Project resource configuration method and system and computer equipment

Abstract

The invention relates to the technical field of project management, and discloses a project resource allocation method, which comprises the following steps: acquiring project resource demand data, resource allocation historical data and multi-project association constraint data; extracting hierarchical resource features through project resource demand data and resource configuration historical data; extracting association evidence features through the multi-project association constraint data; and generating a resource configuration decision basis based on the hierarchical resource features and the associated evidence features. Through the project resource allocation mode, the resource allocation scheme is generated, and through the project resource demand data and the resource allocation historical data, key information is extracted and the decision scheme is generated, the problems of resource redundancy or capability gap and low cooperation efficiency caused by manual allocation during current project resource allocation are solved; and the resource configuration efficiency, the cost control and the progress management are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of project management technology, specifically to project resource allocation methods, systems, and computer equipment. Background Technology

[0002] With the increasing specialization of labor in society, more and more projects require the participation of multiple parties. For example, integrated investment, construction, and operation projects are often undertaken by consortia composed of entities each specializing in investment, construction, and operation. During project implementation, in accordance with relevant requirements, multiple entities participate and collaborate, inevitably encountering issues related to the allocation and coordination of multiple legal entities, tasks, and resources.

[0003] When multiple entities participate in a single project, their capabilities and resources (such as manpower, equipment, materials, and funds) vary. To leverage collective strengths and effectively reduce resource waste, it's crucial to comprehensively assess the capabilities of each participating entity, focusing on project requirements. Traditional manual allocation of resources, given the differing capabilities of each entity, can easily lead to resource redundancy or capability gaps, resulting in low collaboration efficiency. Furthermore, compliance requirements and collaborative efficiency must also be considered when allocating project resources. Inappropriate resource allocation can cause problems related to capability, qualifications, and efficiency. Summary of the Invention

[0004] This invention provides a project resource allocation method to solve the problems of resource redundancy or capacity gaps and low collaboration efficiency caused by manual allocation in current project resource allocation.

[0005] Firstly, a method for allocating project resources, the method comprising: The system collects project resource requirement data, historical resource allocation data, and multi-project correlation constraint data. The project resource requirement data includes the qualification data, performance data, and resource allocation data of project suppliers. The historical resource allocation data refers to past construction data stored in supplier databases, expert databases, and bidding databases. The multi-project correlation constraint data refers to the dependency relationships and resource sharing constraints between projects extracted from national standards, national regulations, industry standards, enterprise standards, bidding documents, and contract texts. Hierarchical resource features are extracted from project resource demand data and historical resource allocation data. Hierarchical resource features refer to resource attribute features with a hierarchical structure extracted from project resource demand data and historical resource allocation data based on a hierarchical state space decomposition model, which are used to identify the features of the resource dimension space. Features of related evidence are extracted from multi-project related constraint data. Features of related evidence refer to the features extracted from multi-project related constraint data that reflect the interaction between projects and compliance requirements, and are used to represent the features of the rule dimension space. The resource allocation decision-making basis is generated based on the hierarchical resource characteristics and associated evidence characteristics, including: constructing an evidence chain knowledge model; performing resource allocation reasoning based on the model, and generating a set of resource allocation suggestions.

[0006] The resource allocation method presented in this project, when applied to integrated investment, construction, and operation projects, has significantly improved resource allocation efficiency, cost control, and schedule management. Resource utilization increased by 20%, project costs decreased by 15%, and the tasks for the construction and operation phases were completed ahead of schedule, validating the effectiveness and practicality of this invention.

[0007] In one alternative implementation, multi-project association constraint data supports the fusion collection of structured rule bases and unstructured contract texts.

[0008] In one optional implementation, after the steps of collecting project resource requirement data, historical resource configuration data, and multi-project correlation constraint data, a data preprocessing step is further included, as follows: Data cleaning involves filling in missing values ​​using an interpolation method based on the chain of evidence, combined with the experience of domain experts. Outlier detection employs a density-based local outlier factor algorithm to identify and correct resource consumption anomalies. Data standardization processing; Data fusion maps structured data into evidence chain attributes, while unstructured data is transformed into evidence chain rules through natural language processing.

[0009] In one optional implementation, the step of extracting hierarchical resource features from project resource demand data and historical resource allocation data includes: Construct a full data space architecture dimension space, including project requirement dimension space, undertaking entity dimension space, rule dimension space and resource dimension space; Based on the full data space architecture dimensional space, a subset space is constructed; Construct a hierarchical state-space decomposition model; Based on the aforementioned hierarchical state space decomposition model, a three-level feature structure is constructed using the hierarchical state space decomposition model to extract resource feature vectors.

[0010] In one optional implementation, the step of extracting association evidence features from multi-item association constraint data includes: Parse the associated constraint data, utilize multi-project associated constraint data, and combine structured rules with unstructured contract text to generate multi-source constraint conditions; Generate evidence chain features by transforming multi-source constraints into computable rule feature vectors through logical reasoning, which serve as associated evidence features.

[0011] In one optional implementation, the step of generating resource allocation decision criteria based on the hierarchical resource characteristics and associated evidence characteristics includes: Construct a knowledge model of the chain of evidence; Based on the model, resource allocation reasoning is performed to generate a set of resource allocation suggestions.

[0012] In one optional implementation, the process of constructing the evidence chain knowledge model includes: Unified knowledge is represented as a chain of evidence structure: i = (C i R i， π i， ω i ), where: C i For the first i The set of conditions for a chain of evidence consists of a combination of hierarchical resource characteristics and related evidence characteristics; R i For the corresponding historical cases / rule conclusions; π i ∈[0,1] represents the prior confidence of this chain of evidence, based on case success rate or expert evaluation; ω i ∈[0,1] represents the applicable scope or information source coverage of this evidence chain.

[0013] In one alternative implementation, the condition set C i The construction process includes: Feature normalization and hierarchical resource features and associated evidence features are normalized according to their value range to obtain standard vectors; Candidate condition generation is based on rule template parsing and data-driven analysis of historical cases to generate a candidate condition pool. Initial thresholds are determined for candidate conditions based on their sources, and initial weights are determined using the analytic hierarchy process or historical frequency. Candidate conditions are merged and deredundant according to priority to generate a condition set for each chain of evidence.

[0014] In one optional implementation, the process of performing resource allocation reasoning based on the model to generate a set of resource allocation suggestions includes: Calculate the similarity and marginal reliability of each chain of evidence; Resource configuration suggestions are generated by summarizing all matching evidence chains based on their credibility weights and outputting configuration scheme suggestions.

[0015] Secondly, the present invention provides a project resource allocation system, comprising: The data acquisition unit is used to collect project resource requirement data, historical resource allocation data, and multi-project correlation constraint data. The project resource requirement data includes the qualification data and performance data of project suppliers, as well as resource allocation data. The historical resource allocation data refers to past construction data stored in supplier databases, expert databases, and bidding databases. The multi-project correlation constraint data refers to the dependency relationships and resource sharing constraints between projects extracted from national standards, national regulations, industry standards, enterprise standards, bidding documents, and contract texts. The first extraction unit extracts hierarchical resource features from project resource demand data and historical resource allocation data. Hierarchical resource features refer to resource attribute features with a hierarchical structure extracted from project resource demand data and historical resource allocation data based on a hierarchical state space decomposition model, which are used to identify the features of the resource dimension space. The second extraction unit extracts associated evidence features from multi-project association constraint data. Associated evidence features refer to the features extracted from multi-project association constraint data that reflect the interaction between projects and compliance requirements, and are used to represent the features of the rule dimension space. The generation unit generates resource allocation decision-making criteria based on the hierarchical resource characteristics and associated evidence characteristics, including: constructing an evidence chain knowledge model; performing resource allocation reasoning based on the model; and generating a set of resource allocation suggestions.

[0016] Thirdly, the present invention provides a computer device, comprising: The system includes a memory and a processor, which are interconnected. The memory stores computer instructions, and the processor executes these computer instructions to perform the project resource allocation method described in the first aspect or any of its corresponding embodiments.

[0017] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to perform the project resource allocation method described in the first aspect or any corresponding embodiment thereof.

[0018] Fifthly, the present invention provides a computer program product, including computer instructions for causing a computer to execute the project resource configuration method described in the first aspect or any corresponding embodiment thereof. Attached Figure Description

[0019] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating a resource allocation method according to an embodiment of the present invention; Figure 2 This is another flowchart illustrating the resource allocation method according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the data space structure according to an embodiment of the present invention; Figure 4 This is a structural block diagram of a resource allocation system according to an embodiment of the present invention; Figure 5 This is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] In current integrated investment, construction, and operation projects, projects are often undertaken by a consortium of multiple entities, each with its own strengths in investment, construction, and operation. With multiple entities participating in the same project, each possessing different capabilities and resources, the following factors need to be considered during resource allocation: (1) Multi-entity resource allocation: The members of the consortium have large differences in their legal qualifications, performance, funds, construction resources, etc. Traditional manual allocation is prone to resource redundancy or capacity gap. Therefore, it is necessary to first assess the legal entity's capabilities to determine what resources each legal entity should provide.

[0023] (2) The capabilities of legal entities are difficult to quantify. How to determine a reasonable structured indicator system to evaluate the comprehensive capabilities of legal entities (coordination of human resources, materials and machinery in three dimensions).

[0024] (3) Dynamic resource adaptation, effective allocation of management resources between the construction management level and the project company's investment and operation management level.

[0025] (4) Compliance constraints, including meeting qualification access requirements (different rules apply to different project implementation areas) and other legal and regulatory requirements, as well as contractual restrictions such as consortium agreements and EPC (engineering, procurement, and construction) general contracting contracts, and adjusting resource allocation according to the actual situation of the legal entity; (5) Bottleneck in collaborative efficiency: Traditional allocation mechanisms rely on manual negotiation and are difficult to respond to changes in resource demand in real time.

[0026] According to an embodiment of the present invention, a project resource allocation method is provided, which is applied in a project resource allocation system. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0027] The project resource allocation method in this embodiment mainly includes the following steps: S1. Collect project resource requirement data, historical resource allocation data, and multi-project correlation constraint data; wherein, the project resource requirement data includes the qualification data, performance data, and resource allocation data of project suppliers; the historical resource allocation data refers to past construction data stored in the supplier database, expert database, and bidding database; the multi-project correlation constraint data refers to the dependency relationships and resource sharing constraints between projects extracted based on national standards, national regulations, industry standards, enterprise standards, bidding documents, and contract texts. S2. Extract hierarchical resource features from project resource demand data and historical resource allocation data. Hierarchical resource features refer to resource attribute features with a hierarchical structure extracted from project resource demand data and historical resource allocation data based on a hierarchical state space decomposition model. These features are used to identify the characteristics of the resource dimension space. S3. Extracting correlation evidence features from multi-project correlation constraint data; Correlation evidence features refer to the features extracted from multi-project correlation constraint data that reflect the interaction between projects and compliance requirements, and are used to represent the features of the rule dimension space. S4. Generate resource allocation decision-making basis based on the hierarchical resource characteristics and associated evidence characteristics, including: constructing an evidence chain knowledge model; performing resource allocation reasoning based on the model, and generating a set of resource allocation suggestions.

[0028] By using the above-mentioned project resource allocation methods, a resource allocation plan is generated. By extracting key information from project resource demand data and historical resource allocation data, the optimal plan is obtained, which solves the problems of resource redundancy or capacity gaps and low collaboration efficiency caused by manual allocation in current project resource allocation.

[0029] like Figure 1The flowchart shown illustrates a project resource allocation method. This method allocates project resources based on hierarchical association evidence reasoning and can be used on computer devices, such as computer terminals or servers, as well as mobile terminals, such as mobile phones and tablets. The method includes the following steps: Step S100: Collect project resource requirement data, historical resource allocation data, and multi-project correlation constraint data.

[0030] The above data can be collected through a data acquisition unit, which is used to collect project resource requirement data, historical resource configuration data, and multi-project related constraint data.

[0031] For example, the project resource requirement data X={x1,x2,...,x} is obtained through the data acquisition unit. t}; Where x t ∈R d Let R be the d-dimensional resource requirement index at time t, where d refers to the dimension data. The resource requirement index R refers to the standard and quantity of human, financial, and mechanical resources required by the project construction from various legal entities when the project reaches time t.

[0032] The data acquisition supports the integration of multi-source heterogeneous data. The system is configured with an incremental acquisition interface, which can obtain resource status change data in real time and automatically verify the logical consistency of the data based on the Petri net rule engine.

[0033] The project resource requirements data include: supplier qualification data, performance data, and resource allocation data. Qualification data refers to industry qualifications issued by government administrative departments, such as a Class A general contracting qualification for water conservancy and hydropower construction; performance data refers to historical data provided by suppliers that are similar in scale, industry, and construction content to this project; and resource allocation data refers to the personnel, materials, and machinery data provided by suppliers based on the project requirements.

[0034] The above resource requirements data can be entered by suppliers into the procurement cloud platform. This cloud platform can be an information management system that can store information about the entire life cycle of a project (the whole process) by using technologies such as GIS, the Internet, IoT sensing, and 3D digitization. It can establish an information ecosystem for construction projects that is interconnected, collaborative, information-sharing, construction-monitoring, and scientifically managed, and assist management in realizing intelligent control and decision analysis of engineering construction.

[0035] Historical resource allocation data supports dynamic mapping across project case libraries. This historical data refers to past construction data stored in the supplier database, expert database, and bidding database, collected through the procurement cloud platform.

[0036] Multi-project correlation constraint data supports the fusion collection of structured rule bases and unstructured contract texts. Multi-project correlation constraint data refers to data such as dependencies and resource sharing constraints between projects extracted from national standards, national regulations, industry standards, enterprise standards, bidding documents, and contract texts.

[0037] The project resource requirement data, historical resource allocation data, and multi-project correlation constraint data collected above may contain missing, duplicate, abnormal, redundant, or irregular data during the data collection process. Therefore, data preprocessing is necessary. The data preprocessing process includes: (1) Data cleaning: missing values ​​are filled using an interpolation method based on the chain of evidence, combined with the experience of domain experts.

[0038] (2) Outlier detection: The density-based Local Outlier Factor (LOF) algorithm is used to identify and correct resource consumption anomalies.

[0039] (3) Data standardization: Z-score standardization is used for continuous data, and one-hot encoding is used for discrete data.

[0040] (4) Data fusion: Structured data is mapped to evidence chain attributes. Structured data in project resource demand data and resource allocation history data are directly mapped to evidence chain attributes (such as "material availability" corresponding to inventory data). Unstructured data is transformed into evidence chain rules through natural language processing. Contract texts and expert experience are also transformed into evidence chain rules through natural language processing, such as extracting "compensation clauses for construction delays" from contracts as constraints.

[0041] Ultimately, a comprehensive data space for integrated investment, construction, and operation projects will be built.

[0042] This step collects project resource demand data, historical resource allocation data, and multi-project correlation constraint data in the manner described above. Through preprocessing, the data is transformed into analyzable data, forming a complete data space for the integrated investment, construction, and operation project, which facilitates subsequent processing.

[0043] Step S200: Extract hierarchical resource characteristics using project resource demand data and historical resource allocation data.

[0044] Hierarchical resource features are extracted through a feature extraction unit connected to the data acquisition unit, which is used to extract hierarchical resource features from the acquired data.

[0045] Hierarchical resource characteristics refer to the hierarchical resource attribute features extracted from project resource demand data and historical resource allocation data based on a hierarchical state space decomposition model. These features are used to identify the characteristics of the resource dimension space, i.e., the resource dimension in the investment, construction, and operation full data space architecture. This dimension focuses on the attributes and states of the resources themselves, and through hierarchical features, it achieves a structured description of resource demand, supply, and historical allocation patterns.

[0046] The process of extracting hierarchical resource characteristics from project resource demand data and historical resource allocation data is as follows, see below. Figure 2 As shown.

[0047] Step S201: Construct the full data space architecture of the integrated investment, construction and operation project. Z is an N-dimensional vector space structure model used to describe the full data space of the integrated investment, construction and operation project, which mainly includes the project demand dimension space, the undertaking entity dimension space, the rule dimension space and the resource dimension space.

[0048] Z can be represented as ;in, These are the project requirements (entity) dimension, the undertaking entity (entity) dimension, the rules and constraints (entity) dimension, and the resource attributes (entity) dimension.

[0049] S202, based on the Z-space, further constructs a subset space. It is a vector space structure model of feature subsets that exists in any dimension of the full data space of an N-dimensional integrated investment, construction and operation project.

[0050] Each subset sei corresponds to a specific feature domain under a certain dimension. The attribute space is defined as follows:

[0051] in for The corresponding set of attribute vectors; For set A single attribute vector in.

[0052] Each attribute component is weighted using the Analytic Hierarchy Process (AHP) based on domain expert experience and historical data to enhance the model's credibility and interpretability. Meanwhile, to ensure the scalability and adaptability of the attribute space, a dynamic attribute management mechanism can be employed to adjust and optimize attribute vector combinations in real time based on project progress and data updates.

[0053] When constructing the subset space, the method of dividing the feature subsets can be dynamically adjusted according to the complexity of the project and the scale of the data. For small projects, a broader feature subset division can be adopted, such as dividing the project requirement dimension into subsets such as time requirements, quality requirements, and cost requirements. For large and complex projects, the feature subsets can be further refined, such as subdividing time requirements into time node requirements for each stage and total project duration requirements. At the same time, in order to improve the accuracy and efficiency of subset space construction, the experience and knowledge of domain experts can be introduced, and data mining algorithms can be used to screen and determine the feature subsets.

[0054] The above data space structure is as follows Figure 3 As shown: Step S203: Construct a hierarchical state-space decomposition model ; Where Se v For the resource demand value range space; Se d The project portfolio requirements set includes specific task requirements for each phase; the industry standard knowledge base Se c Standards and best practices in the storage field. Se represents the total decision-making experience of domain experts in this type of project. i This represents the i-th decision scheme composed of the attribute characteristics in the expert's decision-making experience.

[0055] Resource demand range space middle: V 1: Time resource value range, used to specify the minimum-maximum allowable range of schedule-related indicators (e.g., minimum time for stage nodes, maximum total schedule); V 2: Cost resource value range, used to define the upper and lower limits of budgets, procurement costs, etc. (e.g., budget ceiling, material procurement price fluctuation range); V 3: Quality resource value range, used to define the effective value range of construction quality and acceptance standards (such as construction quality indicators and acceptance standards). k Number the subspaces. K This represents the number of dimensions for total resource requirements.

[0056] The values ​​V1, V2, and V3 mentioned above are used to define the range of values, which are used to describe the legal intervals for each type of resource demand indicator. 。

[0057] In the resource demand index vector X 1:T ={x1,x2,...,x T},x t ∈R d In, each xt Corresponding to Se v The specific range of values ​​within a subspace of the expression. Example: If x... t If the "schedule requirement" indicator is included, then Se v v in the "time subspace" k min v is the minimum time for a stage node. k max This represents the maximum total construction period.

[0058] Step S204: Combining the above structural model, a hierarchical state-space decomposition model is adopted to construct a three-level feature structure and extract resource feature vectors. φ =Φ(X, H rec )∈R p ， in φ The data from the S100 preprocessing includes five key resource indicators: manpower, materials, equipment, funds, and schedule. Each indicator has undergone missing data imputation, normalization, or standardization. These are the resource input dimensions used in subsequent inference models. For example, representative fields selected based on project type and management needs include: human resource skill matching (assessing the fit between manpower and task requirements), material resource availability (real-time status of material inventory, procurement cycle, etc.), and cash flow stability (fluctuation analysis of historical fund allocation records). Resource indicators of the same type are aggregated, weighted (weights derived from expert experience and AHP assessment results), and normalized. The resulting resource feature vector is the hierarchical resource feature.

[0059] Specific process: (1) Constructing the basic feature layer of resources Input data: Project resource requirements Historical data on resource allocation (Hrec).

[0060] After performing missing value imputation, outlier correction, standardization (Z-score), and discrete variable one-hot encoding on all original fields, we obtain:

[0061] For example: =Number of personnel on site; =Skill level numericalization; =Materials inventory level...... (2) Divide the subspace according to the state-space model Using the subset space definition of S203: Human Resources Subspace Equipment resource subspace Materials Resources Subspace Capital Resources Subspace Schedule and duration subspace Each subspace corresponds to a certain range V under Sev. k Put the first level They are assigned to these subspaces:

[0062] ...... Features within a subspace need to be normalized to the corresponding range V. k (V1, V2, and V3 as defined in S203), select core features, and assign weights based on AHP (or expert weights) to form subspace feature vectors:

[0064] in: Let j be the feature set of the j-th subspace; It is a weight matrix (derived from expert experience or historical stability assessment).

[0065] (3) Generate the comprehensive layer: construct resource feature vectors φ : φ =Φ( X,Hrec =Concat

[0066] Concat(·) represents the feature concatenation function, which is used to concatenate sub-resource feature vectors from different dimensions in sequence to form a unified comprehensive resource feature representation.

[0067] Step S300: Extract association evidence features from multi-project association constraint data. Association evidence features refer to the features extracted from multi-project association constraint data that reflect the interactions between projects and compliance requirements. These features are used to represent the rule dimension space (Sr), i.e., the rule dimension in the investment, construction, and operation full data space architecture (Z). This dimension focuses on external constraints and relationships during project implementation, ensuring that resource allocation complies with laws, regulations, contractual agreements, and industry standards.

[0068] Step S301 Parses the association constraint data: This step utilizes multi-project association constraint data, combined with structured rules and unstructured contract text, to generate constraint relationships of evidence chain features for compliance verification and collaborative efficiency optimization, reflecting the external constraint relationships in the rule space (Sr).

[0069] Specifically, this includes the parsing of rules and contract texts. The parsing content includes, but is not limited to, the following two data sources: (1) a structured rule base, including industry standards, regional qualification requirements, environmental protection regulations, and internal management systems of enterprises; and (2) unstructured contract texts, including EPC general contracting contracts, consortium agreements, tender documents, and penalty clauses. Through natural language processing (NLP) and a rule extraction engine, unstructured texts are transformed into structured constraint items. For example: compliance constraints: regional qualification requirements and environmental protection standards. Collaborative efficiency constraints: cross-unit resource scheduling processes and task dependencies. Multi-source constraints are formed through the above (1) and (2).

[0070] S302 generates evidence chain features: transforming multi-source constraints into computable rule feature vectors through logical reasoning. ψ =Ψ(rule base, contract text)∈R q ,in ψ It can include metrics such as the "Compliance-Efficiency Balance Index": a comprehensive assessment of whether resource allocation simultaneously meets qualification requirements and cross-project collaboration efficiency. "Contractual Constraint Satisfaction": quantifiable indicators are generated based on contractual restrictions on resource allocation (such as compensation clauses for project delays).

[0071] The system adopts a modular design, with each unit interacting with each other through standard interfaces; the data acquisition unit can connect to various data sources, including project management software APIs, databases, and file systems; the system as a whole adopts a layered architecture, which facilitates subsequent expansion and maintenance.

[0072] Step S400: Generate resource allocation decision basis based on the hierarchical resource characteristics and associated evidence characteristics.

[0073] This section describes how a hierarchical correlation evidence reasoning engine generates resource allocation decision-making criteria based on the hierarchical resource features and correlated evidence features. The hierarchical correlation evidence reasoning engine is a key tool for generating resource allocation decision-making criteria; it operates based on hierarchical resource features and correlated evidence features.

[0074] The specific process of generating resource allocation decision-making criteria based on the hierarchical resource features and associated evidence features through the hierarchical association evidence reasoning engine includes: Step S401: Construct a knowledge model of the chain of evidence To achieve resource allocation reasoning across data sources and under multiple constraints, it is first necessary to construct an evidence chain knowledge representation model.

[0075] The knowledge representation structure adopts a unified format to represent knowledge from different sources, including: (1) logical rules (IF-THEN form): derived from industry standards and contract terms; (2) case sequences: derived from historical project configuration records; and (3) real-time observation information: derived from current project input features.

[0076] The current status of the project is represented by the query vector: q=[ φ ; ψ ] Unified knowledge is represented as a chain of evidence structure: i = (C i R i， π i， ω i ), where: C i ∈R p+q For the first i The set of conditions for a chain of evidence (consisting of a combination of hierarchical resource features and related evidence features); R i For the corresponding historical cases / rule conclusions; π i ∈[0,1] represents the prior confidence of this chain of evidence, based on case success rate or expert evaluation; ω i ∈[0,1] represents the applicable scope or information source coverage of the evidence chain. Each evidence chain can be viewed as a "knowledge path" used to provide candidate solutions and trust weights during reasoning.

[0077] In order to transform the hierarchical resource features φ and the associated evidence features ψ into a condition set C that can be used for reasoning. i This step includes four sub-steps: feature normalization, candidate condition generation, threshold / weight parameterization, and condition merging and simplification. First, φ and ψ are normalized according to the value range Vk to obtain a standardized vector; then, the following (1) and (2) are used in parallel to generate the condition candidate pool. (1) Parsing based on rule templates (automatically generating compliance and contract terms from structured rule base and contract text) (2) Data-driven mining based on historical cases (extracting highly significant conditional patterns from historical successful / failed solutions) Determine the initial threshold for candidate conditions based on their source. (Contract / standard adoption is used directly; historical data is analyzed using quantiles or statistical significance thresholds), and initial weights are determined using the Analytic Hierarchy Process (AHP) or historical frequency. Finally, candidate conditions are merged and deredundant according to priority (compliance > contract > history) to generate a condition set Ci for each chain of evidence. For ease of numerical reasoning, Ci is transformed into a condition satisfaction vector. (Each component is the normalized satisfaction of the condition multiplied by its weight), and this vector is then used for similarity matching of the evidence chain, marginal confidence update, and confidence weighted summation.

[0078] The hierarchical resource features φ and associated evidence features ψ are mapped to a unified condition representation space. Through rule template / vectorization, threshold segmentation, and weighted aggregation, a condition set Ci for each evidence chain is generated, defined as a set of several Boolean or numerical conditions.

[0079] Each of these conditions It can be represented as a triple or a quad:

[0080] in: The evidence term being examined (which can be a single component of φ or a single component of ψ, or a combination of both). For comparison operators (e.g., ≥, ≤, ∈, ==, ... (match, etc.) For thresholds or rule sets (numerical thresholds, intervals, or text matching templates); This represents the relative weight of the condition in Ci (used for weighted similarity / confidence calculation).

[0081] Example: c=(\text{human_match},\ge,0.8,0.7) means "human matching degree ≥ 0.8", with a weight of 0.7.

[0082] The relationship between this step and the process in step S402 is that, using C... i The transformed condition-satisfied vector Calculate the similarity between the initial match score and the condition vector of the current query q. Adjust the match score based on historical success rates and expert evaluations. (Prior reliability), marginal reliability is updated based on the current matching degree (see S402). (Update mechanism). The final output is a set of configuration suggestions, calculated by a weighted summary of all matching chains of evidence.

[0083] Step S402: Perform resource allocation reasoning based on the FUER model Based on the knowledge model, a FUER (Fused Uncertainty Evidence Reasoning) model is used for dynamic decision generation. This model integrates similarity calculation, reliability update, and ranking mechanisms to generate a set of resource allocation suggestions.

[0084] S402a: Similarity Calculation and Marginal Reliability Update For each chain of evidence Calculate the similarity and marginal confidence between the current project and its feature q. The feature vector of the current project is represented as: q = [φ; ψ]. Where φ is the hierarchical resource feature vector obtained in S204. ψ is the associated evidence feature vector obtained in S300. The condition set of the evidence chain Ci is numerically processed to form a condition vector. Similarity Using cosine similarity:

[0085] The formula for calculating marginal reliability is:

[0086] in: The prior confidence of the chain of evidence is based on the success rate of historical cases or expert evaluation. α, β, η: are the weights for the applicability of the evidence chain; α, β, η: are the fusion coefficients for similarity, prior confidence, and applicability, respectively. The marginal confidence level of the evidence chain is used to rank the credibility of resource allocation schemes.

[0087] S402b: The system generates resource allocation recommendations based on the marginal confidence of each chain of evidence. For each chain of evidence... The marginal reliability calculated using step S402a The resource allocation conclusions provided by this chain of evidence Weighted summaries are performed to obtain the overall resource allocation score:

[0088] in, This indicates the credibility of the chain of evidence in the current project context. This indicates the resource allocation scheme or parameters corresponding to the evidence chain. To obtain the final recommended resource allocation scheme, the system normalizes all weighted results:

[0089] The resulting R represents the optimal resource allocation scheme for the current project. It can output the top K suboptimal schemes in three categories: "most feasible (maximum Score)," "most economical (minimum cost subspace objective)," and "most compliant (highest constraint satisfaction)." For example, it can be based on the most feasible configuration (maximum confidence), the most economical configuration (minimum cost subspace), or the most compliant configuration (highest contract fit).

[0090] By implementing this invention, the integrated investment, construction, and operation project achieved significant improvements in resource allocation efficiency, cost control, and schedule management. Resource utilization increased by 20%, project costs decreased by 15%, and the tasks of the construction and operation phases were completed ahead of schedule, verifying the effectiveness and practicality of the invention.

[0091] Finally, the resource allocation plan was adjusted, the task execution order was optimized, resource checks were set for key time nodes, and the final solution was formed.

[0092] The integrated project resource allocation method in this embodiment can solve resource shortage problems in the short term without affecting project progress; medium-term effects include a 20% increase in resource utilization efficiency and a 15% reduction in project costs; long-term effects include the solution being solidified as an organizational-level process asset for subsequent project management optimization. This solution fully demonstrates the application process and effects of this method in a real-world project. Through systematic data collection, feature extraction, resource allocation, and solution generation, it successfully addresses the risk of project resource shortages and establishes reusable project management optimization experience. Particularly in handling complex multi-project interrelationship constraints, this method demonstrates strong practical value.

[0093] Taking a large-scale urban rail transit integrated investment, construction and operation project as an example, the project involves the construction of multiple stations, the operation of the line and the development of related supporting facilities, and lasts for up to 10 years.

[0094] The actual application of the system includes: Data acquisition: Collecting data on project requirements, responsible entities, rules, and resources from multiple data sources such as the project management system, financial system, and equipment management system to ensure the comprehensiveness and accuracy of the data. Feature extraction: Based on the aforementioned integrated investment, construction, and operation project data space architecture, the collected data is divided into dimensions and subset spaces are constructed to extract features from dimensions such as project requirements, responsible entities, rules, and resources, forming a hierarchical feature structure. Resource allocation decision-making: Based on the extracted features, using the associated evidence chain reasoning model and the evidence chain reasoning mixed integer optimization model, combined with the actual situation and objectives of the project, a reasonable resource allocation plan is formulated. For example, based on the features of the project requirement dimension, the capital investment and equipment configuration at each stage are determined; based on the features of the responsible entity dimension, the rights, responsibilities, and interests of all parties are coordinated; based on the features of the rule dimension, relevant resource allocation rules and constraints are followed; and based on the features of the resource dimension, the allocation and utilization of resources are optimized.

[0095] Ultimately, the system successfully optimized resource allocation for the project, improving construction efficiency and operational effectiveness while reducing resource waste and costs. This example demonstrates the system's value in large-scale real-world projects, particularly its advantages in resource allocation and decision-making for complex projects. The system's modular design and scalability also allow for flexible configuration and adjustment based on the characteristics of different projects.

[0096] This embodiment provides a project resource allocation system, such as Figure 4 As shown, it includes: Data acquisition unit 401 is used to collect project resource requirement data, historical resource configuration data, and multi-project correlation constraint data; The first extraction unit 402 is used to extract hierarchical resource features through project resource demand data and historical resource allocation data; The second extraction unit 403 extracts features of related evidence through multi-item association constraint data; The generation unit 404 is used to generate resource allocation decision basis based on the hierarchical resource characteristics and associated evidence characteristics.

[0097] The project resource allocation system in this embodiment adopts a modular design, with each unit interacting with each other through standard interfaces; the data acquisition unit can connect to various data sources, including project management software APIs, databases, and file systems; the system as a whole adopts a layered architecture, which facilitates subsequent expansion and maintenance.

[0098] The project resource requirement data includes the project supplier's qualification data, performance data, and resource allocation data; The historical data of resource allocation refers to the past construction data stored in the supplier database, expert database, and bidding database; The multi-project association constraint data refers to the dependency relationships and resource sharing constraints between projects extracted from national standards, national regulations, industry standards, enterprise standards, bidding documents, and contract texts; the multi-project association constraint data supports the fusion collection of structured rule bases and unstructured contract texts.

[0099] In an optional implementation, the above-mentioned project resource allocation system further includes a data preprocessing unit, comprising: The data cleaning subunit uses an evidence chain-based interpolation method to fill in missing values, combined with domain expert experience; outlier detection uses a density-based local outlier factor algorithm to identify and correct resource consumption anomalies. Data standardization processing unit; The data fusion unit maps structured data to evidence chain attributes, and transforms unstructured data into evidence chain rules through natural language processing.

[0100] In one optional implementation, the first feature extraction unit includes: The first sub-unit is used to construct the full data space architecture dimension space, including the project requirement dimension space, the undertaking entity dimension space, the rule dimension space, and the resource dimension space. The second sub-unit is used to construct a subset space based on the full data space architecture dimension space; The third sub-unit is used to construct a hierarchical state-space decomposition model; Extract sub-units, which are used to combine the hierarchical state space decomposition model, construct a three-level feature structure, and extract resource feature vectors.

[0101] In one optional implementation, the second feature extraction unit includes: The constraint generation subunit is used to parse associated constraint data, and use multi-item associated constraint data, combined with structured rules and unstructured contract text, to generate multi-source constraint conditions. The feature generation subunit is used to generate evidence chain features. Through logical reasoning, it transforms multi-source constraints into computable rule feature vectors, which serve as associated evidence features.

[0102] In one optional implementation, the above-mentioned generating subunit includes: The first construction subunit is used to construct an evidence chain knowledge model; the process of constructing the evidence chain knowledge model includes: Unified knowledge is represented as a chain of evidence structure: ε i= (C i R i， π i， ω i ), where: C i For the first i The set of conditions for a chain of evidence consists of a combination of hierarchical resource characteristics and related evidence characteristics; R i For the corresponding historical cases / rule conclusions; π i ∈[0,1] represents the prior confidence of this chain of evidence, based on case success rate or expert evaluation; ω i ∈[0,1] represents the applicable scope or information source coverage of this evidence chain.

[0103] A set generation subunit is used for resource allocation reasoning based on the model to generate a set of resource allocation suggestions. This unit includes: The computational subunit is used to calculate the similarity and marginal confidence of each chain of evidence; The suggestion sub-unit is used to generate resource configuration suggestions. It summarizes all matching evidence chains by weighted reliability and outputs configuration scheme suggestions.

[0104] Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0105] In this embodiment, the project resource allocation system is presented in the form of functional units. Here, a unit refers to an ASIC (Application Specific Integrated Circuit), a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0106] This invention also provides a computer device having the above-described features. Figure 5 The project resource allocation system is shown. Please refer to [link / reference]. Figure 5 , Figure 5 This is a schematic diagram of the structure of a computer device provided in an optional embodiment of the present invention, such as... Figure 5 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 5 Take a processor 10 as an example.

[0107] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0108] The memory 20 stores instructions executable by at least one processor 10 to cause the at least one processor 10 to perform the method shown in the above embodiments.

[0109] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0110] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0111] The computer device also includes an input device 30 and an output device 40. The processor 10, memory 20, input device 30, and output device 40 can be connected via a bus or other means. Figure 5 Taking the example of a connection between China and Israel via a bus.

[0112] Input device 30 can receive input numerical or character information, and generate key signal inputs related to user settings and function control of the computer device, such as a touchscreen, keypad, mouse, trackpad, touchpad, joystick, one or more mouse buttons, trackball, joystick, etc. Output device 40 may include display devices, auxiliary lighting devices (e.g., LEDs), and haptic feedback devices (e.g., vibration motors). The aforementioned display devices include, but are not limited to, liquid crystal displays, light-emitting diodes, displays, and plasma displays. In some alternative embodiments, the display device may be a touchscreen.

[0113] The computer device also includes a communication interface 30 for communicating with other devices or communication networks.

[0114] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0115] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0116] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for allocating project resources, characterized in that, The method includes: The system collects project resource requirement data, historical resource allocation data, and multi-project correlation constraint data. The project resource requirement data includes the qualification data, performance data, and resource allocation data of project suppliers. The historical resource allocation data refers to past construction data stored in supplier databases, expert databases, and bidding databases. The multi-project correlation constraint data refers to the dependency relationships and resource sharing constraints between projects extracted from national standards, national regulations, industry standards, enterprise standards, bidding documents, and contract texts. Hierarchical resource features are extracted from project resource demand data and historical resource allocation data. Hierarchical resource features refer to resource attribute features with a hierarchical structure extracted from project resource demand data and historical resource allocation data based on a hierarchical state space decomposition model, which are used to identify the features of the resource dimension space. Features of related evidence are extracted from multi-project related constraint data. Features of related evidence refer to the features extracted from multi-project related constraint data that reflect the interaction between projects and compliance requirements, and are used to represent the features of the rule dimension space. The resource allocation decision-making basis is generated based on the hierarchical resource characteristics and associated evidence characteristics, including: constructing an evidence chain knowledge model; performing resource allocation reasoning based on the model, and generating a set of resource allocation suggestions.

2. The method according to claim 1, characterized in that, Multi-project association constraint data supports the fusion collection of structured rule bases and unstructured contract texts.

3. The method according to claim 2, characterized in that, Following the steps of collecting project resource demand data, historical resource allocation data, and multi-project correlation constraint data, a data preprocessing step is also included, as follows: Data cleaning involves filling in missing values ​​using an interpolation method based on the chain of evidence, combined with the experience of domain experts. Outlier detection employs a density-based local outlier factor algorithm to identify and correct resource consumption anomalies. Data standardization processing; Data fusion maps structured data into evidence chain attributes, while unstructured data is transformed into evidence chain rules through natural language processing.

4. The method according to claim 1, 2, or 3, characterized in that, The process of extracting hierarchical resource characteristics from project resource demand data and historical resource allocation data includes: Construct a full data space architecture dimension space, including project requirement dimension space, undertaking entity dimension space, rule dimension space and resource dimension space; Based on the full data space architecture dimensional space, a subset space is constructed; Construct a hierarchical state-space decomposition model; A hierarchical state-space decomposition model is adopted to construct a three-level feature structure and extract resource feature vectors.

5. The method according to claim 4, characterized in that, The extraction of association evidence features through multi-project association constraint data includes: Parse the associated constraint data, utilize multi-project associated constraint data, and combine structured rules with unstructured contract text to generate multi-source constraint conditions; Generate evidence chain features by transforming multi-source constraints into computable rule feature vectors through logical reasoning, which serve as associated evidence features.

6. The method according to claim 5, characterized in that, The process of constructing the knowledge model of the chain of evidence includes: Unified knowledge is represented as a chain of evidence structure: i = (C i R i， π i， ω i ), where: C i For the first i The set of conditions for a chain of evidence consists of a combination of hierarchical resource characteristics and related evidence characteristics; R i For the corresponding historical cases / rule conclusions; π i ∈[0,1] represents the prior confidence of this chain of evidence, based on case success rate or expert evaluation; ω i ∈[0,1] represents the applicable scope or information source coverage of this evidence chain.

7. The method according to claim 6, characterized in that, The process of constructing the condition set Ci includes: Feature normalization and hierarchical resource features and associated evidence features are normalized according to their value range to obtain standard vectors; Candidate condition generation: Based on rule template parsing and historical cases, a candidate condition pool is generated. Initial thresholds are determined for candidate conditions based on their sources, and initial weights are determined using the analytic hierarchy process or historical frequency. Candidate conditions are merged and deredundant according to priority to generate a condition set for each chain of evidence.

8. The method according to claim 7, characterized in that, The process of performing resource allocation reasoning based on the model to generate a set of resource allocation suggestions includes... Calculate the similarity and marginal reliability of each chain of evidence; Resource configuration suggestions are generated by summarizing all matching evidence chains based on their credibility weights and outputting configuration scheme suggestions.

9. A project resource allocation system, characterized in that, include: The data acquisition unit is used to collect project resource requirement data, historical resource allocation data, and multi-project correlation constraint data. The project resource requirement data includes the qualification data and performance data of project suppliers, as well as resource allocation data. The historical resource allocation data refers to past construction data stored in supplier databases, expert databases, and bidding databases. The multi-project correlation constraint data refers to the dependency relationships and resource sharing constraints between projects extracted from national standards, national regulations, industry standards, enterprise standards, bidding documents, and contract texts. The first extraction unit extracts hierarchical resource features from project resource demand data and historical resource allocation data. Hierarchical resource features refer to resource attribute features with a hierarchical structure extracted from project resource demand data and historical resource allocation data based on a hierarchical state space decomposition model, which are used to identify the features of the resource dimension space. The second extraction unit extracts associated evidence features from multi-project association constraint data. Associated evidence features refer to the features extracted from multi-project association constraint data that reflect the interaction between projects and compliance requirements, and are used to represent the features of the rule dimension space. The generation unit generates resource allocation decision-making criteria based on the hierarchical resource characteristics and associated evidence characteristics, including: constructing an evidence chain knowledge model; performing resource allocation reasoning based on the model; and generating a set of resource allocation suggestions.

10. A computer device, characterized in that, include: A memory and a processor are communicatively connected, the memory stores computer instructions, and the processor executes the project resource allocation method according to any one of claims 1 to 8 by executing the computer instructions.

Images