Engineering co-counseling body intelligent recommendation, account distribution, evaluation method and system, and storage medium

Abstract

This invention discloses a method, system, and storage medium for intelligent recommendation, revenue sharing, and evaluation of engineering consortia, relating to the field of intelligent engineering management technology. It includes: constructing a three-layer evaluation model integrating geographical collaboration, collaborative networks, and value costs to achieve multi-dimensional intelligent recommendation of consortium entities; establishing a dynamic revenue sharing matrix based on the impact coefficient of engineering changes, and using smart contracts to automate revenue sharing at payment nodes and dynamically adapt proportions after changes; integrating customer evaluations, reverse checks and balances evaluations, and performance data to construct a full-dimensional dynamic credit evaluation system; and deeply linking credit scores with intelligent recommendations, financial pricing, and platform fees to form a credit-driven ecological cycle. The system includes functional units such as recommendation, revenue sharing, and credit, and is implemented using computer-readable storage media. This invention improves the adaptability of consortium formation and the stability of revenue sharing, building a healthy credit ecosystem.

Description

Technical Field

[0001] This invention relates to the field of intelligent engineering management technology, specifically to intelligent recommendation, revenue sharing, and evaluation methods, systems, and storage media for engineering consortia. Background Technology

[0002] As the construction industry deepens its digital and intelligent transformation, the Engineering, Procurement, and Construction (EPC) model has become mainstream, its core lying in the deep integration and collaboration of design, procurement, and construction. However, in actual project management, business management and technical execution are often disconnected, creating serious data silos. Existing project resource scheduling and team building rely heavily on manual experience or static bidding processes, lacking in-depth quantitative assessment of the geographical synergy, historical cooperation, and dynamic value-cost ratio among multiple stakeholders. This crude matching mechanism leads to long project setup cycles, high communication costs, and is highly susceptible to delays due to unfamiliarity between stakeholders. More seriously, the project's fund settlement and revenue sharing processes have long suffered from low transparency and delayed response. Faced with frequent design changes and on-site approvals, the existing manual revenue sharing model struggles to dynamically adjust the rights, responsibilities, and benefits of all parties in real time, resulting in an imbalance in profit distribution and cash flow difficulties for upstream and downstream companies. Furthermore, the industry lacks a dynamic credit evaluation system spanning the entire lifecycle; the performance of stakeholders cannot be converted into credit assets in real time, hindering the formation of a healthy ecosystem of survival of the fittest.

[0003] While some engineering management systems based on information technology have emerged in the current technology field, attempting to solve resource integration and payment issues through digital tools, most remain at the level of process digitization or single-dimensional optimization, failing to address the core logic of business-technology two-way coupling and dynamic credit closure in engineering collaboration. When building engineering teams, existing technologies often only consider the qualifications or pricing of a single entity, neglecting the synergistic effects at the combined level. Key variables such as the historical cooperation between the design and construction parties and the implicit impact of geographical location on project response speed are not included in the evaluation model. In the revenue sharing and scheduling stages, most systems still rely on preset static rules, lacking adaptive adjustment mechanisms based on real-time progress deviations, cost performance, and resource load rates. Especially when facing engineering changes, existing systems cannot automatically reconstruct the revenue sharing matrix or dynamically adjust resource scheduling strategies, resulting in poor management flexibility and weak anti-interference capabilities. More fundamentally, the industry lacks a mechanism to deeply bind credit evaluation with business rules (such as recommendation weights and service rates), causing credit scores to become static display data, unable to drive the self-optimization of the ecosystem. These technological bottlenecks make it difficult for engineering projects to achieve breakthroughs in resource allocation efficiency, risk management capabilities, and ecological health.

[0004] To address the aforementioned disconnect between business and technology, Chinese invention patent application publication number CN120875484A (EPC Engineering Information Management System Supporting Two-Way Matching of Business Strategy and Technology Path) proposes an improved solution. This patented technology acquires business, technology, and external environment data through a data acquisition and integration unit. It then uses a business strategy unit to construct a resource optimization allocation model and generate resource allocation schemes. Next, through a technology path unit combined with a BIM model, resources are mapped to construction nodes, generating a candidate set of technology paths that meet business constraints. Finally, a two-way matching unit constructs a two-way coupled optimization model, employing a non-dominated sorting genetic algorithm (NSGA-II) for multi-objective solution, yielding the optimal implementation scheme in terms of cost, risk, schedule, and efficiency. This technology clearly points out the drawbacks of the disconnect between business decisions and technical execution in existing systems. By introducing constraints such as contract cycles and risk coefficients, it achieves rigid constraints of business strategies on technology paths and feedback optimization of business decisions based on technical data, significantly improving the overall feasibility of the solution. However, in-depth analysis reveals significant shortcomings in this technical solution: First, its core focus is on the "business-technology" two-way matching within a single project, lacking an intelligent recommendation mechanism for multi-entity combinations such as "project consortia." It fails to consider external variables affecting team collaboration efficiency, such as geographical collaboration indices and collaboration network indices among different enterprises, potentially leading to high costs associated with the recommended implementation entities during actual cooperation. Second, its revenue-sharing logic primarily serves internal resource allocation, lacking a dynamic revenue-sharing matrix reconstruction mechanism based on the proportion of responsibility contributions among multiple entities. In particular, it cannot automatically adjust the revenue distribution ratio of each participant when project changes occur, still requiring manual intervention and negotiation. Finally, the system has not established a dynamic credit model based on multi-party reverse checks and balances evaluation, resulting in lagging credit data updates and an inability to form a business closed loop driving ecological optimization, making it difficult to incentivize entities to proactively improve performance quality.

[0005] To further address the dynamic adaptability issue of resource scheduling, Chinese invention patent application publication number CN120525244A (A Dynamic Resource Scheduling Method for EPC Projects Based on Multi-Objective Optimization) proposes a new technical approach. This patent introduces a real-time feedback-driven adaptive adjustment mechanism based on existing static scheduling. By constructing a multi-objective function encompassing time, cost, and resource utilization, an optimal resource scheduling scheme is generated using a genetic algorithm. During execution, schedule deviation rate, cost performance index, and resource load rate are collected in real time, and a comprehensive evaluation value is calculated after normalization. Once the evaluation value falls below a threshold, the weights of the multi-objective function are dynamically adjusted based on the indicator with the highest deviation (e.g., increasing the weight of the time objective if the schedule deviation is high), and a new scheduling scheme is generated. This technology effectively solves the problem that existing static planning techniques struggle to respond to dynamic changes on-site, achieving adaptive optimization of resource scheduling schemes as the project environment changes, and significantly reducing the risks of project delays and cost overruns. However, this technical solution still reveals significant limitations in practical applications: First, its optimization is limited to the matching of activities and resources within the project, completely neglecting the formation of a consortium and the distribution of benefits among multiple stakeholders, thus failing to resolve the selection difficulties in the project contracting phase and subsequent disputes over revenue sharing; Second, its dynamic adjustments only target resource input and schedule arrangements, failing to delve into the revenue sharing matrix, i.e., it cannot dynamically adjust the revenue distribution ratio of each stakeholder based on the changed actual rights, responsibilities, and interests, easily leading to new conflicts of interest in multi-party collaboration; Third, the system also lacks a mechanism to convert performance into dynamic credit scores, and it does not embed credit scores as a core factor into the business rules of recommendation ranking, financial pricing, and platform fees, lacking a credit-driven business flywheel mechanism and failing to achieve self-reinforcing ecosystem growth.

[0006] In summary, while CN120875484A has made progress in the two-way matching of business and technology, and CN120525244A has achieved breakthroughs in dynamic resource scheduling, neither has fundamentally solved the problems of the scientific nature of the engineering consortium formation, the fairness of revenue sharing under complex changes, and the endogenous driving force of the credit ecosystem. CN120875484A focuses on scheme optimization within a single project, lacking a multi-entity collaborative assessment and dynamic revenue sharing mechanism; CN120525244A focuses on adaptive resource scheduling within a single project, lacking the distribution of benefits among multiple entities and the construction of a credit closed loop. Existing technologies lack a comprehensive model that fully evaluates candidate combinations from three dimensions—geographical space, historical cooperation networks, and value cost—when dealing with the combined effects of multi-entity collaboration. When facing high-frequency, dynamic engineering changes, there is a lack of automated mechanisms for dynamically reconstructing the revenue-sharing matrix based on responsibility contributions, making it difficult to balance revenue-sharing efficiency and accuracy. In terms of credit system construction, a dynamic evaluation model integrating customer assessment, internal checks and balances, and objective performance data has not yet been formed, nor has credit score been embedded as a core factor into the business closed loop of recommendation weighting, financial service pricing, and platform service fee adjustments. These technological bottlenecks lead to inefficient resource allocation in engineering projects, high collaboration risks, and a failure to foster a healthy industry ecosystem based on merit.

[0007] Therefore, there is an urgent need for an intelligent management technology for engineering consortia that can adapt to the needs of specific engineering scenarios, achieve dynamic and stable adjustment of change-based revenue sharing, and build an ecosystem-driven mechanism based on multi-dimensional dynamic credit scores. This technology would enable intelligent management of the entire process of engineering consortia, from demand analysis, combination recommendation, automatic revenue sharing to credit assessment and ecosystem-driven management. Summary of the Invention

[0008] Based on the aforementioned technical problems, this application discloses a method, system, and storage medium for intelligent recommendation, revenue sharing, and evaluation of engineering consortia; the method for intelligent recommendation, revenue sharing, and evaluation of engineering consortia includes: Obtain project requirements information, which includes at least the project type, construction location, technical parameters, cost budget, and schedule requirements; Based on the project requirements information, candidate service providers in the categories of design, construction, production and processing, and supervision are selected from the engineering service resource library to construct multiple candidate combinations of engineering contracting entities; A fusion evaluation model is constructed, and the comprehensive recommendation score of each candidate combination of the project contracting entity is obtained through the fusion evaluation model. The project contracting entity recommendation results are then sorted in descending order according to the comprehensive recommendation score and output. Based on the internal collaboration agreement of the selected project consortium, the payment node smart contract and the revenue sharing matrix are initialized, wherein the revenue sharing matrix defines the revenue distribution ratio of each service entity in each payment node; Perform project obligations, collect performance evidence verified by multiple parties, and trigger the smart contract execution of the corresponding payment node for automatic revenue distribution; At project performance milestones and upon project completion, we collect customer evaluation data on the engineering consortium and reverse checks and balances evaluation data among the various service providers within the consortium. We then combine this data with objective performance data to generate dynamic credit scores for each service provider. The dynamic credit score is embedded into a pre-built business flywheel model of the engineering consortium to achieve credit-driven operation of the engineering consortium ecosystem.

[0009] Furthermore, the fusion evaluation model adopts a three-layer fusion evaluation architecture, including a geographical collaboration index, a collaboration network index, and a value cost index, to calculate the comprehensive recommendation score for each candidate combination of the project contractors, specifically as follows: Geographic Coordination Index Collaborative Network Index Value Cost Index Normalize each value separately to obtain the normalized value. , , ; Through formula Calculate the comprehensive recommendation score ,in For geographical collaboration weighting coefficient, For the weight coefficients of the collaborative network, These are the value cost weighting coefficients, and each weighting coefficient is dynamically configured according to the project type.

[0010] Furthermore, the geographical coordination index The formula is: ,in The number of service providers in the project consortium. As the reference time constant, The time normalization coefficient is... For the first Real-time access time from each service provider to the project site For the first The same-city service load factor of each service entity For the first The and the first Real-time availability between service providers.

[0011] Furthermore, the formula for the cooperative network index N is: ,in Candidate combinations for engineering consortia Historical average weight of cooperation Candidate combinations The density of collaborative networks, Candidate combinations The maximum path length between nodes. Candidate combinations The average path length between nodes.

[0012] Furthermore, the value cost index The formula is: ,in For the first The weight of each cost item For the first The industry benchmark value for each cost item. For the first The price quote for each cost item. For the set of value points, For the first A rating of each value point. For the first The weight index of each value point This is a penalty function based on project risk.

[0013] Furthermore, the automatic revenue sharing also includes dynamically adjusting the amounts and revenue sharing matrix of the affected payment nodes when engineering changes occur, specifically: Identify the affected payment nodes corresponding to engineering changes Through formula Calculate the new contract amount for this node. ,in This is the original contract amount for that node. The total amount may be increased or decreased due to project changes. This is the proportionality coefficient by which the workload of this node is affected by the change. Based on the responsibility contribution allocation plan after the project changes, adjust the revenue sharing matrix of the affected payment nodes to ensure that the allocation ratio of each service entity matches the actual rights, responsibilities and benefits.

[0014] Furthermore, the generation of dynamic credit scores for each service provider by combining objective performance data includes: Through formula computing service providers At any moment Dynamic credit score ,in Historical credit decay factor For service providers At any moment Credit score, Customer evaluation weighting, To provide a reverse check and balance on the evaluation weights, For objective performance data weighting, Customer rating index As a reverse check and balance evaluation index, This is an objective performance data index.

[0015] Furthermore, the feature is that the engineering consortium's business flywheel model includes configuration rules for intelligent recommendation weights, financial service pricing, and platform service fees, specifically: Dynamic credit score The comprehensive recommendation score calculation model, which incorporates core factors into intelligent recommendation, enables combinations of high-credit service providers to receive recommendation weight bonuses. Through formula Determine the service provider Supply chain loan interest rates ,in Based on the basic lending rate, This refers to the credit premium coefficient. Through formula Determine the service provider Platform service fee rate ,in Basic service fee rate, This is the credit discount coefficient.

[0016] The aforementioned intelligent recommendation, revenue sharing, and evaluation system for engineering consortia includes: The project requirements acquisition unit receives project requirements information and performs structured parsing. The project consortium construction unit selects candidate service providers from the engineering service resource library and constructs multiple sets of engineering consortium candidate combinations; The fusion evaluation unit calculates the comprehensive recommendation score of candidate combinations and outputs the recommendation results through a three-layer architecture model; The smart contract revenue sharing unit initializes the payment node smart contract and revenue sharing matrix, triggers automatic revenue sharing after verification of performance evidence, and realizes dynamic adjustment of revenue sharing after engineering changes. The multi-dimensional credit assessment unit collects customer evaluations, reverse checks and balances evaluations, and objective performance data to calculate the dynamic credit score of each service provider. The business flywheel model unit embeds dynamic credit scores into the configuration rules of intelligent recommendation, financial pricing, and platform fees, and executes corresponding weight and fee adjustments.

[0017] A computer-readable storage medium having a computer program stored thereon, characterized in that, when executed by a processor, the program implements any one of the following methods for intelligent recommendation, revenue sharing, and evaluation of an engineering consortium.

[0018] Compared with the prior art, the technical solution of this application has the following technical effects: This invention constructs a three-layer evaluation model that integrates geographical collaboration, collaborative networks, and value cost to achieve multi-dimensional intelligent recommendation of engineering consortium combinations. It breaks through the limitations of single-dimensional screening, accurately matches project needs with service provider capabilities, significantly improves the scientific nature and adaptability of consortium formation, and lays the foundation for efficient project advancement.

[0019] This invention achieves automated payment node allocation and dynamic adjustment after engineering changes by deeply binding smart contracts and revenue sharing matrices. It can complete amount calculation and ratio adaptation without manual intervention, ensuring the fairness, transparency, accuracy and efficiency of the revenue sharing process, effectively reducing revenue sharing disputes and improving project performance and collaboration efficiency.

[0020] By integrating customer evaluations, reverse checks and balances evaluations, and objective performance data, this invention constructs a multi-dimensional dynamic credit evaluation system, generating a credit score that reflects the performance capabilities of service providers in real time. This breaks through the rigid limitations of static credit evaluations and provides reliable credit support for the healthy development of the ecosystem.

[0021] This invention embeds dynamic credit scores into a business flywheel model, enabling deep integration of credit with intelligent recommendations, financial pricing, and platform fees. This creates a self-reinforcing cycle of high credit, superior resources, strong performance, and even higher credit, continuously optimizing the resource allocation efficiency of the engineering consortium ecosystem and driving the industry towards intelligence and standardization.

[0022] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the preferred embodiments of this application are described in detail below with reference to the accompanying drawings.

[0023] The above and other objects, advantages and features of this application will become more apparent to those skilled in the art from the following detailed description of specific embodiments in conjunction with the accompanying drawings. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In all drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0025] Based on the description of the figures and their corresponding technical content in the document, the titles of the figures are as follows: Figure 1A schematic diagram illustrating the entire process of intelligent recommendation, revenue sharing, and evaluation methods for project consortia; Figure 2 A three-tiered evaluation model architecture diagram integrating geographical collaboration, collaborative networks, and value cost index; Figure 3 Schematic diagram of the overall structure of the intelligent management system for the project consortium; Figure 4 A diagram showing the data interaction between computer-readable storage media, processor, and external systems; Figure 5 A comparison chart of the accuracy of recommendations for HVAC system construction projects and various technological trends; Figure 6 A comparative chart of the trends in revenue sharing time and accuracy for HVAC system construction projects; Figure 7 Contribution of the Integrated Evaluation Index for HVAC System Construction Projects and Comparison Chart of Multiple Technologies; Figure 8 Comparison chart of three primary indicators and multiple technologies in HVAC fresh air system renovation projects; Figure 9 A multi-technology growth curve chart of four indicators driving the credit ecosystem of HVAC fresh air system renovation projects; Figure 10 Comprehensive trend chart of key technical indicators for steel structure and steel-concrete building construction projects. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. In the following description, specific details such as specific configurations and components are provided merely to help fully understand the embodiments of this application. Therefore, those skilled in the art should understand that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this application. In addition, for clarity and brevity, descriptions of known functions and structures are omitted in the embodiments.

[0027] It should be understood that the phrase "an embodiment" or "this embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of this application. Therefore, "an embodiment" or "this embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments.

[0028] Furthermore, reference numerals and / or letters may be repeated in different examples within this application. Such repetition is for the purpose of simplification and clarity and does not in itself indicate a relationship between the various embodiments and / or settings discussed.

[0029] In this article, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, B exists alone, and A and B exist simultaneously. The term " / and" in this article describes another type of relationship between related objects, indicating that two relationships can exist. For example, A / and B can mean: A exists alone, and A and B exist alone. In addition, the character " / " in this article generally indicates that the related objects before and after it are in an "or" relationship.

[0030] In this article, the term "at least one" is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, "at least one of A and B" can mean: A exists alone, A and B exist simultaneously, or B exists alone.

[0031] It should also be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion.

[0032] Example 1

[0033] This embodiment mainly describes an intelligent recommendation, revenue sharing, and evaluation method for engineering consortia, such as... Figure 1 As shown, it specifically includes: Obtain project requirements information, which includes at least the project type, construction location, technical parameters, cost budget, and schedule requirements; Based on the project requirements information, candidate service providers in the categories of design, construction, production and processing, and supervision are selected from the engineering service resource library to construct multiple candidate combinations of engineering contracting entities; A fusion evaluation model is constructed, and the comprehensive recommendation score of each candidate combination of the project contracting entity is obtained through the fusion evaluation model. The project contracting entity recommendation results are then sorted in descending order according to the comprehensive recommendation score and output. Based on the internal collaboration agreement of the selected project consortium, the payment node smart contract and the revenue sharing matrix are initialized, wherein the revenue sharing matrix defines the revenue distribution ratio of each service entity in each payment node; Perform project obligations, collect performance evidence verified by multiple parties, and trigger the smart contract execution of the corresponding payment node for automatic revenue distribution; At project performance milestones and upon project completion, we collect customer evaluation data on the engineering consortium and reverse checks and balances evaluation data among the various service providers within the consortium. We then combine this data with objective performance data to generate dynamic credit scores for each service provider. The dynamic credit score is embedded into a pre-built business flywheel model of the engineering consortium to achieve credit-driven operation of the engineering consortium ecosystem.

[0034] Furthermore, the quantitative and structured collection of project requirement information is achieved by defining a quantitative representation system for project requirement information, with a unique identifier for each project. (Project-specific code, character type), project type is quantified using industry-standard coding. ( These correspond to the primary categories and sub-categories of engineering projects such as construction, municipal engineering, industry, transportation, and water conservancy (coded type). These correspond to primary categories such as architecture, municipal engineering, and industry, as well as their sub-categories. Construction sites utilize geospatial quantification models to collect precise latitude and longitude coordinates. ,in Longitude of the project location; The latitude of the project location, along with the administrative region level code. ( It is a provincial-level administrative code. It is a municipal-level administrative code. (This refers to the district / county level administrative code, where || is the code concatenation character), and calculates the distance from the project location to the regional transportation hub. ,in Longitude of regional transportation hub The latitude of regional transportation hubs is used, and the result is a spherical straight-line distance, providing basic parameters for geographic collaborative computing; technical parameters are used to construct a multi-dimensional quantitative index matrix. ,in The number of dimensions for technical indicators; The number of indicator attributes. These are technical indicators (such as structural strength, environmental standards, and process precision). These are the indicator attributes (standard value, lower limit of acceptance, upper limit of acceptance). ( This is the industry standard value for the indicator. This is the lower limit of the indicator's qualification level. (This refers to the upper limit of the acceptable value for each indicator), and each indicator is assigned an importance weight. (No. (Importance weight of each technical indicator), and at the same time Cost budget quantified into total budget (Overall project cost budget) and itemized budget vector ,in To design itemized budgets, For the construction item budget, For the production and processing item budget, The budget for each supervision item is divided into four core cost categories: design, construction, production and processing, and supervision. Each item's budget includes a fluctuation factor. , No. If the allowable fluctuation percentage for each item's budget is specified, then the budget range for each item is as follows: The project duration requirement is quantified as the total project duration. (Overall project duration from commencement to completion) and key milestone duration vectors , The number of key performance nodes; The planned completion times for each key milestone are as follows: For design delivery nodes, For each completion milestone, a time flexibility coefficient is marked. , For the first The allowable flexibility in the project duration for each key node is as follows: the actual completion window for each node is... All quantitative data are uniformly entered into the project requirements database. (Structured project requirements data repository), and validated using data validation formulas. ,in For indicator functions, hour ,otherwise ; For the valid value range of each data item, Time data is verified; Furthermore, a candidate combination of engineering contracting entities is constructed. Candidate entities are screened and combined from the engineering service resource library. Precise screening is achieved through matching degree calculation, and the engineering service resource library is defined. The quantitative structure of the (structured repository of full-dimensional information on engineering service providers), with each service provider in the repository... Information is represented as vectors ( A unique character-based code for the service provider. Identifier for the main body type; A quantitative coefficient for qualification level; For technical parameter matching degree; To serve the service area coverage radius; The coordinates of the main permanent residence location; This is the quotation coefficient; (This is the construction period guarantee coefficient), where (1 = Design, 2 = Construction, 3 = Production and Processing, 4 = Supervision, numerical identifiers) correspond to the four main categories of design, construction, production and processing, and supervision, respectively. A quantitative coefficient for qualification level (mapped from low to high according to industry qualification level, with values ​​ranging from low to high). ), Technical parameter matching degree (the degree of matching between the main technical capabilities and the project's technical parameters, with a value of...) ), The service area coverage radius (the geographical range within which the main body can provide effective services). The latitude and longitude of the main permanent residence ( Longitude as the main axis; (main latitude) This is the pricing coefficient (the ratio of the main item price to the industry benchmark price). The construction period guarantee coefficient (the coefficient of the main structure's ability to complete the construction period according to plan, with a value of...) The core of single-category subject screening is calculating the matching degree. (No. The overall matching degree between the service provider and the project, with a value of [value]. The formula for matching the design subject is: ,in For qualification weighting, As a weight for technical matching degree, For regional matching degree weight, As a weight for price matching degree, Weighting for project schedule matching degree; To ensure the matching degree of the main technical parameters of the design category; The straight-line distance from the main body to the project site; This refers to the pricing coefficient for design-related main components; (This refers to the main construction period guarantee coefficient for design-related projects), where... , The longitude of the main permanent residence; The latitude of the main building's permanent location is given, and the result is a straight-line distance over a spherical surface. The straight-line distance from the main building to the project site is given. The formula for the matching degree of construction-related main structures is as follows: ,in For the matching degree of the main technical parameters of the construction project; This refers to the pricing coefficient for the main construction components; This is the construction period guarantee factor for the main construction project; the definitions of other parameters are the same. The formula for matching the main body of manufacturing and processing is as follows: ,in For the matching degree of main technical parameters of production and processing products; The pricing coefficient for manufacturing and processing entities; This is a core production and delivery milestone for the project. For production and processing main components, the project schedule guarantee coefficient is... These are natural constants; the other parameters are defined the same way. ),in For core production and delivery nodes; the formula for matching the main body of the supervision category is: ,in For the matching degree of the main technical parameters of the supervision category; For the pricing coefficient of the main body of the supervision category; This is the main construction period guarantee factor for supervision projects; the definitions of other parameters are the same. ); Set the filtering threshold (Minimum matching degree of the subject to be selected into the candidate pool, dimensionless), to select The main body forms four categories of candidate subject sets. (Collection of candidate subjects for design category) (Collection of candidate entities for construction projects) (Set of candidate entities in the production and processing category) (Candidate subject set for supervision). In the combination construction stage, candidate combinations are generated using the Cartesian product rule. Let the combination... ( For design-related main body, For construction-related main structures, For manufacturing and processing entities, (for supervision entities), among which , , , Then the set of all candidate combinations is ( (where is the total number of candidate combinations, a positive integer) ( for The number of elements in the set (and so on), each combination is assigned a unique identifier. (No. Each candidate combination is assigned a unique character encoding, and the quantized information vectors of all entities within the combination are associated with it to form a basic database of the combinations. (A structured repository of full-dimensional information on candidate combinations for engineering projects).

[0035] Furthermore, such as Figure 2 As shown, a fusion evaluation model is constructed to calculate the comprehensive recommendation score. Through three-layer index calculation, normalization processing, and weighted fusion, the combined comprehensive recommendation score is obtained, and the geographical synergy index is calculated. (Quantitative index of geospatial synergy capability of candidate combinations; the larger the value, the stronger the synergy), the formula is: ,in The number of main components within the combination (fixed to four categories: design, construction, production and processing, and supervision). This is the benchmark response time for the same city (the industry-standard basic response time for same-city engineering services). This is the time normalization coefficient. For the first The real-time reachability time from an entity to the project site is given by the formula. calculate( (The average driving speed on roads within the administrative region to which the project is located). This represents the average driving speed within the project area. The main urban service load factor (the first) The service resource occupancy coefficient of each entity within the same city as the project is calculated using the following formula: ( The number of projects under construction in the same city as the main entity; The core resource occupancy rate; This represents the maximum number of projects under construction in the same city for any entity within the industry. (the maximum core resource utilization rate of the main players in the industry), among which The main body is the number of projects under construction in the same city. For core resource utilization rate, , This is the industry maximum value. For the first With the The real-time reachability time between entities is given by the formula: ,in For the first The main longitude; For the first Each main latitude, For the first The main longitude, For the first Each main latitude, For the first The main longitude, For the first Individual dimensions. Calculate the collaborative network index. (A quantitative index of the historical cooperation and synergy capabilities of the main entities within the candidate combination; the larger the value, the stronger the synergy), the formula is as follows: ,in The average weight of historical cooperation in the combination (the overall quality coefficient of historical cooperation between subjects within the candidate combination, dimensionless, with values ​​ranging from...) The calculation formula is: ( The number of pairs of elements within the group. ; For the first With the (Cooperation weight of each entity) ( for and Number of collaborations; This represents the highest number of collaborations among entities within the industry. for and Average customer ratings for collaborative projects; for and The average timeframe for resolving issues through negotiation during cooperation; Standard timeframes for resolving industry-wide issues through negotiation; for and (Completion rate of cooperation agreement) for and Number of collaborations Rate the collaborative projects For the timeliness of problem resolution, To assess the degree of fulfillment of contractual obligations. ( (The number of historical cooperative relationships that actually exist within the group). This represents the number of actual cooperative relationships within the group. This is the maximum path length of the combined nodes (the longest path hop count between entities within the combined node that is connected through historical cooperation). ( For the first With the The length of the cooperation path between two entities, i.e., the minimum number of hops required for two entities to establish a cooperative connection through other entities (infinite if there is no direct cooperation). for and The path length. Next, the value cost index is calculated. (A quantitative index of the cost-effectiveness and value creation capability of candidate combinations; the higher the value, the stronger the cost-effectiveness and value creation capability.) The formula is as follows: ,in For the weight of the component costs (the first item) The weight of each item cost in the overall cost, dimensionless, with a value of [value missing]. ),satisfy . For the industry benchmark cost of the sub-items (the first The reasonable market cost of each item in the industry is calculated using the following formula: ( The average cost of each item in the project's region and for similar projects; (This is a project scale coefficient, quantified according to the project construction scale). This represents the average cost of similar projects in the region. This represents the project scale coefficient. For the combined itemized pricing (candidate combination for the first item) (Overall price for each item) ( (Itemized pricing for the corresponding type of main body within the combination). Rate the value (the first) Ability score for each value-added point, with a value... ), Value point weight (the first) The importance weight of each value point, with values... ),satisfy . (Project risk penalty coefficient, value) The higher the value, the higher the risk. For the first Risk weights for individual entities, with values... ; For the first The comprehensive risk coefficient of each entity, taking a value of (Weighted by credit, technology, and performance risk) As the main risk coefficient, Risk weights are then applied. Normalization is then performed, mapping the three major indices to... The interval, the formula is ( This represents the maximum geographical synergy index among all candidate combinations. (Minimum of the geographical synergy index for all candidate combinations) ( The maximum value of the collaborative network index for all candidate combinations. (Minimum of the collaborative network index for all candidate combinations) ( The maximum value cost index among all candidate combinations. (Minimum of the value cost index of all candidate combinations) ,but , , Similarly, calculate the overall recommendation score. (The comprehensive recommendation index of candidate combinations; the higher the value, the higher the recommendation priority), the formula is: ,in For geographical collaboration weighting coefficient, For the weight coefficients of the collaborative network, The value cost weighting coefficient satisfies Dynamically configured based on project type. (By...) The values ​​are sorted in descending order to generate recommendation results. ( The combination with the highest overall recommendation score. (for the lowest possible combination), satisfying .

[0036] Furthermore, the initialization of the payment node smart contract and the revenue sharing matrix enables the programmatic binding of payment rules. Specifically, this is achieved by dividing the project into payment nodes, assuming the project's payment node set is... ( (This represents the number of payment nodes, a positive integer, corresponding to key project performance nodes.) Each node represents a key node in the entire project performance process. The contract amount is (No. The contractual amount for each payment node must meet the following requirements. Initialize the payment node smart contract; the core parameters of the contract are: ( For the first Each payment node's smart contract has a unique character-based code. The corresponding project consortium combination code; For payment node identification; The amount of the node contract; This is the set of conditions for triggering the contract. Encode the corresponding revenue sharing matrix; (Contract status, character type), where The trigger condition set (quantified values ​​of preconditions for contract execution) is quantified as follows: ( The number of types of evidence of performance; For the first The node The verification score of the type of evidence; (to verify the threshold for eligibility of evidence). For the first Verification score for evidence of performance, To verify the threshold, This is an indicator function. The initialization of the accounting matrix uses... Two-dimensional matrix ( The number of main components within the combination is fixed at 4; Number of payment nodes; For the first The subject in the first (the profit distribution ratio of each node), where The number of main components within the combination. For the number of payment nodes, For the first The subject in the first The allocation ratio of each node. The calculation of the allocation ratio follows the principle of equal rights, responsibilities, and benefits, and the formula is as follows: ( For the first The subject in the first (responsibility weight of each node), where For the first The subject in the first The responsibility weight of each node ( As a workload weight, As a weight of responsibility, As a contribution weight; For the first The subject in the first The workload percentage of each node; For the first The subject in the first The responsibility percentage of each node; For the first The subject in the first (Contribution percentage of each node) , As a percentage of workload, As for the proportion of responsibility, This represents the percentage of contribution. Simultaneously, the constraints must be met. , The revenue sharing matrix With smart contracts Through unique identifier (Revenue sharing matrix-specific encoding, character type) Associated, the revenue sharing execution trigger formula is written into the contract. ( As a trigger identifier, When the revenue sharing is triggered, It is not triggered at times; (for the initial state of the contract), when At that time, the automatic revenue sharing process is initiated. All contract and revenue sharing matrix data are written to the blockchain, generating hash values. (SHA256 is a hash algorithm, and || is a data concatenation operator. It generates a unique and irreversible hash value, which is used for data anti-tampering verification) to ensure that the data cannot be tampered with.

[0037] Furthermore, the system implements smart contract triggering and automatic revenue sharing for project performance, automating performance evidence verification, contract triggering, and revenue sharing calculation. This is achieved by collecting performance evidence. The evidence set of each node is ( For the first The node (Original data of performance evidence), the formula for evidence verification score is as follows: ( To verify the subject's view on the first The node Scoring of class of evidence; (The maximum score for evidence) To verify the subject's rating, The formula for the comprehensive score based on multi-party verification is as follows: ( To verify the number of subjects; For the first The scoring weight of each verification subject; For the first The verification subject verifies the first... The node (Verification score of class evidence) To verify the subject weight, satisfy .when At that time, the contract trigger condition The smart contract status has been updated to "triggered". The core formula for calculating the automatic revenue sharing amount is: ( For the first The subject in the first (actual revenue amount of each node), of which For the first The subject in the first The node (i.e., the node) The actual revenue of each node. After the revenue sharing instruction is generated, the transfer is executed through the payment gateway. Upon completion, the contract status is updated to "executed" and the revenue sharing hash is recorded. ( This is the actual time of revenue sharing, in timestamp format, used for tracing revenue sharing records. If engineering changes occur, first calculate the total amount of the changes. (Increase or decrease in the total contract amount due to engineering changes, if it is an additional item) If the item is reduced, then Identify the set of affected nodes. ( (Number of payment nodes affected by engineering changes), calculate the workload impact coefficient for each affected node. (No. (the proportion of workload affected by the change for each affected node), satisfying... The formula for the new contract amount for affected nodes is as follows: ( For the first New contract amount for each affected node; For the first (Original contract amount for each affected node). Simultaneously, the liability weights of the affected nodes are recalculated. (After the engineering change) The subject in the first The new responsibility weights of each affected node), and the new allocation ratio formula is: ( For the first time after the engineering change The subject in the first (new allocation ratio for each affected node), satisfying The updated version and Synchronize to smart contracts, and subsequent revenue sharing will be based on... ( For the first time after the engineering change The subject in the first The actual revenue amount of each affected node will be executed.

[0038] Furthermore, a dynamic credit score is generated for the service provider. Through multi-dimensional data fusion, the dynamic credit score is calculated, along with three core indices. The customer evaluation index formula is as follows: ( For the first Customer evaluation index of each service provider; For the project requester to the first (Comprehensive score of each subject) For the project requester to the first The score for each subject. The formula for the reverse checks and balances evaluation index is as follows: ( For the first The reverse checks and balances evaluation index for each service entity; For the combination excluding the first The number of other entities besides the main entity; For the first The subject to the first The evaluation weight of each subject; For the first The subject to the first Anonymous ratings from individual subjects, with a maximum score of 100. for right Evaluation weights, satisfying , for right Anonymous ratings. The objective performance data index formula is: ( For the first Objective performance data index of each service entity; For the first The weight of each objective indicator; For the first The first subject The quantitative value of the objective indicator), among which The quantitative values ​​are for five objective indicators (payment on time rate, project completion rate, first-pass quality rate, timely evidence submission rate, and rectification completion rate). As the indicator weight, satisfying . (Payment on-time rate), the formula is: ( The number of times the main entity receives payments on time; The total number of payments that the entity should receive. (Completion rate of project schedule), the formula is: ( The actual completion period of the main work; (The construction period for completing the work as stipulated in the main contract). (First-pass yield rate), the formula is: ( The number of times the main work results pass a quality acceptance test; (Total number of quality acceptance tests for the main work results) (Timeliness of evidence submission), the formula is: ( The number of times the main entity submits evidence of performance on time; (Total number of times the subject should submit performance evidence). (Rectification Completion Rate), the formula is: ( The number of times the main entity completes rectification as required; (The total number of times the entity has been required to rectify its practices). The core formula for calculating the dynamic credit score is: ,in This is the historical credit decay factor (the coefficient of influence of historical credit score on current credit score; the closer the value is to 1, the greater the historical credit impact). The credit score for the previous evaluation period (the 1st evaluation period) individual entities (Dynamic credit score at any time), new entity (Industry benchmark credit score). Customer evaluation weighting, To provide a reverse check and balance on the evaluation weights, To ensure objective performance data weighting, and to meet the following requirements. The formula for updating the credit score during the project node evaluation cycle is as follows: ( For the first Evaluation time for each performance node; For the first The subject in the first Customer evaluation index for each node; For the first The subject in the first The reverse checks and balances evaluation index of each node; For the first The subject in the first (Objective performance data index of each node), among which , , For the first The corresponding index for each node. The final credit score formula after project completion is: ( The time for project completion evaluation; The evaluation time for the last performance milestone; For the first Customer evaluation index after the completion of each main project; For the first The reverse checks and balances evaluation index after the completion of each main project; For the first The objective performance data index of each main project after completion is calculated and synchronized to the engineering service resource database to update the credit information of the main entity.

[0039] Furthermore, by embedding dynamic credit scores into the engineering consortium's business flywheel model, credit-driven mechanisms are achieved. Through intelligent recommendation weighting, the revised comprehensive recommendation score formula is as follows: ( This is the overall recommendation score after credit enhancement; This is the original comprehensive recommendation score; This is the credit bonus weighting coefficient; (The average credit score of the main entities within the portfolio), where The average credit score of the subjects within the portfolio (the first) (Dynamic credit score of each entity) Credit bonus weighting coefficient, adjusted according to The recommended rankings are regenerated in descending order. Secondly, financial service pricing is dynamically adjusted; the formula for supply chain loan interest rates is... ,in This is the base loan interest rate for the platform's partner financial institutions. This is the credit premium factor (the interest rate reduction corresponding to each 0.1 increase in credit score), which satisfies... (The minimum loan interest rate set by financial institutions). Platform service fees are dynamically adjusted; the fee formula is as follows: ,in The platform's basic service fee rate is charged as a percentage of the project contract amount. This is the credit discount factor (the percentage reduction in fees corresponding to a 0.1 increase in credit score), which satisfies... (The platform's minimum service fee rate). The self-reinforcing cycle of the business flywheel is reflected through a closed-loop formula: ,in , , The optimization evaluation and performance data generated after the main body obtains high-quality resources ( The customer evaluation index after the main body obtains high-quality resources. The reverse checks and balances evaluation index after the main body obtains high-quality resources. It serves as an objective performance data index for entities that have acquired high-quality resources (and the value is better than the original index), forming a quantitative cycle of high credit score, high-quality resources, better performance, and even higher credit score, thereby realizing the credit-driven operation of the project consortium ecosystem.

[0040] This implementation details the realization of intelligent management of the entire project consortium process. By accurately quantifying demand and matching resources, it improves the scientific and accurate nature of subject selection and combination recommendation; it achieves automated revenue sharing based on smart contracts and revenue sharing matrices, ensuring fair, transparent and efficient payments; and it allows for dynamic adaptation and adjustment to project changes; it integrates multi-dimensional data to generate dynamic credit scores, embedding them into a business flywheel model to form a credit-driven ecological cycle, continuously optimizing the efficiency of consortium collaboration and the industry credit system, and comprehensively improving the intelligence, standardization and ecological level of project cooperation.

[0041] Example 2 describes in detail an intelligent recommendation, revenue sharing, and evaluation system for engineering consortia, used to implement intelligent recommendation, revenue sharing, and evaluation methods for engineering consortia, such as... Figure 3 As shown, specifically: The project requirements acquisition unit receives project requirements information and performs structured parsing. The project consortium construction unit selects candidate service providers from the engineering service resource library and constructs multiple sets of engineering consortium candidate combinations; The fusion evaluation unit calculates the comprehensive recommendation score of candidate combinations and outputs the recommendation results through a three-layer architecture model; The smart contract revenue sharing unit initializes the payment node smart contract and revenue sharing matrix, triggers automatic revenue sharing after verification of performance evidence, and realizes dynamic adjustment of revenue sharing after engineering changes. The multi-dimensional credit assessment unit collects customer evaluations, reverse checks and balances evaluations, and objective performance data to calculate the dynamic credit score of each service provider. The business flywheel model unit embeds dynamic credit scores into the configuration rules of intelligent recommendation, financial pricing, and platform fees, and executes corresponding weight and fee adjustments.

[0042] Furthermore, the project requirements acquisition unit receives project data in various formats, including structured forms, unstructured text, and engineering standard documents. It collects information on project type, construction location, technical parameters, cost budget, and schedule requirements. In accordance with the unified requirements of engineering industry standards and subsequent algorithm calculations of the system, it performs data cleaning, missing value warning, and outlier correction on the collected information, transforming various heterogeneous information into standardized structured data that can be directly used for calculation and matching, thus forming a project requirements dataset. Furthermore, the project consortium construction unit is the core execution unit for accurately matching engineering service providers and building project consortium combinations. It connects to an engineering service resource library, which gathers comprehensive information on the qualifications, technical capabilities, service areas, historical performance, cost pricing, and schedule guarantees of all types of service providers, including design, construction, production and processing, and supervision. The unit uses the standardized project requirement dataset output by the project requirement acquisition unit as its core, and formulates differentiated multi-dimensional screening rules for different types of service providers. It screens the providers in the resource library from dimensions such as qualification matching degree, technical parameter fit, and service area coverage, and selects providers that meet the basic requirements of the project to form a categorized candidate provider pool. Then, according to the requirements of the full role collaboration configuration of the project consortium, cross-category provider combination matching is carried out, and the collaboration compatibility between the providers in terms of qualifications, technology, and region is strictly verified to ensure that each combination has the basic capabilities for project undertaking and collaborative operation, and to build multiple candidate combinations of project consortia that meet the project requirements.

[0043] Furthermore, the fusion evaluation unit is built on a three-layer fusion evaluation architecture, which includes a geographic collaboration index sub-unit, a collaboration network index sub-unit, and a value cost index sub-unit. After each sub-unit independently completes the calculation of its corresponding index, the unit then obtains a comprehensive recommendation score and outputs the recommendation result through normalization and weighted fusion.

[0044] The geographic synergy index sub-unit automatically extracts relevant data such as the geospatial coordinates, local service load, and real-time traffic conditions of all service entities within each candidate combination of project co-contractors. According to the preset calculation logic, it quantifies and calculates the geographic synergy index, which reflects the overall spatial synergy capability of the combination, from dimensions such as the response efficiency of the service entity to the project site, the internal coordination space cost between entities, and the sufficiency of local service resources.

[0045] The Collaborative Network Index sub-unit retrieves all historical cooperation records between service entities within the platform's storage, and performs quantitative analysis from dimensions such as the number of cooperations, cooperation evaluation, problem-solving timeliness, and collaborative network structure to calculate a collaborative network index that reflects the degree of cooperation tacit understanding, the tightness of the collaborative network, and the structural integrity among the entities within the platform.

[0046] The Value Cost Index sub-unit integrates relevant data such as the cost quotations of each component of the candidate portfolio, the industry benchmark cost of the corresponding category, the value-added points of the main body in the portfolio, and the overall risk level of the project. Through professional calculation logic, it quantifies and calculates the Value Cost Index, which reflects the comprehensive competitiveness of the portfolio at the economic level, from dimensions such as cost-effectiveness, value creation capability, and risk hedging capability.

[0047] After the sub-unit completes the index calculation, the fusion evaluation unit will perform unified normalization on the three indices to eliminate the differences in the dimensions between the indices. Then, it will dynamically configure the weight coefficients of each index according to the type characteristics of the engineering project, calculate the comprehensive recommendation score of each candidate combination through weighted fusion, sort all candidate combinations in descending order of comprehensive recommendation score, generate standardized and visualized engineering project recommendation results, and output them to the project requester.

[0048] Furthermore, the smart contract revenue sharing unit automates project performance payments, ensures precise revenue sharing, and dynamically adapts to changes. By deeply integrating blockchain and smart contract technologies, and based on the project's full-process performance nodes and the internal collaboration agreement of the project consortium, the overall project contract is broken down into multiple independently verifiable and quantifiable phased payment nodes. Each payment node completes the initial configuration of a dedicated smart contract, while a corresponding revenue sharing matrix is ​​constructed to clarify the revenue distribution ratio of each service provider at different payment nodes. The smart contract and revenue sharing matrix are precisely bound together and stored on the blockchain, ensuring that all payment and revenue sharing rules are tamper-proof and fully traceable. During project execution, this unit automatically receives standardized performance evidence submitted by various service providers, such as design acceptance documents, project completion reports, and supervision verification documents. It then initiates a multi-party cross-verification process involving the project requester, supervisor, and other relevant parties in the consortium. When the performance evidence passes verification by all relevant parties and fully meets the smart contract's preset trigger conditions, the corresponding payment node's smart contract is automatically triggered. Automated payment distribution is then executed according to the preset rules of the payment distribution matrix, ensuring accurate and synchronized fund transfers among the service providers. If engineering changes occur during project execution, this unit can quickly identify the scope of payment nodes affected by the changes. Based on the engineering change agreement signed and confirmed online by all parties, it automatically calculates and adjusts the contract amount for the affected nodes. Simultaneously, based on the revised rights, responsibilities, and benefits allocation scheme, it dynamically adjusts the payment distribution matrix for the corresponding nodes and synchronously updates the relevant smart contract configurations. This ensures a high degree of matching between the payment distribution rules and the actual situation after the engineering changes, achieving real-time dynamic adjustment and execution of the payment distribution system.

[0049] Furthermore, the multi-dimensional credit assessment unit, encompassing the entire project performance process, automatically triggers a multi-channel, multi-dimensional data collection mechanism upon completion of each performance node and the overall project completion. This mechanism collects customer evaluation data from the project client regarding the overall project consortium and individual service providers in terms of work quality, service level, schedule guarantee, and cost control. It also collects role-specific reverse checks and balances evaluation data formed by actual business collaboration among the various service providers within the project consortium. Simultaneously, it automatically extracts objective performance data from relevant units such as the system's smart contract revenue sharing unit and integrated evaluation unit, including payment on time rate, schedule completion rate, first-pass quality rate, timely submission rate of performance evidence, and rectification completion rate for each service provider. This unit performs professional standardization and quantitative transformation on the collected multi-source and heterogeneous data, converting various subjective evaluation data into evaluation indices with unified dimensions, and objective performance data into performance indices that can be directly used in calculations. Then, through a preset credit rating model, it deeply integrates customer evaluation indices, reverse checks and balances evaluation indices, and objective performance data indices. At the same time, it combines the decay rules of historical credit scores to achieve accurate calculation and real-time updates of dynamic credit scores for each service entity, and synchronizes the updated dynamic credit scores to the engineering service resource library.

[0050] Furthermore, the Business Flywheel Model Unit uses the dynamic credit scores of each service provider calculated by the multi-dimensional credit assessment unit as a core control factor, deeply embedding it into the platform's business rule system. This achieves a deep binding and coordinated adjustment between credit scores and the platform's intelligent recommendations, financial service pricing, and platform service fee setting. The dynamic credit scores are integrated into the comprehensive recommendation score calculation logic of the fusion assessment unit, giving exclusive recommendation weight bonuses to candidate combinations of engineering contracting entities composed of high-credit service providers. This allows high-credit combinations to obtain higher priority in recommendation ranking, significantly increasing their probability of being selected by project clients. Simultaneously, in conjunction with the platform's partner financial institutions, the dynamic credit scores are used as a core basis for pricing financial products such as supply chain loan interest rates and engineering insurance premiums. Service providers with higher credit scores can obtain lower financing costs, effectively reducing their project operating costs. In addition, differentiated platform service fee standards are formulated based on the dynamic credit scores of service providers, offering tiered fee discounts to high-credit service providers, further reducing their platform transaction costs.

[0051] By implementing business rules, high-credit service providers can obtain more high-quality business opportunities, lower overall transaction costs, and better platform resources. This will incentivize service providers to proactively improve their performance capabilities, collaboration levels, and service quality, and continuously optimize their credit status. Further optimization of their credit status will in turn allow them to obtain more high-quality resources, thus forming a self-reinforcing business flywheel cycle.

[0052] This embodiment details how the collaborative operation of functional units enables intelligent management of the entire project consortium, from requirements analysis, combination matching, and intelligent recommendation to automatic revenue sharing, dynamic evaluation, and credit-driven processes. A three-layer integrated evaluation architecture ensures the consortium's combination recommendations, while the smart contract revenue sharing unit automates and makes payment revenue sharing transparent and dynamically adapts to project changes. The multi-dimensional evaluation credit unit constructs a full-dimensional dynamic credit system, and combined with the business flywheel model unit, it forms a deep binding between credit and resource allocation, creating a self-reinforcing credit-driven ecosystem and improving the collaborative efficiency and project management standardization of the project consortium.

[0053] The present invention also provides a computer-readable storage medium, such as Figure 4 As shown, it stores a computer program that, when executed by a processor, implements an intelligent recommendation, revenue sharing, and evaluation method for engineering consortia. This computer-readable storage medium serves as the program carrier for the intelligent recommendation, revenue sharing, and evaluation method for engineering consortia. It can be read and executed by hardware devices with data processing capabilities, such as computers, servers, and industrial control terminals. The computer program stored on it is written using modular programming logic and encapsulates the algorithm logic, data processing rules, unit interaction instructions, and various parameter configuration systems for the entire process management of engineering consortia.

[0054] The storage medium can adopt a common computer-readable storage form, including but not limited to solid-state drives, hard disk drives, USB flash drives, portable storage disks, read-only memory (ROM), random access memory (RAM), etc. It can stably store all the code of computer programs and related configuration files, and support repeated reading, modification and updating of data, and adapt to the operating environment of different hardware platforms.

[0055] Furthermore, when the processor reads and executes the computer program on the storage medium, it can sequentially trigger all steps, including the acquisition and structured parsing of project requirements information, the screening of candidate service entities and the construction of consortium combinations, the three-layer index calculation and comprehensive recommendation score generation of the fusion evaluation model, the initial configuration of payment node smart contracts and revenue sharing matrices, automatic revenue sharing after performance evidence verification and dynamic adjustment of revenue sharing after project changes, multi-dimensional evaluation and performance data collection and dynamic credit score calculation, and the ecological operation of embedding dynamic credit scores into the business flywheel model. This fulfills the technical logic and execution requirements of all methods, including the index calculation, normalization processing and weighted fusion rules of the three-layer fusion evaluation architecture, the dynamic adjustment algorithm of revenue sharing amount and matrix during project changes, the fusion calculation and decay rules of dynamic credit scores, and the configuration and application logic of credit scores in intelligent recommendation, financial pricing, and platform fee rates.

[0056] Furthermore, during execution, the computer program can seamlessly connect with external systems such as engineering service resource databases, blockchain evidence storage systems, payment gateways, and financial institution interfaces to complete data interaction, retrieval, and storage, ensuring the method is implemented in practical applications.

[0057] Based on Embodiment 1 or 2, this embodiment describes in detail the implementation and verification of the intelligent recommendation, revenue sharing, and evaluation methods for engineering consortia. It employs two types of standardized public datasets in the HVAC engineering field to conduct full-process technical testing. Both datasets have passed the certification of the building HVAC engineering information data standard, covering structured data across all dimensions, including HVAC project requirements, service provider qualifications, historical collaborations, performance settlements, and credit evaluations. The testing hardware is an industrial-grade intelligent engineering analysis server (4*Intel Xeon Gold 6430 / 256GB DDR4 / 8TB NVMe SSD), and the software environment is a dedicated algorithm engine for HVAC engineering management (built based on Python 3.10 / Spark 3.4 / TensorFlow 2.15). By comparing the results of this application's methodology with the core indicators of three types of advanced engineering intelligent management technologies (denoted as Tech-A, Tech-B, and Tech-C) already implemented in the industry, where Tech-A is the Fanpu Software Engineering Management System, Tech-B is the Glodon BIM5D Platform, and Tech-C is the DeepSYS Energy-Saving System from Shenzhikong, all test data are actual computational outputs. This verification obtained a dataset of HVAC system construction data, which contains complete structured requirements data for 40 HVAC system construction projects, covering the entire process of HVAC system design, construction, equipment manufacturing, and supervision. It also includes full-dimensional information on 140 candidate service providers in the HVAC field (35 design providers, 45 construction providers, 40 equipment manufacturers, and 20 supervisors, all with HVAC engineering professional qualification certifications). The data covers core dimensions such as the provider's qualification level, technical service capabilities, local service scope, historical collaboration records, performance, and cost pricing system. This application method completes the entire process of demand analysis, consortium combination construction, integrated evaluation, intelligent revenue sharing, and multi-dimensional credit assessment. It quantitatively compares the method with existing advanced technologies of Tech-A, Tech-B, and Tech-C from three core dimensions: recommendation accuracy, revenue sharing efficiency, and credit assessment relevance, to verify the comprehensive technical implementation effect of this application method in the construction scenario of HVAC system.

[0058] This application's method performs a full-dimensional structured analysis of the requirements of 40 project groups within the dataset, generating three optimal engineering contract combinations for each project group. It also statistically analyzes the accuracy of recommendations based on the actual implementation results of HVAC engineering project parties. For the 38 implemented projects, it initializes smart contracts for 156 payment nodes and constructs a revenue-sharing matrix, statistically analyzing revenue-sharing execution efficiency. Furthermore, it calculates dynamic credit scores for the 133 service providers involved in the 38 projects, quantifying the correlation between credit rating results and actual performance using the Pearson correlation coefficient. All indicators are calculated using the average of actual calculations, and the results are shown in Table 1. All indicators in the table are positive (higher values ​​indicate better technical performance), while indicators related to revenue sharing time are negative (lower values ​​indicate better technical performance). The abbreviations for each indicator are: RLR (Recommended Combination Implementation Rate), TPR (Optimal Combination Selection Rate), TCR (Project Completion Rate), CCR (Cost Control Rate), QPR (First-Time Quality Qualification Rate), RST (Revenue Sharing Time at Regular Nodes), CST (Revenue Sharing Time at Change Nodes), SAR (Revenue Sharing Accuracy Rate), and COR (Credit-Performance Correlation Coefficient). The benchmark for calculating the relative optimal technical improvement rate is the optimal value of the corresponding indicator in Tech-A, Tech-B, and Tech-C. A negative revenue sharing time improvement rate indicates a reduction in time consumption, and the larger the negative value, the more significant the efficiency improvement.

[0059] Table 1. Comparison of Key Technical Implementation Indicators in HVAC System Construction Data Set

[0060] Based on the core indicator data of recommendation accuracy in Table 1, a multi-technology trend comparison curve of recommendation accuracy for HVAC system construction projects is constructed, as follows: Figure 5 As shown, multiple smooth curves are used to represent the numerical trends of five indicators—RLR, TPR, TCR, CCR, and QPR—in this application's method and three existing advanced technologies. The curves are distinguished by different line types and colors, and numerical labels are added. The curve trends clearly show that all curves corresponding to this application's method are at their highest points, and the trends are stable with no significant fluctuations. In contrast, the curves corresponding to Tech-A, Tech-B, and Tech-C all show varying degrees of low-level distribution, with the Tech-A curve showing the lowest overall trend. This intuitively demonstrates that this application's method possesses significant and stable technical advantages in all dimensions of recommendation accuracy, and the effect of multi-dimensional fusion evaluation is far superior to the single / dual-dimensional evaluation modes of existing advanced technologies.

[0061] Based on the revenue sharing efficiency index data in Table 1, a multi-technology dual-trend model for revenue sharing efficiency in HVAC system construction projects is constructed, such as... Figure 6As shown, two trend curves are used to represent the changing trends of revenue sharing time (average of regular nodes and changed nodes) and revenue sharing accuracy. The revenue sharing time curve adopts a decreasing line type, while the revenue sharing accuracy curve adopts an increasing line type. The curve trends clearly show that the revenue sharing time curve of the proposed method is at a precipitous low level, while the revenue sharing accuracy curve is at a high and stable level of 98% and 99%, forming an optimal curve combination of low time consumption and high accuracy. In contrast, the revenue sharing time curves of the three existing advanced technologies are all distributed at a high level, while the revenue sharing accuracy curve shows a slight downward trend as the time consumption decreases. This intuitively verifies that the smart contract and revenue sharing matrix system of the proposed method achieves the dual optimization of revenue sharing execution efficiency and accuracy, solving the technical pain point of difficulty in balancing time consumption and accuracy in existing advanced technologies.

[0062] To further quantify the contribution of the three indices in the integrated evaluation model of this application to the accuracy of HVAC system construction project recommendations and the rationality of their weighting, the optimal combination of 38 projects implemented in the dataset was selected. The normalized scores (out of 100) of the Geographic Collaboration Index (G), Collaboration Network Index (N), and Value Cost Index (V), as well as the contribution percentage of each index to the comprehensive recommendation score (S), were statistically analyzed. Simultaneously, the single-dimensional evaluation scores of three existing advanced technologies (Tech-A focusing on qualification matching, Tech-B focusing on cost value, and Tech-C focusing on geographic collaboration) were compared. The results are shown in Table 2. The weighting configuration of this application's method is based on HVAC engineering industry technical specifications and project implementation requirements. The contribution percentage of each index reflects the model's adaptability to the core requirements of HVAC project local response efficiency, multi-entity collaboration synergy, and cost-value balance. The three existing advanced technologies only possess single / dual-dimensional evaluation capabilities and lack a comprehensive recommendation score.

[0063] Table 2. Dataset Fusion Evaluation Index Score and Contribution Ratio for HVAC System Construction

[0064] Based on the fusion evaluation index scores in Table 2, the contribution of the fusion evaluation index for HVAC system construction projects is constructed, such as... Figure 7As shown, a stacked area curve represents the process of forming a comprehensive recommendation score by superimposing the scores of the three indices of the proposed method. Simultaneously, four independent trend curves represent the score changes of the proposed method and three existing advanced technologies on each individual index. The stacked area curve clearly shows that the area formed by the superposition of the three indices of the proposed method is much larger than that of the existing technologies, and the scores of each index are evenly distributed. The independent trend curves show that the proposed method has a significant advantage in the G and N indices, and its V index score is on par with Tech-B. This fully reflects the rationality of the weight configuration of the integrated evaluation model of the proposed method, which takes into account the index weights of the core needs of HVAC projects while achieving a balance in the evaluation of each dimension, avoiding the limitations of the single-dimensional evaluation of existing technologies.

[0065] The average time to complete the structured parsing of the requirements of 40 HVAC system construction projects in the dataset was 0.82 hours, with 100% data completeness. The parsing results can be directly connected to the fusion evaluation unit for joint entity combination matching. In the joint entity combination construction stage, 15-20 high-quality service entities were accurately selected for each project from 140 candidate entities through multi-dimensional screening rules, with a combination construction efficiency of 2.4 groups / hour, which is far higher than the average combination construction efficiency of 0.75 groups / hour of the three existing advanced technologies. During the intelligent revenue sharing phase, the revenue sharing matrices constructed for the 38 implemented projects were highly compatible with the internal collaboration agreements of the HVAC engineering consortium. The smart contracts of the 156 payment nodes were automatically triggered after multi-party verification of performance evidence. Among them, 30 nodes underwent engineering changes due to HVAC system design optimization and equipment model adjustments. After the change agreement was confirmed, the method of this application completed the amount adjustment and revenue sharing matrix update of the affected nodes in an average of 2.18 hours, and simultaneously completed the on-chain storage of smart contracts. All changes in revenue sharing achieved 100% accuracy, with no data deviations or rule conflicts. During the credit assessment phase, the dynamic credit score calculation for 133 service providers integrated the Customer Evaluation Index (CEI), the Reverse Balance Evaluation Index (REI), and the Objective Performance Data Index (ODI). The ODI includes six HVAC engineering-specific performance indicators such as payment on time rate, project completion rate, and first-pass quality rate. The weights of each index were configured according to the industry standard for HVAC engineering credit assessment (α=0.3, β=0.3, γ=0.4). The correlation coefficient between the calculated credit score and the actual performance of the service providers reached 0.90, showing a high positive correlation. This indicates that the credit assessment results can accurately reflect the actual performance capabilities of service providers in the HVAC field, providing reliable credit data support for the subsequent operation of the business flywheel model.

[0066] To further verify the adaptability of the proposed method to the needs of HVAC engineering renovation scenarios, the stability of revenue sharing dynamic adjustment in high-frequency local engineering change scenarios, and the credit-driven implementation effect of the business flywheel model in the HVAC engineering sub-sector ecosystem, this verification obtained a dataset of HVAC fresh air system renovation. This dataset is a related sub-dataset of HVAC system construction, containing complete structured demand data for 35 HVAC fresh air system renovation projects, covering the entire process of fresh air system design optimization, equipment replacement, construction renovation, and supervision and acceptance. It is accompanied by full-dimensional information of 120 candidate service providers in the HVAC field (30 design, 40 construction, 35 equipment manufacturing and processing, and 15 supervision), of which 65% of the service providers have HVAC system renovation special service qualifications. The data covers core dimensions such as renovation project-specific technical capabilities, equipment adaptability, local construction experience, and historical collaboration records of renovation projects. By implementing the method described in this application, the entire process of technical execution is completed. From three core dimensions—adaptability to specific scenarios, stability of change-based revenue sharing, and credit ecosystem-driven approach—it is quantitatively compared with three existing advanced technologies: Tech-A, Tech-B, and Tech-C. This verifies the technical implementation effect and scenario adaptability of the method in specific HVAC engineering renovation scenarios.

[0067] Based on the detailed needs of the HVAC system renovation projects in the dataset, the weight configuration of the integrated evaluation model was dynamically adjusted (increasing the weight of same-city renovation service response in the G index, the weight of renovation collaboration between design, construction and equipment in the N index, and the weight of renovation technology matching degree and equipment adaptability in the V index), generating 3 optimal engineering co-contract combinations for each of the 35 projects; for the 34 projects that were implemented, the smart contracts of 126 payment nodes were initialized, of which 52 nodes underwent engineering changes due to partial design adjustments of the ventilation system and equipment model adaptation optimization, and the stability of change-based revenue sharing indicators were statistically analyzed; dynamic credit scores were calculated for the 119 service providers involved in the 34 projects, and the credit scores were embedded into the business flywheel model to achieve deep binding with intelligent recommendation, financial pricing, and platform fees, and the driving indicators of the credit ecosystem were statistically analyzed. All indicators were taken as the actual calculated average, and the results are shown in Table 3. The abbreviations for each indicator are as follows: PSA (Professional Renovation Entity Matching Rate), PLR (Project Performance Efficiency), CIR (Change Node Identification Rate), CAR (Change Adjustment Accuracy Rate), CFR (Change Revenue Sharing Smoothness, the higher the value, the more stable the execution), RPP (High Credit Entity Recommendation Probability Increase Rate), CPR (High Credit Entity Cooperation Repurchase Rate), FCR (High Credit Entity Financial Cost Reduction Rate), and SFR (High Credit Entity Platform Fee Rate Reduction Rate). All indicators are positive indicators. High credit entities refer to HVAC service entities with a credit score of ≥80 points (out of 100) calculated by this application method. Project performance efficiency is the weighted comprehensive value of the three indicators of schedule, cost, and quality. Change revenue sharing smoothness is the percentage of change revenue sharing execution without any delays, setbacks, or errors.

[0068] Table 3 Comparison of Core Implementation Indicators for Multiple Technologies in HVAC Fresh Air System Retrofit Data Set

[0069] Based on the full set of core indicator data in Table 3, a comprehensive trend of multiple technical core indicators for HVAC fresh air system renovation projects is constructed, such as... Figure 8 As shown, three core trend curves represent the numerical changes of three primary indicators: adaptability to specific scenarios, stability of change-based revenue sharing, and credit ecosystem driving force. Each curve integrates the weighted average of its corresponding secondary indicators, using differentiated gradient colors and line types for distinction. Numerical labels and trend slope indicators are added to key nodes of the curves. The curve trends clearly show that the three curves corresponding to the proposed method exhibit a steep upward trend and are all at their highest values. The curves for adaptability to specific scenarios and stability of change-based revenue sharing approach 100, while the curve for credit ecosystem driving force shows explosive growth compared to existing technologies. In contrast, the three curves corresponding to Tech-A, Tech-B, and Tech-C show a gentle upward trend and are at low values. The curve for credit ecosystem driving force is the gentlest, directly demonstrating that the proposed method possesses stronger scenario adaptability and change processing stability in HVAC engineering subdivision renovation scenarios, especially achieving a leapfrog breakthrough over existing advanced technologies in terms of credit ecosystem driving force.

[0070] For the four core indicators of credit ecosystem driving force in Table 3, a multi-technology growth curve for the credit ecosystem driving force of HVAC fresh air system renovation projects is constructed, such as... Figure 9 As shown, multiple exponential growth curves are used to represent the numerical changes of the proposed method and three existing advanced technologies in four indicators: RPP, CPR, FCR, and SFR. The curves use color gradients and node markers to highlight the growth rate of each indicator of the proposed method. The curve trends clearly show that all curves corresponding to the proposed method exhibit high-speed exponential growth, with all indicator values ​​significantly higher than those of existing technologies. Furthermore, the growth trends of the four curves are balanced, with no obvious weaknesses. In contrast, the curves corresponding to Tech-A, Tech-B, and Tech-C all exhibit slow linear growth, with significant numerical differences. Among these, the FCR and SFR curves show the slowest growth. This directly verifies that the proposed method's business flywheel model can achieve a deep integration of credit scores with the HVAC engineering sub-ecosystem. Through the linkage of credit with intelligent recommendation, financial pricing, and platform fees, an efficient credit-driven growth cycle is formed. Existing advanced technologies, lacking a complete credit linkage mechanism, cannot achieve this kind of exponential ecosystem-driven effect.

[0071] To address the segmented requirements of HVAC system renovation projects in the dataset, a dedicated HVAC renovation engineering analysis module was added during the requirements analysis phase. This module achieved 100% completeness in analyzing specific indicators such as the compatibility of the fresh air system with the existing HVAC system, localized renovation construction techniques, and equipment replacement standards, with an analysis time of 0.98 hours, fully aligning with industry technical specifications for HVAC renovation projects. In the fusion evaluation model weight configuration phase, the weights of the Geographic Collaboration Index (G) were adjusted to 30%, the Collaborative Network Index (N) to 40%, and the Value Cost Index (V) to 30%. The sub-weights of renovation technology matching and equipment compatibility within the V index accounted for 65%, ensuring that the recommended combinations met the professional requirements of HVAC system renovation projects. Ultimately, the professional renovation subject matching rate for the 34 implemented projects reached 94.12%, significantly higher than the three existing advanced technologies, fully validating the accuracy and adaptability of the proposed method in segmented HVAC engineering renovation scenarios. During the change-based revenue sharing phase, 46 engineering changes occurred across 34 projects, involving 52 payment nodes. The proposed method achieved a 100% recognition rate for all change nodes, completing node amount adjustments and revenue sharing matrix updates in an average of 1.75 hours. The accuracy rates for change adjustment and revenue sharing execution were both 100%, with a smoothness of 99.04% and no stuttering, delays, or rule conflicts. In contrast, the three existing advanced technologies lack a dedicated change-based revenue sharing rule library for HVAC renovation projects, resulting in issues such as missing change node identification and deviations in adjustment ratios. This verifies the stability of the proposed method in dynamic revenue sharing adjustments under high-frequency local change scenarios in HVAC engineering. In the credit-driven ecosystem phase, this application method deeply embeds the dynamic credit scores of 119 service providers into three major business rules: intelligent recommendation, financial pricing, and platform fees. It provides high-credit providers with exclusive benefits such as increased recommendation weight, reduced financial costs, and reduced platform fees. In subsequent HVAC system renovation projects, high-credit providers saw a 49.02% increase in recommendation probability, a 65.69% repurchase rate, a 34.15% decrease in supply chain loan financial costs, and a 54.12% decrease in platform service fees. After obtaining these benefits, high-credit providers further improved their HVAC renovation project performance efficiency, with an average credit score increase of 8.8 points, forming a self-reinforcing business flywheel cycle. This achieved a virtuous cycle of credit-driven industry ecosystem development in the HVAC engineering sub-sector of HVAC system renovation. In contrast, three existing advanced technologies, lacking a complete linkage mechanism between credit and business rules, cannot achieve this kind of ecosystem-driven effect.

[0072] Through full-process implementation verification of two types of standardized certification datasets in the field of HVAC engineering—HVAC system construction dataset and HVAC fresh air system renovation dataset—and combined with quantitative comparison with three types of existing advanced intelligent management technologies for HVAC engineering (Tech-A, Tech-B, and Tech-C), the proposed method demonstrates significant technical advantages in six core dimensions: recommendation accuracy, revenue sharing execution efficiency, credit assessment relevance, adaptability to specific scenarios, stability of revenue sharing changes, and credit ecosystem driving force. Compared with the existing best advanced technologies, the recommended combination implementation rate is increased by 8.57%, revenue sharing time is reduced by 69.05%, the credit-performance correlation coefficient is increased by 13.92%, the professional segmented subject matching rate is increased by 14.30%, and the core indicators driven by the credit ecosystem are increased by more than 72%. All calculation results are actual values ​​with low data dispersion, high stability, and are authentic and reliable.

[0073] The proposed method can dynamically adjust the weight configuration of the fusion evaluation model and the adjustment rules of the revenue sharing matrix according to the needs of different sub-scenarios in HVAC engineering. It has shown good scenario adaptability in scenarios such as HVAC system construction and HVAC fresh air system renovation. At the same time, the smart contract revenue sharing system constructed by the proposed method has achieved high-efficiency and zero-error revenue sharing execution in HVAC engineering. The dynamic credit rating system can accurately reflect the actual performance capability of service providers in the HVAC field. The business flywheel model realizes the deep binding of credit score with the HVAC engineering ecosystem, forming a self-reinforcing credit-driven cycle.

[0074] To further verify the universality and reliability of the proposed method in large-scale construction projects, especially its adaptability to steel structure and reinforced concrete building construction scenarios requiring steel structure professional qualifications or Class A general contracting qualifications, this verification uses a standardized dataset for steel structure and reinforced concrete building construction to conduct full-process technical testing. This dataset includes complete structured requirements data for 42 steel structure and reinforced concrete building construction projects, covering different types of projects such as high-rise office buildings, large industrial plants, and public venues, encompassing the core requirements of the entire process from design, steel structure fabrication, main construction, to supervision and acceptance. It also includes comprehensive information on 156 candidate service providers with corresponding qualifications (38 design providers, 42 steel structure fabrication providers, 51 construction providers, and 25 supervision providers). All providers hold Class A professional contracting qualifications for steel structures or Class A general contracting qualifications for construction projects. The data covers core dimensions such as qualification level, steel structure fabrication accuracy, high-altitude operation capabilities, historical collaboration records, performance, and cost pricing systems. The test hardware used an industrial-grade engineering intelligent analysis server (4*Intel Xeon Platinum 8470C / 512GBDDR5 / 16TB NVMe SSD), and the software environment was a dedicated algorithm engine for intelligent management of building engineering (built on Python 3.11 / Spark 3.5 / TensorFlow 2.16). The core indicators of the proposed method were quantitatively compared with those of three mainstream advanced engineering intelligent management technologies in the industry (Tech-A: Fanpu Software Engineering Management System, Tech-B: Glodon BIM5D Platform, Tech-C: DeepSYS Deep Energy Saving System) to verify the technical implementation effect of the proposed method in the construction scenario of steel structure and reinforced concrete buildings.

[0075] For the technical characteristics of steel structure and reinforced concrete building construction projects, such as high requirements for steel structure processing precision, high difficulty in high-altitude operation coordination, and strong time sequence of component transportation and installation, the method of this application dynamically adjusts the weight configuration of the fusion evaluation model: the transportation coordination weight between the steel structure processing entity and the project site in the geographical coordination index (G) is increased to 32%, the steel structure installation coordination weight of the design-processing-construction entity in the coordination network index (N) is adjusted to 43%, and the sub-weights of steel structure processing precision matching degree and high-altitude operation safety assurance capability in the value cost index (V) account for 68%, ensuring that the recommended combination accurately adapts to the specific needs of the scenario. For each of the 42 projects, three optimal engineering consortium combinations were generated. Based on the actual implementation results of the projects, the accuracy of the recommendations was statistically analyzed. For the 40 implemented projects, the smart contracts of 182 payment nodes were initialized and the revenue sharing matrix was constructed. Among them, 47 nodes underwent engineering changes due to adjustments in steel structure component models and optimization of installation processes. The stability of revenue sharing due to these changes was statistically analyzed. Dynamic credit scores were calculated for the 148 service providers involved in the 40 projects. The credit scores were embedded into the business flywheel model, and the driving force of the credit ecosystem was statistically analyzed. All indicators were calculated using the average of actual calculations. The results are shown in Table 4 below. The table recommends the following accuracy dimensions: RLR (Recommended Combination Implementation Rate), TPR (Optimal Combination Selection Rate), STR (Steel Structure Processing Accuracy Compliance Rate), PCR (Component Installation Coordination Rate), and QPR (Quality First-Time Pass Rate). The revenue sharing execution efficiency dimension includes RST (Routine Node Revenue Sharing Time), CST (Change Node Revenue Sharing Time), and SAR (Revenue Sharing Accuracy Rate). The credit rating correlation dimension includes COR (Credit-Performance Correlation Coefficient). All indicators are positive (higher values ​​indicate better technical performance), while revenue sharing time-related indicators are negative (lower values ​​indicate better technical performance). The benchmark for calculating the relative optimal technical improvement rate is the optimal value of the corresponding indicator among Tech-A, Tech-B, and Tech-C. A negative revenue sharing time improvement rate indicates reduced time consumption; the larger the negative value, the more significant the efficiency improvement.

[0076] Table 4 Comparison of Core Implementation Indicators for Multiple Technologies in Steel Structure and Reinforced Concrete Building Construction Data Set

[0077] Based on the core indicator data of recommendation accuracy and revenue sharing execution efficiency in Table 4, a comprehensive trend of multiple key technical indicators for steel structure and reinforced concrete building construction projects is constructed, such as... Figure 10As shown, five differentiated curves are used to characterize the changing trends of five core indicators: RLR, STR, PCR, QPR, and SAR. The RLR curve is represented by a solid blue line with a circular marker, the STR curve by a solid red line with a triangle marker, the PCR curve by a dashed green line with a square marker, the QPR curve by a dashed purple line with a diamond marker, and the SAR curve by a dotted line with a pentagram marker. All curves are labeled with numerical values ​​and trend slope indicators. The curve distribution clearly shows that all curves corresponding to the method in this application are at the highest point and have a stable trend. The STR and QPR curves are close to 100, and the SAR curve maintains a perfect score of 100. In contrast, the curves corresponding to Tech-A, Tech-B, and Tech-C are all distributed at lower levels to varying degrees, with larger fluctuations, especially in the STR and PCR indicators. This clearly demonstrates that the method in this application has significant advantages in the accuracy of recommended combinations, the coordination of processing and installation, and the accuracy of revenue sharing in steel structure and reinforced concrete building construction scenarios, far surpassing existing advanced technologies.

[0078] To more intuitively demonstrate the dynamic adjustment mechanism of revenue sharing in engineering change scenarios, this application focuses on the most common type of change in steel structure component models in steel structure and reinforced concrete building construction projects. The process includes: change triggering (submission of component model adjustment application), identification of affected payment nodes (automatic matching of three related nodes involving component processing, transportation, and installation), recalculation of responsibility weights (adjusting the main responsibility weight of steel structure processing from 28% to 33%, and the main responsibility weight of construction from 45% to 40%), dynamic updating of the revenue sharing matrix (synchronously updating the revenue distribution ratio of the three nodes), and smart contract triggering. The complete accounting chain is marked with the corresponding processing time (change identification time 0.12h, responsibility weight calculation time 0.35h, accounting matrix update time 0.28h, contract triggering accounting time 0.1h) and data verification results (100% data verification pass rate for all nodes). The change accounting processing flow of the method in this application forms a closed loop, and the collaboration between each node is smooth. In contrast, the change processing flow of the comparative technology has problems such as node omission and weight calculation lag. This further verifies the stability of the dynamic adjustment of accounting in the high-frequency change scenario of steel structure and steel-concrete building construction projects.

[0079] In actual performance, the method described in this application took an average of 0.76 hours to structurally analyze the requirements of 42 project groups, with 100% data completeness. The analysis results could be directly connected to the fusion evaluation unit for joint project combination matching. The efficiency of joint project combination construction reached 2.8 groups / hour, far exceeding the average combination construction efficiency of 0.82 groups / hour of the three comparative technologies. In the intelligent revenue sharing stage, the revenue sharing matrix constructed for 40 implemented projects achieved 100% matching with the internal collaboration agreement of the steel structure and steel-concrete building joint project. The smart contracts of 182 payment nodes were automatically triggered after multi-party verification of performance evidence. The 47 change nodes completed the amount adjustment and revenue sharing matrix update on an average of 1.85 hours after the agreement was confirmed, and simultaneously completed the blockchain on-chain notarization. The accuracy and smoothness of the change revenue sharing reached 100%, with no data deviation or rule conflict. During the credit assessment phase, the dynamic credit score calculation for 148 service providers integrated the Customer Evaluation Index (CEI), the Reverse Balance Evaluation Index (REI), and the Objective Performance Data Index (ODI). The ODI includes six performance indicators specific to steel structure and reinforced concrete buildings, such as the steel structure processing accuracy compliance rate, high-altitude operation safety rate, and component installation on-time rate. The weights of each index are configured according to the industry standard for credit evaluation of construction projects (α=0.3, β=0.3, γ=0.4). The correlation coefficient between the calculated credit score and the actual performance of the service providers reached 0.92, showing a high positive correlation. After embedding dynamic credit scores into the business flywheel model, the probability of recommending high-credit entities (credit scores ≥ 80) in subsequent steel structure and reinforced concrete building projects increased by 51.36%, the repurchase rate reached 68.42%, the financial cost of supply chain loans decreased by 37.28%, and the platform service fee rate decreased by 56.74%, forming a significant credit-driven ecosystem effect. In contrast, the three comparative technologies, due to the lack of dedicated assessment modules and credit linkage mechanisms for steel structure scenarios, could not achieve the same level of scenario adaptability and ecosystem-driven effect.

[0080] This verification, through full-process testing in a steel structure and reinforced concrete building construction scenario, further demonstrates that the proposed method is not only adaptable to specific scenarios such as HVAC engineering, but also meets the needs of large-scale construction projects requiring specific qualifications. It exhibits significant technical advantages in core dimensions such as recommendation accuracy, change order revenue sharing stability, and credit ecosystem-driven performance. Compared to the existing best-in-class advanced technologies, it improves the implementation rate of recommended combinations, reduces the time spent on change order revenue sharing, and increases the credit-performance correlation coefficient. All computational data exhibits low dispersion, high stability, and is reliable, fully verifying the universality and practicality of the proposed method.

[0081] The above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. For those skilled in the art, the present invention can have various modifications and variations. Any changes, modifications, substitutions, integrations, and parameter changes made to these embodiments within the spirit and principles of the present invention, without departing from the principles and spirit of the present invention, through conventional substitutions or to achieve the same function, fall within the scope of protection of the present invention.

Claims

1. A method for intelligent recommendation, revenue sharing, and evaluation of engineering consortia, characterized in that, include: Obtain project requirements information, which includes at least the project type, construction location, technical parameters, cost budget, and schedule requirements; Based on the project requirements information, candidate service providers in the categories of design, construction, production and processing, and supervision are selected from the engineering service resource library to construct multiple candidate combinations of engineering contracting entities; A fusion evaluation model is constructed, and the comprehensive recommendation score of each candidate combination of the project contracting entity is obtained through the fusion evaluation model. The project contracting entity recommendation results are then sorted in descending order according to the comprehensive recommendation score and output. Based on the internal collaboration agreement of the selected project consortium, the payment node smart contract and the revenue sharing matrix are initialized, wherein the revenue sharing matrix defines the revenue distribution ratio of each service entity in each payment node; Perform project obligations, collect performance evidence verified by multiple parties, and trigger the smart contract execution of the corresponding payment node for automatic revenue distribution; At project performance milestones and upon project completion, we collect customer evaluation data on the engineering consortium and reverse checks and balances evaluation data among the various service providers within the consortium. We then combine this data with objective performance data to generate dynamic credit scores for each service provider. The dynamic credit score is embedded into a pre-built business flywheel model of the engineering consortium to achieve credit-driven operation of the engineering consortium ecosystem.

2. The intelligent recommendation, revenue sharing, and evaluation method for engineering consortia according to claim 1, characterized in that, The fusion evaluation model adopts a three-layer fusion evaluation architecture, including a geographical collaboration index, a collaboration network index, and a value cost index, to calculate the comprehensive recommendation score for each candidate combination of the project contractors, specifically as follows: Geographic Coordination Index Collaborative Network Index Value Cost Index Normalize each value separately to obtain the normalized value. , , ; Through formula Calculate the comprehensive recommendation score ,in For geographical collaboration weighting coefficient, For the weight coefficients of the collaborative network, These are the value cost weighting coefficients, and each weighting coefficient is dynamically configured according to the project type.

3. The intelligent recommendation, revenue sharing, and evaluation method for engineering consortia according to claim 2, characterized in that, The geographical synergy index The formula is: ,in The number of service providers in the project consortium. As the reference time constant, The time normalization coefficient is... For the first Real-time access time from each service provider to the project site For the first The same-city service load factor of each service entity For the first The and the first Real-time availability between service providers.

4. The intelligent recommendation, revenue sharing, and evaluation method for engineering consortia according to claim 2, characterized in that, The formula for the collaborative network index N is: ,in Candidate combinations for engineering consortia Historical average weight of cooperation Candidate combinations The density of collaborative networks, Candidate combinations The maximum path length between nodes. Candidate combinations The average path length between nodes.

5. The intelligent recommendation, revenue sharing, and evaluation method for engineering consortia according to claim 2, characterized in that, The value cost index The formula is: ,in For the first The weight of each cost item For the first The industry benchmark value for each cost item. For the first The price quote for each cost item. For the set of value points, For the first A rating of each value point. For the first The weight index of each value point This is a penalty function based on project risk.

6. The intelligent recommendation, revenue sharing, and evaluation method for engineering consortia according to claim 1, characterized in that, The automatic revenue sharing also includes dynamically adjusting the amounts and revenue sharing matrix of the affected payment nodes when engineering changes occur, specifically: Identify the affected payment nodes corresponding to engineering changes Through formula Calculate the new contract amount for this node. ,in This is the original contract amount for that node. The total amount may be increased or decreased due to project changes. This is the proportionality coefficient by which the workload of this node is affected by the change. Based on the responsibility contribution allocation plan after the project changes, adjust the revenue sharing matrix of the affected payment nodes to ensure that the allocation ratio of each service entity matches the actual rights, responsibilities and benefits.

7. The intelligent recommendation, revenue sharing, and evaluation method for engineering consortia according to claim 1, characterized in that, The generation of dynamic credit scores for each service provider by combining objective performance data includes: Through formula computing service providers At any moment Dynamic credit score ,in Historical credit decay factor For service providers At any moment Credit score, Customer evaluation weighting, To provide a reverse check and balance on the evaluation weights, For objective performance data weighting, Customer rating index As a reverse check and balance evaluation index, This is an objective performance data index.

8. The intelligent recommendation, revenue sharing, and evaluation method for engineering consortia according to claim 7, characterized in that, The business flywheel model for the project consortium includes configuration rules for intelligent recommendation weights, financial service pricing, and platform service rates, specifically: Dynamic credit score The comprehensive recommendation score calculation model, which incorporates core factors into intelligent recommendation, enables combinations of high-credit service providers to receive recommendation weight bonuses. Through formula Determine the service provider Supply chain loan interest rates ,in Based on the basic lending rate, This refers to the credit premium coefficient. Through formula Determine the service provider Platform service fee rate ,in Basic service fee rate, This is the credit discount coefficient.

9. A smart recommendation, revenue sharing, and evaluation system for engineering consortia, characterized in that, include: The project requirements acquisition unit receives project requirements information and performs structured parsing. The project consortium construction unit selects candidate service providers from the engineering service resource library and constructs multiple sets of engineering consortium candidate combinations; The fusion evaluation unit calculates the comprehensive recommendation score of candidate combinations and outputs the recommendation results through a three-layer architecture model; The smart contract revenue sharing unit initializes the payment node smart contract and revenue sharing matrix, triggers automatic revenue sharing after verification of performance evidence, and realizes dynamic adjustment of revenue sharing after engineering changes. The multi-dimensional credit assessment unit collects customer evaluations, reverse checks and balances evaluations, and objective performance data to calculate the dynamic credit score of each service provider. The business flywheel model unit embeds dynamic credit scores into the configuration rules of intelligent recommendation, financial pricing, and platform fees, and executes corresponding weight and fee adjustments.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the intelligent recommendation, revenue sharing, and evaluation method for engineering consortia as described in any one of claims 1 to 8.

Images