consumption and logistics transportation data; among them, the bidder's ERP system integrates data from various links such as enterprise procurement, production, sales, logistics, etc., and is the core storage and management system for internal enterprise data.

[0069] Then, with the help of Apache NiFi in the ETL tool (Extract, Transform, Load, i.e. data extraction, conversion and loading), production energy consumption and logistics transportation data can be extracted regularly; at the same time, because Apache NiFi has a visual operation interface and powerful data processing and scheduling functions, it can extract the required supply chain data from the ERP system according to the set time period (such as early morning every day), and perform pre-processing operations such as format conversion and data cleaning during the extraction process to further ensure data quality.

[0070] It should be noted that when obtaining the target latest environmental protection policy feature set based on the latest industry policy text and the target supply chain feature set based on the bidder's production energy consumption and logistics transportation data, it is necessary to clean the latest industry policy text and supply chain data, process abnormal data, and unify the data format, thereby improving data quality and mapping heterogeneous data into a standard format (such as JSON data structure) to facilitate subsequent feature extraction.

[0071] When cleaning data, streaming data (data continuously generated in real time) can be cleaned based on a sliding cleaning mechanism of a time window. When data is missing, linear interpolation or mean filling is used to compensate for it to ensure the continuity and integrity of the data.

[0072] For abnormal data, such as negative energy consumption and empty logistics tracks, a rule engine (such as Drools, which can judge and process data according to pre-defined rules) is used to automatically mark these abnormal data and trigger a manual review process, thereby promptly discovering and processing errors or unreasonable aspects in the data to ensure data quality.

[0073] S102. In combination with the target latest environmental policy feature set and the target supply chain feature set, obtain scores corresponding to the multiple environmental indicators of the bidding scheme to generate an environmental score corresponding to the bidding scheme.

[0074] Among them, the multiple environmental protection indicators include carbon emission indicators.

[0075] In the embodiment of the present application, the multiple environmental indicators of the bidding proposal are pre-set indicators that can reflect the environmental friendliness of the current bidding proposal, covering carbon emissions, energy consumption, waste disposal, water resource utilization, material utilization, etc. For example, the carbon emission indicator focuses on carbon dioxide emissions in the production and transportation links; the energy consumption indicator considers the use of energy such as electricity and coal.

[0076] It should be emphasized that the multiple environmental protection indicators in the embodiment of the present application include a carbon emission indicator; the carbon emission indicator is obtained based on the target carbon emission factor corresponding to the bidding scheme.

[0077] In the context of global climate change response and strategic initiatives, carbon emissions have a significant impact on the environmental performance of companies and projects. They are directly linked to total greenhouse gas emissions, impacting ecological and environmental quality and the achievement of sustainable development goals. Accurately considering carbon emissions in bidding decisions can guide bidders in adopting low-carbon technologies and processes, driving the industry's green transformation.

[0078] Using the target carbon emission factor as a basis, carbon emission indicators are then generated, accurately quantifying the carbon emissions of each bid proposal at each stage (such as raw material procurement, manufacturing, transportation and delivery). Different industries, processes, and energy use correspond to different carbon emission factors. This calculation allows carbon emission accounting to be more closely aligned with the actual bid proposal, improving the accuracy and scientific nature of environmental protection scores and providing reliable data support for tenderers to select low-carbon and environmentally friendly bid proposals.

[0079] Specifically, the target carbon emission factor corresponding to the bid proposal can be understood as a quantitative coefficient used to measure the carbon dioxide emissions generated per unit of output (such as per unit of product produced or per unit of service provided) in the bid proposal during a specific activity or process. For example, in the manufacturing industry, the carbon dioxide emissions corresponding to the production of each ton of steel or each electronic product; in the construction industry, the carbon emissions generated per square meter of building construction can all be calculated using the corresponding carbon emission factor.

[0080] Furthermore, the target carbon emission factor corresponding to the bidding scheme can be obtained through the following steps 1 to 3:

[0081] Step 1: Obtain the bill of materials and process flow chart in the bidding proposal.

[0082] The bill of materials in the bid proposal will record the various materials required to produce the product or provide the service, including material name, quantity, source, etc. It is one of the basic data for subsequent carbon emissions calculations. By clarifying the material composition, it provides a basis for querying carbon emission factors.

[0083] The process flow chart shows the entire production process from raw material input to final product output, which can help identify key energy consumption nodes in the production process, such as heating, cooling, processing, etc. These nodes are important sources of carbon emissions and are crucial for accurately calculating carbon emission factors.

[0084] Step 2: Calculate a first carbon emission factor based on the bill of materials, and calculate a second carbon emission factor based on the energy consumption nodes in the process flow chart.

[0085] Next, based on the bill of materials, the carbon emission factor corresponding to each material is searched from a professional material database (such as Ecoinvent). This is then combined with the material quantity used to calculate a first carbon emission factor based on the material. For example, if the production of a product requires 10 kg of steel, and the carbon emission factor for steel is found in the database to be x kg CO2 / kg steel, then the carbon emissions of the steel alone are 10 x kg CO2. Similar calculations can be used to calculate the total carbon emissions of all materials, thereby obtaining the first carbon emission factor.

[0086] Similarly, based on the key energy consumption nodes identified in the process flow chart, analyze the energy consumption of each energy consumption node. Combined with real-time data sources (such as energy market prices and changes in transportation routes), dynamically adjust the carbon emission factor corresponding to the energy. For example, if a heating process uses coal as an energy source, based on real-time energy market prices, the company may switch to natural gas energy. The carbon emission factor of natural gas is different from that of coal, and the carbon emissions of this energy consumption node need to be recalculated. After calculating the carbon emissions of all key energy consumption nodes, the second carbon emission factor is obtained.

[0087] It should be noted that when obtaining the first carbon emission factor and the second carbon emission factor, taking into account the changes in the production processes of various materials, as well as the impact of dynamic factors such as fluctuations in the market prices of various materials and energy, changes in transportation routes, etc. on carbon emissions, it is necessary to dynamically adjust the first carbon emission factor and the second carbon emission factor in real time to ensure the accuracy of the carbon emissions calculation.

[0088] Step 3: Add the first carbon emission factor and the second carbon emission factor to obtain a target carbon emission factor.

[0089] The target carbon emission factor is the sum of the first and second carbon emission factors. This factor comprehensively considers the carbon emissions generated by material use and energy consumption in the production process, reflecting the carbon emission level of the bidding proposal more comprehensively and accurately, providing key data support for the subsequent environmental protection score calculation.

[0090] Based on the target's latest environmental policy feature set and the target supply chain feature set, comprehensively evaluate the environmental protection level of each environmental indicator of the current bidding plan, quantify the environmental protection level of each environmental indicator, and obtain the corresponding score for each environmental indicator. Specifically, corresponding scoring models can be constructed for different environmental indicators. For quantifiable indicators (such as carbon emissions and energy consumption), linear scoring or piecewise function scoring is used. If the carbon emissions are lower than the standard, it will be worth a high score, and if it exceeds the standard, points will be deducted proportionally. For qualitative indicators (such as the environmental protection of waste treatment methods), expert experience and industry standards are combined to divide the scores into grades to evaluate the scores.

[0091] It should be noted that a corresponding environmental scoring model can be developed based on the above-mentioned method for obtaining the environmental score corresponding to a bidding proposal. After inputting the bidding proposal into the environmental scoring model, the corresponding environmental score can be obtained. After obtaining the scores corresponding to each environmental indicator of the bidding proposal, a fuzzy logic and evidence theory fusion algorithm can be used to unify the qualitative and quantitative environmental indicators, and then the scores of these individual environmental indicators are summed to obtain the environmental score corresponding to the bidding proposal. Further, according to the above method, the environmental score corresponding to each bidding proposal is obtained, so that a comprehensive score for each bidding proposal can be obtained later.

[0092] S103: Obtain the technical score corresponding to the bidding scheme, and determine the initial weight coefficients corresponding to the environmental score and the technical score of the bidding scheme respectively based on a double-weight game model.

[0093] In some embodiments, the technical score corresponding to the bidding proposal can be obtained by evaluating and scoring the bidding proposal from multiple dimensions such as the advancement of the bidding proposal's technical parameters, the completeness of the proposal, the feasibility of the implementation plan, etc., so as to evaluate the bidding proposal's technical advancement, technical maturity, technical feasibility, etc., and give a technical score.

[0094] It should be noted that this application does not impose any limitation on the method of obtaining the technical score corresponding to the bidding proposal.

[0095] Furthermore, for a bidding proposal, after obtaining the environmental protection score and technical score corresponding to the bidding proposal, the environmental protection score and technical score can be incorporated into the game framework based on the double-weighted game model to determine the weight coefficient corresponding to the score of the environmental protection score and technical score dimension.

[0096] The dual-weighted game model primarily considers the interrelationship and trade-offs between environmental and technical scores in bidding decisions. Environmental and technical goals are considered as two aspects of the game, and a payoff function is established for both parties. This payoff function reflects the benefits associated with achieving these goals under different weightings. For example, the benefits of environmental goals may be related to reduced carbon emissions and resource savings, while the benefits of technical goals may be related to increased production efficiency and product quality.

[0097] Then, using game theory methods, such as Nash equilibrium theory, we sought weighting coefficients that achieve a relative balance between environmental protection and technical objectives under the current circumstances. At a Nash equilibrium point, unilaterally changing the weights of either party would not improve its own benefits. Through calculation and analysis, we determined the initial weighting coefficients corresponding to the environmental and technical scores. For example, the model calculations determined an initial weighting coefficient of 0.4 for the environmental score and 0.6 for the technical score. This means that in the current bidding situation, technical factors are relatively more important in decision-making, but environmental factors also play a certain role.

[0098] Specifically, we can construct a game matrix between technical score (T) and environmental score (E), define the payoff function U(T,E)=αT+βE, and determine the initial weight coefficients corresponding to the environmental score and technical score respectively.

[0099] Here, α and β are weighting coefficients, respectively determining the contribution of technology and environmental protection points to the overall benefits. This is solved using Nash equilibrium theory. Nash equilibrium refers to a game in which each participant chooses the optimal strategy, and unilaterally changing strategy by either party will not increase its own benefits. In this model, the goal is to find the right α and β to maximize both technology and environmental benefits.

[0100] S104. Adjust the initial weight coefficients corresponding to the environmental protection score and the technical score respectively based on the historical bidding behavior of the bidder corresponding to the bidding scheme, and generate target weight coefficients corresponding to the environmental protection score and the technical score respectively.

[0101] It should be noted that after determining the initial weight coefficients corresponding to the environmental score and technical score of the bidding scheme through the dual-weight game model, the dual-weight game model can also adjust the initial weight coefficients corresponding to the environmental score and the technical score based on the historical bidding behavior of the bidder corresponding to the bidding scheme, and generate target weight coefficients corresponding to the environmental score and the technical score.

[0102] Specifically, historical bidding behavior can include bidders' bid price records, bid winning and contract performance, response time data, and environmental and technical performance. Bid price records can be obtained by obtaining bidders' past bids, including information such as bid amounts and the degree of deviation from the market average. For example, statistics can be collected to determine whether bids for similar projects were above, below, or close to the market average, as well as the price fluctuation range. Bid winning and contract performance can be obtained by reviewing bidders' bid winning records, such as the number of bids won and the types of projects won. Furthermore, detailed information can be obtained on the contract performance of the winning projects, including whether products or services were delivered on time, whether the quality of the products or services met standards, and whether there were any breach of contract disputes. Response time data records the time interval between the release of tender information and the submission of bid documents. This allows analysis of the speed of a bidder's response to tenders and the degree of project priority and the efficiency of its internal organizational coordination. Environmental and technical performance focuses on the implementation of environmental measures in past projects, such as whether environmental standards were met and whether any innovative environmental initiatives were implemented. Technical performance can also be assessed, such as whether advanced technologies were adopted and the effectiveness of their application.

[0103] Furthermore, the bidder's historical performance is evaluated based on the aforementioned multiple historical bidding behaviors, and the adjustment range of the initial weight coefficients of the environmental protection score and technical score is determined. For example, if the weight coefficients of the environmental protection score and technical score of the current bid proposal are 0.4 and 0.6, respectively; based on the aforementioned historical bidding behaviors, it is known that the bidder has excellent environmental performance, the initial weight coefficient of the environmental protection score can be appropriately increased, such as from 0.4 to 0.5; the initial weight coefficient of the technical score remains unchanged, and the target weight coefficients corresponding to the environmental protection score and technical score are obtained, thereby improving the overall score of the current bid proposal. For another example, if the weight coefficients of the environmental protection score and technical score of the current bid proposal are 0.4 and 0.6, respectively; based on the aforementioned historical bidding behaviors, it is known that the bidder has poor performance in terms of bid winning and contract fulfillment, the initial weight coefficient of the technical score can be appropriately reduced, such as from 0.6 to 0.4; the initial weight coefficient of the environmental protection score remains unchanged, and the target weight coefficients corresponding to the environmental protection score and technical score are obtained, thereby reducing the overall score of the current bid proposal.

[0104] For the dual-weight game model, we can introduce the decay function α(t) = α0·e -λt .

[0105] Among them, α0 is the initial weight, which represents the proportion of a certain factor (such as the technical score weight or the environmental protection score weight) in the comprehensive evaluation at the initial moment. λ is the decay rate, which determines how fast the weight decreases over time. It is a fixed parameter set according to actual conditions and experience. t is the time interval, which represents the time elapsed from the initial moment to the current moment. As time t increases, e -λt The value of will gradually decrease, so that α(t), that is, the current weight, will continue to decrease, realizing the weight decay over time.

[0106] This is because the value of a bidder's historical bidding behavior in current decision-making changes over time. Some older data reflects past technological advancements, environmental standards, and other factors, which may no longer be relevant to current realities. The decay mechanism reduces the impact of this outdated data on current weighting, making the weighting more relevant to current conditions and improving decision-making accuracy.

[0107] Through the above steps, the initial weight coefficients corresponding to the environmental protection score and the technical score are adjusted based on the historical bidding behavior of the bidder corresponding to the bidding scheme, so that the weight distribution is more in line with the actual capabilities of the bidder and the characteristics of project requirements, thereby improving the accuracy of the comprehensive score.

[0108] S105. Perform weighted calculation on the environmental protection score and the technical score according to the target weight coefficient to obtain a comprehensive score, and screen out a target winning bid from multiple bidding schemes based on the comprehensive scores corresponding to the respective bidders.

[0109] Furthermore, after determining the target weight coefficients corresponding to the environmental protection score and the technical score, the environmental protection score and the technical score can be weighted calculated based on the determined target weight coefficients.

[0110] Through the above steps, a comprehensive score corresponding to each bid proposal is obtained. The comprehensive scores of multiple bidders are then ranked from high to low. The bid proposal with the highest comprehensive score is selected as the target winning proposal. In practice, if multiple bidders have similar comprehensive scores, further consideration can be given to the individual scores of the environmental and technical scores, or a comprehensive assessment can be made based on factors such as the bidder's corporate reputation and after-sales service to ultimately select the most suitable target winning proposal.

[0111] This application targets any one of the multiple bidding schemes, obtains the target latest environmental policy feature set and target supply chain feature set corresponding to the bidding scheme; then, in combination with the target latest environmental policy feature set and target supply chain feature set, obtains the scores corresponding to the multiple environmental indicators of the bidding scheme to generate the environmental score corresponding to the bidding scheme; the multiple environmental indicators include carbon emission indicators; and then, by obtaining the target latest environmental policy feature set and target supply chain feature set corresponding to the bidding scheme, closely combining the latest environmental policy and actual supply chain data to evaluate the bidding scheme, it is possible to accurately evaluate the environmental protection level of the bidding scheme, avoid the limitations of the traditional static evaluation method, make the environmental evaluation more in line with reality, and provide a basis for screening the winning scheme. Then, based on the double-weighted game model, the initial weight coefficients of the environmental score and the technical score are determined, and the weights are adjusted in combination with the bidder's historical bidding behavior. Then, in combination with the actual capabilities of the bidder, the initial weight coefficients corresponding to the environmental score and the technical score of the bidding scheme are adjusted to improve the accuracy of the bidding decision. Therefore, this application obtains a comprehensive score by weighted calculation of the environmental protection score and the technical score, which can comprehensively consider the environmental protection and technical levels of the bidding scheme, help to screen out bidding schemes that better meet the requirements in terms of environmental protection and technology, and improve the decision-making accuracy in the bidding process.

[0112] As an extension and refinement of the above embodiment, refer to Figure 2 As shown, the embodiment of the present application provides another intelligent bidding decision method, the specific steps of which include the following:

[0113] S201. For any one of the multiple bidding schemes, obtain the target latest environmental protection policy feature set and the target supply chain feature set corresponding to the bidding scheme.

[0114] S202: Based on the target latest environmental policy feature set and the target supply chain feature set, score the multiple environmental indicators of the bidding scheme respectively to obtain the initial score corresponding to each environmental indicator.

[0115] Among them, the multiple environmental protection indicators include carbon emission indicators.

[0116] In the embodiment of the present application, taking into account the different contributions of different environmental indicators to the environmental impact and the overall environmental protection level of the project under the industry and policy background, after obtaining the initial score corresponding to each environmental indicator, a corresponding weight coefficient will be set according to the importance of each indicator. Therefore, in this step, the target latest environmental policy feature set and the target supply chain feature set are combined to score the multiple environmental indicators of the bidding scheme separately, and the initial scores corresponding to each environmental indicator are preliminarily obtained.

[0117] S203. Perform weighted calculation on the initial scores corresponding to the environmental protection indicators according to the weight coefficients corresponding to the environmental protection indicators to obtain the environmental protection score corresponding to the bidding scheme.

[0118] In an embodiment of the present application, corresponding weight coefficients are set according to the different degrees of influence of different environmental indicators on the environmental protection level of the bidding scheme. After obtaining the initial score corresponding to each environmental indicator in the above step S202, a weighted calculation is performed to obtain the environmental protection score corresponding to the bidding scheme.

[0119] Through the above-mentioned weighted calculation, the role of important environmental indicators can be more highlighted. For example, under the current environmental situation, carbon emission indicators, as an important feature for measuring environmental protection, will be given a higher weight, which can make the environmental protection score more in line with actual needs and policy orientation, and thus provide the tendering party with a comprehensive and objective environmental protection score, which will help to select solutions with better environmental performance.

[0120] For example, in bidding for high-energy-consuming manufacturing industries, carbon emissions have a significant environmental impact and attract significant policy attention, far outweighing water resource utilization. By weighting carbon emissions and water resource utilization, this approach allows for more realistic scoring, highlights the role of key indicators, and accurately measures the environmental performance of bid proposals.

[0121] It should be noted that the embodiment of the present application also includes the following steps:

[0122] The weight coefficients corresponding to the plurality of environmental protection indicators in the environmental protection indicator set are updated according to the latest market policy data.

[0123] Because environmental policies and industry development priorities are subject to change, the weighting of indicators can be dynamically adjusted based on policy timeliness and industry relevance during the weighted calculation. When policies strongly promote energy conservation and emission reduction, the weight of carbon emissions indicators increases; when industries focus on waste recycling, the weight of waste treatment indicators increases.

[0124] Therefore, the environmental scoring model can be updated and optimized based on the latest market policy data and re-weighted at predetermined intervals, such as quarterly, for each environmental indicator. This allows the environmental scoring model to promptly respond to policy and industry changes, guiding bidders to focus on key environmental areas and ensuring bidding decisions meet current requirements. For other environmental indicators, such as water consumption and waste disposal, a dynamic weighting model is designed based on policy guidance and industry priorities to automatically adjust the weight of each indicator in the environmental scoring.

[0125] S204: Obtain the technical score corresponding to the bidding scheme, and determine the initial weight coefficients corresponding to the environmental score and the technical score of the bidding scheme respectively based on a double-weighted game model.

[0126] S205. Adjust the initial weight coefficients corresponding to the environmental protection score and the technical score respectively based on the historical bidding behavior of the bidder corresponding to the bidding scheme, and generate target weight coefficients corresponding to the environmental protection score and the technical score respectively.

[0127] S206. Perform weighted calculation on the environmental protection score and the technical score according to the target weight coefficient to obtain a comprehensive score, and screen out a target winning bid from multiple bidding schemes based on the comprehensive scores corresponding to the respective bidders.

[0128] Design attenuation function α(t) = α0·e -λt , where α0 is the initial weight, λ is the decay rate, and t is the time interval; dynamically adjust the impact of historical data on current decisions to avoid interference from outdated data.

[0129] Among them, α0 is the initial weight, which represents the proportion of a certain factor (such as the technical score weight or the environmental protection score weight) in the comprehensive evaluation at the initial moment. λ is the decay rate, which determines how fast the weight decreases over time. It is a fixed parameter set according to actual conditions and experience. t is the time interval, which represents the time elapsed from the initial moment to the current moment. As time t increases, e -λt The value of will gradually decrease, so that α(t), that is, the current weight, will continue to decrease, realizing the weight decay over time.

[0130] In bidding decisions, the reference value of historical data (such as performance data of past bidders) for current decisions will change over time; older data reflects past technical levels, environmental protection standards, etc., which may no longer be in line with current reality. Through the decay mechanism, the impact of these outdated data on the current weight determination can be reduced, so that the weight distribution is more in line with the current actual situation and the accuracy of decision-making is improved. Adapt to environmental changes: Industry technology is advancing and environmental protection policies are being updated. The decay mechanism allows the model to "forget" old information that is no longer applicable in a timely manner, and pay more attention to recent and more relevant data. For example, after the introduction of a new environmental protection policy, the performance data of bidders that meet the requirements of the new policy in the near future should be given more attention, while the influence of previous data will decrease. This will allow the dual-weight game model to better adapt to environmental changes, optimize weight distribution, and make more reasonable bidding decisions.

[0131] It should be noted that after deciding on the target winning bid among multiple bidding schemes, we can always pay attention to the subsequent implementation of the target winning bid. According to its impact on environmental protection goals, we can optimize and train the dual-weight game model to further improve the accuracy of the dual-weight game model. The specific implementation method is as follows:

[0132] Based on the state-action space algorithm, the target weight coefficients of the technical score and the environmental score are used as the state space, the weight adjustment direction is used as the action space, and the dual-weight game model is optimized and trained according to the reward value; the reward value is set according to the impact of the target bidder on the long-term environmental protection goals after executing the corresponding bidding plan.

[0133] First, we need to define the state and action space. Specifically, we combine the target weight coefficients for the technical score and the environmental score to form the state space. For example, if the current technical score weight is 0.6 and the environmental score weight is 0.4, this weight combination represents a state in the state space. It reflects the emphasis on technology and environmental protection in the bidding decision.

[0134] The weight adjustment direction is set as the action space, such as increasing the weight of technology and reducing the weight of environmental protection. By performing different operations in the action space, a better weight distribution can be explored.

[0135] The reward value is determined based on the impact of the target winning proposal on long-term environmental goals after implementation. If the target winning proposal achieves long-term environmental goals such as reducing carbon emissions and improving resource utilization, for example, by exceeding energy conservation and emission reduction targets or incorporating innovative environmental technologies, a higher reward value may be awarded. Conversely, if the proposal encounters environmental violations or fails to achieve the expected environmental impact during implementation, a lower reward value will be awarded.

[0136] Based on a state-action space algorithm, reinforcement learning is performed using the Deep Q-Learning (DQN) algorithm. Specifically, based on the current state (i.e., the weighted combination of technical and environmental scores), an action (weight adjustment direction) is selected from the action space. After executing the action, the reward value obtained is used to evaluate the effectiveness of the weight adjustment. A high reward value indicates that the current weight adjustment direction is conducive to achieving long-term environmental goals, and the model will strengthen this adjustment strategy. A low reward value will lead the model to explore alternative weight adjustment directions. This dual-weight game model is then optimized and trained by continuously selecting actions, obtaining rewards, and adjusting strategies. After each optimization training, the Q-value is updated based on the selected actions and the rewards obtained. The Q-value records the expected cumulative reward for taking a specific action in a certain state. By continuously updating the Q-value, the model learns which actions in different states yield higher rewards. This enables more precise weighting of technical and environmental scores, improving the accuracy of bidding decisions in terms of the balance between environmental protection and technology, and better serving long-term environmental goals.

[0137] The final winning bid selected by the human can be compared with the target winning bid calculated using the dual-weight game model. The difference in winning results can be fed back to the model as a reward signal. If the winning results match, the model receives positive feedback and further executes the current weight adjustment strategy. Conversely, if the winning results differ, the model further adjusts the weight adjustment strategy for optimization training.

[0138] Furthermore, the embodiment of the present application continuously optimizes the training and feedback iteration of the dual-weight game model through the above-mentioned two model optimization methods, so that it can more accurately determine the target weight coefficients of the technical score and the environmental score, so as to better achieve the balance between environmental protection and technology in bidding decisions, and at the same time promote the achievement of long-term environmental protection goals.

[0139] As an extension and refinement of the above embodiment, the embodiment of the present application also provides an intelligent bidding decision system corresponding to the intelligent bidding decision method. The system specifically includes: a data access module, a computing engine module, and a decision output module, wherein:

[0140] The data access module is deployed on the edge server, close to the data source, which can greatly reduce data transmission delays. In production bidding scenarios, it can quickly obtain real-time energy consumption data of production equipment and provide timely support for subsequent calculations.

[0141] The computing engine module is deployed in the cloud using Docker and Kubernetes containerization technology. Docker implements application encapsulation and isolation, while Kubernetes orchestrates and schedules containers. This allows for flexible scaling of computing resources based on business volume, ensuring efficient system operation and meeting the needs of evaluating large numbers of bid proposals.

[0142] The decision output module uses the WebSocket protocol to push the recommended winning proposal to the front-end interface in real time, ensuring timely communication of information and allowing users to quickly obtain results.

[0143] It's important to note that the intelligent bidding decision-making system also uses Prometheus to collect system operational metrics, such as data throughput and model calculation latency, and then visualizes them in Grafana. This allows operations personnel to promptly monitor system status and identify potential performance bottlenecks. If a module fails, it automatically switches to a backup node, ensuring uninterrupted system operation and preventing disruptions in the bidding decision-making process.

[0144] At the same time, the AES-256 encryption algorithm is used to encrypt sensitive data in transmission (such as bidder information) to prevent the data from being stolen or tampered with during transmission. Data desensitization rules are designed to process public data, remove identifiable information, prevent privacy leaks, and ensure data security for all parties. A visual decision support interface is provided to display environmental protection score comparison heat maps, risk level distribution maps, etc. The heat map intuitively presents the differences in environmental protection scores of different bidding schemes, and the risk level distribution map shows the potential risks of each scheme to assist users in understanding the evaluation results. Users are supported to manually adjust the weight distribution to meet personalized needs, and the feedback results are synchronized to the optimization engine in real time to achieve continuous optimization of the system and improve interactivity.

[0145] Based on the same inventive concept, as an implementation of the above method, an embodiment of the present application also provides an intelligent bidding decision-making device, which corresponds to the aforementioned method embodiment. For ease of reading, this embodiment will no longer repeat the details of the aforementioned method embodiment one by one, but it should be clear that an intelligent bidding decision-making device in this embodiment can correspond to and implement all the contents of the aforementioned method embodiment.

[0146] The present application embodiment provides an intelligent bidding decision-making device, Figure 3 This is a structural diagram of the intelligent bidding decision-making device, as shown in Figure 3 As shown, the intelligent bidding decision-making device 300 includes:

[0147] The first acquisition unit 301 is configured to acquire, for any one of the multiple bidding schemes, a target latest environmental protection policy feature set and a target supply chain feature set corresponding to the bidding scheme;

[0148] A generating unit 302 is configured to combine the target latest environmental policy feature set and the target supply chain feature set to obtain scores corresponding to multiple environmental indicators of the bidding scheme to generate an environmental score corresponding to the bidding scheme; the multiple environmental indicators include a carbon emission indicator;

[0149] The second acquisition unit 303 is used to obtain the technical score corresponding to the bidding scheme, and determine the initial weight coefficients corresponding to the environmental score and the technical score of the bidding scheme based on a double-weight game model;

[0150] An adjusting unit 304 is configured to adjust the initial weight coefficients corresponding to the environmental protection score and the technical score respectively based on the historical bidding behavior of the bidder corresponding to the bidding scheme, and generate target weight coefficients corresponding to the environmental protection score and the technical score respectively;

[0151] The screening unit 305 is configured to perform weighted calculation on the environmental protection score and the technical score according to the target weight coefficient to obtain a comprehensive score, and screen out a target winning bid from multiple bidding schemes based on the comprehensive scores corresponding to the respective bidders.

[0152] As an optional implementation of the embodiment of the present application, the first acquisition unit 401 is specifically used to obtain the latest industry policy text corresponding to the bidding scheme, and perform key feature extraction on the latest industry policy text to generate the target latest environmental protection policy feature set; obtain the supply chain data of the bidder corresponding to the bidding scheme, and perform feature extraction on the supply chain data to generate the target supply chain feature set.

[0153] As an optional implementation of an embodiment of the present application, the generation unit 402 is specifically used to score the multiple environmental indicators of the bidding scheme in combination with the target latest environmental policy feature set and the target supply chain feature set, and obtain the initial score corresponding to each environmental indicator; according to the weight coefficient corresponding to each environmental indicator, the initial score corresponding to each environmental indicator is weightedly calculated to obtain the environmental score corresponding to the bidding scheme.

[0154] As an optional implementation of the embodiment of the present application, the intelligent bidding decision-making device also includes an updating unit, which is specifically used to adjust and update the weight coefficients corresponding to the various environmental protection indicators based on the latest environmental protection policy data characteristics of the target.

[0155] As an optional implementation of the embodiment of the present application, the multiple environmental protection indicators include a carbon emission indicator; the carbon emission indicator is obtained based on the target carbon emission factor corresponding to the bidding scheme.

[0156] As an optional implementation of an embodiment of the present application, the generation unit 402 is also used to obtain the bill of materials and process flow chart in the bidding proposal; calculate the first carbon emission factor based on the bill of materials, and calculate the second carbon emission factor based on the energy consumption nodes in the process flow chart; add the first carbon emission factor and the second carbon emission factor to obtain the target carbon emission factor.

[0157] As an optional implementation of an embodiment of the present application, the intelligent bidding decision-making device also includes an optimization unit, which is specifically used to use the target weight coefficients of the technical score and the environmental score as the state space, the weight adjustment direction as the action space, based on the state-action space algorithm, and optimize the training of the dual-weight game model according to the reward value; the reward value is set according to the impact of the target bidder on the long-term environmental protection goal after executing the corresponding bidding plan.

[0158] Based on the same inventive concept, an embodiment of the present disclosure further provides an electronic device. Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present disclosure, such as Figure 4 As shown, the electronic device provided in this embodiment includes: a memory 401 and a processor 402, wherein the memory 401 is used to store computer programs; the processor 402 is used to execute the intelligent bidding decision method provided in the above embodiment when executing the computer program.

[0159] Based on the same inventive concept, an embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the computing device implements the intelligent bidding decision-making method provided in the above embodiment.

[0160] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code.

[0161] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0162] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0163] Computer-readable media includes both permanent and non-permanent, removable and non-removable storage media. Storage media can implement any method or technology for storing information, which can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media, such as modulated data signals and carrier waves.

[0164] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some or all of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application.

Claims

1. An intelligent bidding decision-making method, characterized in that: include: For any one of the multiple bidding schemes, obtaining a target latest environmental protection policy feature set and a target supply chain feature set corresponding to the bidding scheme; Combining the target latest environmental policy feature set and the target supply chain feature set, obtaining scores corresponding to multiple environmental indicators of the bidding scheme to generate an environmental score corresponding to the bidding scheme; the multiple environmental indicators include a carbon emission indicator; Obtaining the technical score corresponding to the bidding scheme, and determining the initial weight coefficients corresponding to the environmental score and the technical score of the bidding scheme respectively based on a double-weighted game model; Adjusting the initial weight coefficients corresponding to the environmental protection score and the technical score based on the historical bidding behavior of the bidder corresponding to the bidding scheme to generate target weight coefficients corresponding to the environmental protection score and the technical score; The environmental protection score and the technical score are weighted according to the target weight coefficient to obtain a comprehensive score, and based on the comprehensive scores corresponding to the various bidders, a target winning bid is selected from multiple bidding schemes.

2. The method according to claim 1, characterized in that The step of obtaining, for any one of the multiple bidding schemes, a target latest environmental protection policy feature set and a target supply chain feature set corresponding to the bidding scheme includes: Obtaining the latest industry policy text corresponding to the bidding scheme, and extracting key features from the latest industry policy text to generate the target latest environmental protection policy feature set; The supply chain data of the bidder corresponding to the bidding scheme is obtained, and features are extracted from the supply chain data to generate the target supply chain feature set.

3. The method according to claim 1, characterized in that The step of combining the target latest environmental policy feature set and the target supply chain feature set to obtain scores corresponding to the multiple environmental indicators of the bidding scheme to generate an environmental score corresponding to the bidding scheme includes: Based on the target latest environmental policy feature set and the target supply chain feature set, score the multiple environmental indicators of the bidding scheme respectively to obtain the initial score corresponding to each environmental indicator; According to the weight coefficient corresponding to each environmental protection indicator, the initial score corresponding to each environmental protection indicator is weightedly calculated to obtain the environmental protection score corresponding to the bidding scheme.

4. The method according to any one of claims 1 to 3, characterized in that Before performing weighted calculation on the initial scores corresponding to the environmental protection indicators according to the weight coefficients corresponding to the environmental protection indicators to obtain the environmental protection scores corresponding to the bidding schemes, the method further includes: Based on the latest environmental protection policy data characteristics of the target, the weight coefficients corresponding to the various environmental protection indicators are adjusted and updated.

5. The method according to claim 1, wherein The carbon emission index is obtained based on the target carbon emission factor corresponding to the bidding scheme.

6. The method according to claim 5, characterized in that The method further comprises: Obtain the bill of materials and process flow chart in the bidding proposal; Calculating a first carbon emission factor based on the bill of materials, and calculating a second carbon emission factor based on the energy consumption nodes in the process flow chart; The first carbon emission factor and the second carbon emission factor are added to obtain a target carbon emission factor.

7. The method according to claim 1, characterized in that After weighting the environmental protection score and the technical score according to the target weight coefficient to obtain a comprehensive score, and screening a target winning bid from multiple bidding schemes based on the comprehensive scores corresponding to the bidders, the method further includes: Based on the state-action space algorithm, the target weight coefficients of the technical score and the environmental score are used as the state space, the weight adjustment direction is used as the action space, and the dual-weight game model is optimized and trained according to the reward value; the reward value is set according to the impact of the target bidder on the long-term environmental protection goals after executing the corresponding bidding plan.

8. An intelligent bidding decision-making device, characterized in that: include: A first acquisition unit is configured to acquire, for any one of the multiple bidding schemes, a target latest environmental protection policy feature set and a target supply chain feature set corresponding to the bidding scheme; a generating unit, configured to combine the target latest environmental policy feature set and the target supply chain feature set to obtain scores corresponding to the multiple environmental indicators of the bidding scheme, so as to generate an environmental score corresponding to the bidding scheme; The multiple environmental protection indicators include carbon emission indicators; A second acquisition unit is configured to acquire a technical score corresponding to the bidding scheme, and determine initial weight coefficients corresponding to the environmental score and the technical score of the bidding scheme respectively based on a double-weighted game model; an adjusting unit, configured to adjust the initial weight coefficients corresponding to the environmental protection score and the technical score respectively based on the historical bidding behavior of the bidder corresponding to the bidding scheme, and generate target weight coefficients corresponding to the environmental protection score and the technical score respectively; A screening unit is used to perform weighted calculation on the environmental protection score and the technical score according to the target weight coefficient to obtain a comprehensive score, and screen out a target winning bid from multiple bidding schemes based on the comprehensive scores corresponding to the respective bidders.

9. An electronic device, characterized in that: include: A memory and a processor, wherein the memory is used to store a computer program; and the processor is used to enable the electronic device to implement the intelligent bidding decision method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a computing device, the computing device implements the intelligent bidding decision-making method according to any one of claims 1 to 7.

Images