Large-model-driven intelligent bidding document generation method and system for petrochemical energy marine transportation industry

Abstract

The invention belongs to the technical field of shipping and artificial intelligence, and particularly relates to a large-model-driven intelligent bidding document generation method and system for the petrochemical energy marine transportation industry, and the method comprises the following specific steps: obtaining original data from an enterprise internal database and an external industry mechanism database through a multi-source data collection module, and after the storage, key concepts in the petrochemical energy marine transportation field are associated in the knowledge graph through the knowledge graph construction module, so that the system can quickly retrieve or recommend text fragments according to the semantic association degree. According to the system, original data are structurally stored through the multi-source data acquisition module, the knowledge graph construction module is associated with key concepts to quickly retrieve and recommend text fragments, multiple templates are preset in the template library multiplexing module and can be directly retrieved and revised, and the large model creation module generates professional texts and the like; the time investment of manually writing the bidding document is greatly reduced, and the compiling period is effectively shortened.

Description

Technical Field

[0001] The present invention relates to the field of shipping and artificial intelligence technology, and specifically to a large-model-driven intelligent bid generation method and system for the petrochemical energy shipping industry. Background Art

[0002] The petrochemical energy shipping industry refers to the transportation of petrochemical energy and its products by sea. Petrochemical energy, primarily including oil and natural gas, plays a vital role in the national economy and is widely used in industrial production, transportation, and other fields. The petrochemical energy shipping industry involves logistics and transportation from production sites to consumption sites, including the maritime transport of crude oil, refined oil products, and liquefied natural gas (LNG).

[0003] In the petrochemical energy shipping industry, the transportation of bulk liquid cargoes (such as oil and chemicals) places extremely high demands on safety, environmental protection, and operational efficiency. The current bidding process often involves the following difficulties:

[0004] 1. Long preparation cycle: Bids usually include technical specifications, transportation plans, risk assessments, emergency plans, and other aspects, and must meet the format and compliance requirements of the tenderer. Manual preparation takes a long time.

[0005] 2. Insufficient content matching accuracy: The requirements of different bidding projects vary significantly, and manual comparison and selection of relevant chapters are prone to omissions or duplications, especially in risk control and compliance clauses.

[0006] 3. Dispersed data sources: To accurately describe information such as ship performance, transportation routes, and historical operation cases, a large amount of information must be extracted from internal and external databases of the enterprise, and manual organization is inefficient.

[0007] 4. Complicated format and version management: Bids often need to be compiled according to the tenderer's template. Multiple rounds of revisions or internal collaborations can easily lead to version conflicts and format errors, affecting the quality of the final document and submission efficiency.

[0008] Based on the above, a large model-driven intelligent bid generation method and system for the petrochemical energy shipping industry is invented, which uses systematic technical means to automatically analyze bidding requirements, intelligently call professional knowledge and quickly output compliant documents, thereby improving the efficiency of bid preparation and ensuring high quality and consistency. Summary of the Invention

[0009] To solve the above technical problems, according to one aspect of the present invention, the present invention provides the following technical solutions:

[0010] A large model-driven intelligent tender document generation method for the petrochemical energy shipping industry includes the following specific steps:

[0011] S1: The multi-source data acquisition module acquires raw data from internal enterprise databases and external industry organization databases and stores them in a structured manner. After storage, the knowledge graph construction module associates key concepts in the petrochemical energy and shipping fields in the knowledge graph, enabling the system to quickly retrieve or recommend text fragments based on semantic relevance.

[0012] S2: The bidding requirements parsing module uses natural language processing and a large model of the petrochemical energy and shipping industries to perform semantic understanding and keyword extraction on bidding documents to obtain the tenderer's detailed requirements for transportation safety, ship performance, and emergency measures. Once obtained, the function tag dynamic mapping module maps the extracted requirements to the corresponding function tags in the knowledge graph to screen out standard documents that can be directly referenced and new content that needs to be generated.

[0013] S3: The template library reuse module presets bid document templates for various industries or specialized projects. If the requirements match existing template library entries, the relevant content is directly retrieved and necessary revisions or enhancements are made. If the template library cannot meet the requirements or personalized expression is required, the large model creation module calls the deep learning language model to generate professional text, covering technical descriptions, risk control points, and project planning. After generation, it automatically undergoes compliance review to prevent errors or inappropriate content.

[0014] S4: The style engine module enables the customized typesetting engine to automatically format text blocks, charts, and attachments according to the tenderer's specified format, while retaining the original data information to facilitate tracing the source of the reference. Next, the deviation table automation module automatically searches for differences between the tender content and the requirements of the tender documents and organizes the differences into a deviation table. At the same time, it prompts the editor to provide further explanation or evidence in the text, reducing the burden of manual omission detection.

[0015] S5: The cloud-based collaborative editing module supports multi-user concurrent writing and approval. It also uses permission management and online tagging to ensure team members are aware of document updates in real time. Furthermore, the multi-version traceability module automatically records modification logs after each bid revision, retaining all historical versions and supporting difference comparisons, facilitating future project review and quality inspection traceability.

[0016] S6: The regulatory compliance detection module matches relevant industry standards and shipping specifications to test the compliance of the bid documents with respect to safe operations, environmental protection, and emergency plans. If there are any deficiencies, feedback will be provided to the editor for improvement. Then, the bidding decision guidance module will integrate historical bidding data and competitor information to estimate the quality of the bid, the rationality of the bid price, and the probability of winning the bid, and generate a visual report to assist management in decision-making.

[0017] A large-scale model-driven intelligent tender document generation system for the petrochemical energy shipping industry, including:

[0018] The data aggregation and knowledge base construction module is used to first collect multi-source data and then build a knowledge graph;

[0019] The large model parsing and demand matching module is used to first understand the semantics of the bidding documents and extract keywords, and then map the extracted requirements with the corresponding functional labels in the knowledge graph;

[0020] A two-level content generation mechanism module is used to preset bid document templates for various industries or specific projects, and can generate professional text when the bid document template cannot meet the needs or requires personalized expression;

[0021] Automatic typesetting and deviation table generation module, used to program the customized typesetting engine and organize the differences between the bid content and the requirements of the bidding documents into a deviation table;

[0022] Collaboration management and version control module, used to support multi-user concurrent writing and approval, and can record modification logs after each bid revision;

[0023] The compliance review and risk decision support module is used to detect the compliance level of the bid content and can estimate the quality of the bid, the rationality of the bid price and the probability of winning the bid.

[0024] As a preferred solution of the large model driven intelligent tender document generation system for the petrochemical energy shipping industry described in the present invention, the data aggregation and knowledge base construction module includes:

[0025] Multi-source data acquisition module, used to obtain original data from the company's internal database and external industry organization databases and store them in a structured manner;

[0026] The knowledge graph construction module is used to associate key concepts in the field of petrochemical energy and shipping in the knowledge graph, so that the system can quickly retrieve or recommend text fragments based on semantic relevance.

[0027] As a preferred solution of the large model driven intelligent tender document generation system for the petrochemical energy shipping industry described in the present invention, the large model parsing and demand matching module includes:

[0028] The bidding requirements parsing module uses natural language processing and a large model of the petrochemical energy and shipping industry to perform semantic understanding and keyword extraction on bidding documents to obtain detailed requirements from the tenderer regarding transportation safety, vessel performance, and emergency measures.

[0029] The function label dynamic mapping module is used to map the extracted requirements with the corresponding function labels in the knowledge graph to filter out standard documents that can be directly referenced and new content that needs to be generated.

[0030] As a preferred solution of the large model driven intelligent tender document generation system for the petrochemical energy shipping industry described in the present invention, the two-level content generation mechanism module includes:

[0031] The template library reuse module is used to preset bid document templates for various industries or specific projects. If the requirements match the existing template library entries, the relevant content can be directly retrieved and necessary revisions or enhancements can be made.

[0032] The large model creation module is used to call on the deep learning language model to generate professional text when the template library cannot meet the needs or personalized expression is required. It covers technical descriptions, risk control points, and project planning. After generation, it will automatically perform compliance review to prevent errors or inappropriate content.

[0033] As a preferred solution of the large model driven intelligent tender document generation system for the petrochemical energy shipping industry described in the present invention, the automatic typesetting and deviation table generation module includes:

[0034] The style engine module is used to enable the customized typesetting engine to automatically format text blocks, charts, and attachments according to the format specified by the tenderer, while retaining the original data information to facilitate tracing the source of the reference;

[0035] The deviation table automation module is used to automatically retrieve the differences between the bid content and the requirements of the bidding documents, and can organize the differences into a deviation table. At the same time, it can prompt the editor to provide further explanation or evidence in the text, reducing the burden of manual omission detection.

[0036] As a preferred solution of the large model-driven intelligent tender document generation system for the petrochemical energy shipping industry described in the present invention, the collaborative management and version control module includes:

[0037] The cloud-based collaborative editing module supports concurrent writing and approval by multiple users, and ensures that team members can understand document updates in real time through permission management and online markup functions.

[0038] The multi-version traceability module is used to automatically record modification logs after each bid revision, retain all historical versions, and support difference comparison, making it convenient for future project review and quality inspection traceability.

[0039] As a preferred solution of the large model-driven intelligent tender document generation system for the petrochemical energy shipping industry described in the present invention, the compliance review and risk decision support module includes:

[0040] The regulatory compliance testing module is used to check the compliance of bid documents with respect to safe operations, environmental protection, and emergency response plans by matching relevant industry standards and shipping regulations. If any deficiencies are found, feedback will be provided to the editor for improvement.

[0041] The bidding decision guidance module is used to integrate historical bidding data and competitor information, estimate the quality of bid documents, the rationality of bid prices and the probability of winning the bid, and generate visual reports to assist management in decision-making.

[0042] Compared with existing technologies:

[0043] 1. To address the long preparation cycle: The multi-source data acquisition module is used to structure the storage of original data, the knowledge graph construction module associates key concepts to quickly retrieve recommended text fragments, the template library reuse module presets multiple templates that can be directly retrieved and revised, and the large model creation module generates professional texts. The collaborative work of multiple modules greatly reduces the time investment in manual bidding document writing and effectively shortens the preparation cycle.

[0044] 2. Addressing insufficient content matching accuracy: The bidding requirements parsing module uses natural language processing and large models to perform semantic understanding and keyword extraction on bidding documents. The function label dynamic mapping module maps the extracted requirements with knowledge graph function labels, screening standard documents and new content. This can more accurately match the requirements of different bidding projects, reduce omissions and duplications, and improve content matching accuracy, especially in terms of risk control and compliance clauses.

[0045] 3. Addressing the fragmentation of data sources: The multi-source data acquisition module obtains original data from internal enterprise and external industry organization databases and stores them in a structured manner. The knowledge graph construction module associates key concepts, making it convenient for the system to quickly retrieve or recommend text fragments based on semantic relevance. This solves the problem of fragmented data sources, improves the efficiency of data collation, and makes the acquisition of information such as ship performance, transportation routes, and historical operation cases more convenient and efficient.

[0046] 4. Addressing the cumbersome format and version management: The style engine module automatically formats text blocks, charts, and attachments according to the format specified by the tenderer and retains the original data information. The deviation table automation module automatically retrieves the differences, organizes them into a deviation table, and prompts editing. The cloud-based collaborative editing module supports multi-user parallel writing and approval, permission management, and online marking. The multi-version traceability module automatically records modification logs, retains historical versions, and supports difference comparison. These modules work together to effectively solve the cumbersome format and version management problems, improve document quality and submission efficiency, and facilitate project review and quality inspection traceability.

[0047] 5. Regarding bids and tenders: The regulatory compliance detection module matches industry standards and shipping specifications to detect compliance and provides feedback for improvement. The bidding decision guidance module integrates historical bidding data and competitor information to estimate bid quality, bid price rationality, and bid winning probability, and generates visual reports to assist in decision-making. This further improves bid quality and the scientific and rational nature of bids, increasing the chances of winning. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0049] To make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0050] The present invention provides a large model driven intelligent bidding document generation method for the petrochemical energy shipping industry. Figure 1 , including the following specific steps:

[0051] S1: The multi-source data acquisition module obtains raw data from internal enterprise databases (such as ship operation information, historical bidding cases, scheduling records, etc.) and external industry organization databases (such as relevant laws and regulations, environmental policies, international shipping standards, etc.) and performs structured storage. After storage, the knowledge graph construction module associates key concepts in the field of petrochemical energy shipping (such as dangerous goods categories, ship technical indicators, route restrictions, port regulations) in the knowledge graph, allowing the system to quickly retrieve or recommend text fragments based on semantic relevance;

[0052] S2: The bidding requirements parsing module uses natural language processing and a large model of the petrochemical energy and shipping industry to perform semantic understanding and keyword extraction on the bidding documents to obtain the tenderer's detailed requirements for transportation safety, ship performance, and emergency measures. Once obtained, the function label dynamic mapping module maps the extracted requirements to the corresponding function labels in the knowledge graph (such as "loading and unloading risk control" and "emission standard compliance") to screen out standard documents that can be directly referenced and new content that needs to be generated.

[0053] S3: The template library reuse module presets bid document templates for various industries or specialized projects. If the requirements match those of existing template library entries, the relevant content is directly retrieved and necessary revisions or enhancements (such as adding specific data) are made. If the template library cannot meet the requirements or personalized expression is required, the large model creation module calls the deep learning language model to generate professional text, covering technical descriptions, risk control points, and project planning. After generation, compliance review is automatically performed to prevent errors or inappropriate content.

[0054] S4: The style engine module enables the customized typesetting engine to automatically format text blocks, charts, and attachments according to the tenderer's specified format (such as title hierarchy, page layout, table fields, etc.), while retaining the original data information to facilitate tracing the source of the reference. Next, the deviation table automation module automatically searches for differences between the tender content and the requirements of the tender documents and organizes the differences into a deviation table. At the same time, it can prompt the editor to provide further explanation or evidence in the text, reducing the burden of manual omission detection.

[0055] S5: The cloud-based collaborative editing module supports multi-user concurrent writing and approval. It also uses permission management and online tagging to ensure team members are aware of document updates in real time. Furthermore, the multi-version traceability module automatically records modification logs after each bid revision, retaining all historical versions and supporting difference comparisons, facilitating future project review and quality inspection traceability.

[0056] S6: The regulatory compliance detection module matches relevant industry standards and shipping specifications to test the compliance of the bid documents with respect to safe operations, environmental protection, and emergency plans. If there are any deficiencies, feedback will be provided to the editor for improvement. Then, the bidding decision guidance module will integrate historical bidding data and competitor information to estimate the quality of the bid, the rationality of the bid price, and the probability of winning the bid, and generate a visual report to assist management in decision-making.

[0057] A large-model-driven intelligent tender document generation system for the petrochemical energy and shipping industries includes: a data aggregation and knowledge base construction module for first collecting multi-source data and then constructing a knowledge graph; a large-model parsing and demand matching module for first performing semantic understanding and keyword extraction on the tender documents, and then mapping the extracted demands with corresponding functional tags in the knowledge graph; a two-level content generation mechanism module for presetting tender document templates for various industries or segmented projects, and for generating professional text when the tender document templates cannot meet the demands or require personalized expression; an automatic typesetting and deviation table generation module for programming a customized typesetting engine and for organizing the differences between the tender document content and the requirements of the tender document into a deviation table; a collaborative management and version control module for supporting multi-user parallel writing and approval and for recording modification logs after each tender document revision; a compliance review and risk decision support module for detecting the degree of compliance in the tender document content and for estimating the tender document quality, the rationality of the bid price and the probability of winning the bid.

[0058] The data aggregation and knowledge base construction module includes: a multi-source data acquisition module, which is used to obtain original data from the company's internal database and external industry organization database and store it in a structured manner; a knowledge graph construction module, which is used to associate key concepts in the field of petrochemical energy and shipping in the knowledge graph, so that the system can quickly retrieve or recommend text fragments based on semantic relevance.

[0059] The large model parsing and demand matching module includes:

[0060] The bidding requirements parsing module is used to use natural language processing and the petrochemical energy shipping industry big model to perform semantic understanding and keyword extraction on the bidding documents to obtain the tenderer's detailed requirements for transportation safety, ship performance, and emergency measures; the function label dynamic mapping module is used to map the extracted requirements with the corresponding function labels in the knowledge graph to screen out standard documents that can be directly referenced and new content that needs to be generated.

[0061] The two-level content generation mechanism module includes: a template library reuse module, which is used to preset bid templates for various industries or segmented projects. If the requirements match the existing template library entries, the relevant content will be directly retrieved and necessary revisions or enhancements will be made; a large model creation module, which is used to call the deep learning language model to generate professional text when the template library cannot meet the requirements or personalized expression is required. It covers technical descriptions, risk control points, and project planning, and will automatically perform compliance review after generation to prevent errors or inappropriate content.

[0062] The automatic typesetting and deviation table generation module includes: a style engine module, which is used to enable the customized typesetting engine to automatically arrange text blocks, charts and attachments according to the format specified by the tenderer, while retaining the original data information to facilitate tracing the reference source; a deviation table automation module, which is used to automatically retrieve the differences between the bid content and the requirements of the tender documents, and can organize the differences into a deviation table. At the same time, it can prompt the editor to provide further explanation or evidence in the text, reducing the burden of manual omission detection.

[0063] The collaborative management and version control module includes: a cloud-based collaborative editing module, which supports multi-user concurrent writing and approval, and can ensure that team members can understand document updates in real time through permission management and online marking functions; a multi-version traceability module, which can automatically record modification logs after each bid revision, retain all historical versions and support difference comparison, facilitating future project review and quality inspection traceability.

[0064] The compliance review and risk decision support module includes: a regulatory compliance detection module, which is used to detect the compliance level of the bid content regarding safe operation, environmental protection, and emergency plans by matching relevant industry standards and shipping specifications, and provide feedback to the editor for improvement if there are any deficiencies; a bidding decision guidance module, which is used to integrate historical bidding data and competitor information, estimate the quality of the bid, the rationality of the bid price and the probability of winning the bid, and generate a visual report to assist management in decision-making.

[0065] Although the present invention has been described above with reference to embodiments, various modifications may be made thereto and equivalent components may be substituted without departing from the scope of the present invention. In particular, as long as there are no structural conflicts, the various features of the embodiments disclosed herein may be combined with each other in any manner, and the omission of an exhaustive description of such combinations in this specification is solely for the sake of space and resource conservation. Therefore, the present invention is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A large model driven intelligent bidding document generation method for the petrochemical energy shipping industry, characterized by: The specific steps are as follows: S1: The multi-source data acquisition module acquires raw data from internal enterprise databases and external industry organization databases and stores them in a structured manner. After storage, the knowledge graph construction module associates key concepts in the petrochemical energy and shipping fields in the knowledge graph, enabling the system to quickly retrieve or recommend text fragments based on semantic relevance. S2: The bidding requirements parsing module uses natural language processing and a large model of the petrochemical energy and shipping industries to perform semantic understanding and keyword extraction on bidding documents to obtain the tenderer's detailed requirements for transportation safety, ship performance, and emergency measures. Once obtained, the function tag dynamic mapping module maps the extracted requirements to the corresponding function tags in the knowledge graph to screen out standard documents that can be directly referenced and new content that needs to be generated. S3: The template library reuse module presets bid document templates for various industries or specialized projects. If the requirements match existing template library entries, the relevant content is directly retrieved and necessary revisions or enhancements are made. If the template library cannot meet the requirements or personalized expression is required, the large model creation module calls the deep learning language model to generate professional text, covering technical descriptions, risk control points, and project planning. After generation, it automatically undergoes compliance review to prevent errors or inappropriate content. S4: The style engine module enables the customized typesetting engine to automatically format text blocks, charts, and attachments according to the tenderer's specified format, while retaining the original data information to facilitate tracing the source of the reference. Next, the deviation table automation module automatically searches for differences between the tender content and the requirements of the tender documents and organizes the differences into a deviation table. At the same time, it prompts the editor to provide further explanation or evidence in the text, reducing the burden of manual omission detection. S5: The cloud-based collaborative editing module supports multi-user concurrent writing and approval. It also uses permission management and online tagging to ensure team members are aware of document updates in real time. Furthermore, the multi-version traceability module automatically records modification logs after each bid revision, retaining all historical versions and supporting difference comparisons, facilitating future project review and quality inspection traceability. S6: The regulatory compliance detection module matches relevant industry standards and shipping specifications to test the compliance of the bid documents with respect to safe operations, environmental protection, and emergency plans. If there are any deficiencies, feedback will be provided to the editor for improvement. Then, the bidding decision guidance module will integrate historical bidding data and competitor information to estimate the quality of the bid, the rationality of the bid price, and the probability of winning the bid, and generate a visual report to assist management in decision-making.

2. A large model-driven intelligent tender document generation system for the petrochemical energy shipping industry, characterized by: include: The data aggregation and knowledge base construction module is used to first collect multi-source data and then build a knowledge graph; The large model parsing and demand matching module is used to first understand the semantics of the bidding documents and extract keywords, and then map the extracted requirements with the corresponding functional labels in the knowledge graph; A two-level content generation mechanism module is used to preset bid document templates for various industries or specific projects, and can generate professional text when the bid document template cannot meet the needs or requires personalized expression; Automatic typesetting and deviation table generation module, used to program the customized typesetting engine and organize the differences between the bid content and the requirements of the bidding documents into a deviation table; Collaboration management and version control module, used to support multi-user concurrent writing and approval, and can record modification logs after each bid revision; The compliance review and risk decision support module is used to detect the compliance level of the bid content and can estimate the quality of the bid, the rationality of the bid price and the probability of winning the bid.

3. The large model driven intelligent tender document generation system for the petrochemical energy shipping industry according to claim 2 is characterized in that: The data aggregation and knowledge base building module includes: Multi-source data acquisition module, used to obtain original data from the company's internal database and external industry organization databases and store them in a structured manner; The knowledge graph construction module is used to associate key concepts in the field of petrochemical energy and shipping in the knowledge graph, so that the system can quickly retrieve or recommend text fragments based on semantic relevance.

4. The large model driven intelligent tender document generation system for the petrochemical energy shipping industry according to claim 2 is characterized in that: The large model parsing and demand matching module includes: The bidding requirements parsing module uses natural language processing and a large model of the petrochemical energy and shipping industry to perform semantic understanding and keyword extraction on bidding documents to obtain detailed requirements from the tenderer regarding transportation safety, vessel performance, and emergency measures. The function label dynamic mapping module is used to map the extracted requirements with the corresponding function labels in the knowledge graph to filter out standard documents that can be directly referenced and new content that needs to be generated.

5. The large model driven intelligent tender document generation system for the petrochemical energy shipping industry according to claim 2 is characterized in that: The two-level content generation mechanism module includes: The template library reuse module is used to preset bid document templates for various industries or specific projects. If the requirements match the existing template library entries, the relevant content can be directly retrieved and necessary revisions or enhancements can be made. The large model creation module is used to call on the deep learning language model to generate professional text when the template library cannot meet the needs or personalized expression is required. It covers technical descriptions, risk control points, and project planning. After generation, it will automatically perform compliance review to prevent errors or inappropriate content.

6. The large model driven intelligent tender document generation system for the petrochemical energy shipping industry according to claim 2 is characterized in that: The automatic typesetting and deviation table generation module includes: The style engine module is used to enable the customized typesetting engine to automatically format text blocks, charts, and attachments according to the format specified by the tenderer, while retaining the original data information to facilitate tracing the source of the reference; The deviation table automation module is used to automatically retrieve the differences between the bid content and the requirements of the bidding documents, and can organize the differences into a deviation table. At the same time, it can prompt the editor to provide further explanation or evidence in the text, reducing the burden of manual omission detection.

7. The large model driven intelligent tender document generation system for the petrochemical energy shipping industry according to claim 2 is characterized in that: The collaboration management and version control module includes: The cloud-based collaborative editing module supports concurrent writing and approval by multiple users, and ensures that team members can understand document updates in real time through permission management and online markup functions. The multi-version traceability module is used to automatically record modification logs after each bid revision, retain all historical versions, and support difference comparison, making it convenient for future project review and quality inspection traceability.

8. The large model driven intelligent tender document generation system for the petrochemical energy shipping industry according to claim 2 is characterized in that: The compliance review and risk decision support module includes: The regulatory compliance testing module is used to check the compliance of bid documents with respect to safe operations, environmental protection, and emergency response plans by matching relevant industry standards and shipping regulations. If any deficiencies are found, feedback will be provided to the editor for improvement. The bidding decision guidance module is used to integrate historical bidding data and competitor information, estimate the quality of bid documents, the rationality of bid prices and the probability of winning the bid, and generate visual reports to assist management in decision-making.

Images