International engineering cost analysis and management method and system

Abstract

This invention discloses a method and system for international engineering cost analysis and control. The cost analysis and control method includes the following steps: S1, collecting international subcontractor data, material supply chain data, cross-border environmental data, and project engineering data, and standardizing the above data to obtain standardized data; S2, analyzing and evaluating the standardized data to obtain subcontractor cost assessment results, supply chain cost analysis results, and cross-border cost risk analysis results; S3, combining the subcontractor cost assessment results, the supply chain cost analysis results, the cross-border cost risk analysis results, and the standardized data obtained in step S1 with timestamps to package them into a cost control block, and uploading it to the blockchain for notarization. This invention provides a method and system for international engineering cost analysis and control, which achieves accurate control of international engineering costs by collecting and analyzing relevant data.

Description

Technical Field

[0001] This invention belongs to the field of international construction engineering cost management and supply chain optimization technology, specifically involving an international engineering cost analysis and control method and system. Background Technology

[0002] In the field of international construction engineering, projects are characterized by complex cross-border collaboration, large supply chains, and diverse cost influencing factors (such as exchange rate fluctuations, tariff policies, and localization compliance requirements). Traditional cost management methods are no longer sufficient to meet the needs of end-to-end control, and the main technical bottlenecks are as follows: Subcontractor cost control lacks cross-border adaptability: Existing methods are mostly aimed at domestic subcontractors and do not take into account factors such as the differences in the local qualifications of international subcontractors and performance risks (such as price adjustments caused by exchange rate fluctuations). It is difficult to accurately assess the reasonableness of subcontractor quotations and overpayment or performance disputes are likely to occur.

[0003] Insufficient cost coordination in the material supply chain: The international material supply chain involves multiple links such as procurement, transportation, customs clearance, and warehousing. Traditional methods cannot accurately analyze supply chain costs, resulting in a lag in the detection of material cost overrun risks.

[0004] The cross-border cost risk early warning mechanism is rigid: existing early warnings mostly rely on static thresholds (such as the proportion of material price overruns) and do not take into account the dynamic risk factors of international projects (such as exchange rate fluctuations, tariff policy adjustments, and geopolitical impacts), thus failing to achieve early prediction of risks. Summary of the Invention

[0005] This invention discloses an international engineering cost analysis and control method and system. By collecting data on international subcontractors, material supply chains, cross-border environments, and project engineering, and standardizing the above data, the system analyzes the standardized data to obtain subcontractor cost assessment results, supply chain cost analysis results, and cross-border cost risk analysis results, thereby achieving accurate control of international engineering costs.

[0006] This invention discloses a method for international engineering cost analysis and control, comprising the following steps: S1. Collect data on international subcontractors, material supply chains, cross-border environments, and project engineering, and standardize the above data to obtain standardized data. S2. Analyze and evaluate the standardized data to obtain subcontractor cost assessment results, supply chain cost analysis results, and cross-border cost risk analysis results; S3. The subcontractor cost assessment results, the supply chain cost analysis results, the cross-border cost risk analysis results, and the standardized data obtained in step S1 are combined with timestamps and packaged into a cost control block, which is then uploaded to the blockchain for evidence storage.

[0007] A further improvement of the present invention is that the standardization process in step S1 specifically includes the following steps: S11. Currency Unification: All monetary data in the aforementioned international subcontractor data, material supply chain data, cross-border environment data, and project engineering data are converted into the project settlement currency using the real-time exchange rate; S12, Field Unification: Define a core field library for international engineering costs to unify all heterogeneous fields in the international subcontractor data, material supply chain data, cross-border environmental data, and project engineering data; S13. Data cleaning: Use quantile rules to remove extreme outliers from the international subcontractor data, material supply chain data, cross-border environmental data, and project engineering data, and fill in the missing values ​​through cross-border data consistency verification; S14. Feature Engineering: Perform feature engineering on the cleaned data to generate derived features, and combine the cleaned data and the derived features to obtain the standardized data.

[0008] A further improvement of the present invention is that obtaining the subcontractor cost assessment result in step S2 specifically includes the following steps: S211. Establish a subcontractor benchmark database based on the standardized data; S212. Based on the aforementioned subcontractor benchmark database, calculate the three core indicators of the subcontractor and weight them to obtain the subcontractor cost reasonableness score S. sub : S sub = ω1×(1-P price ) + ω2×P qual + ω3×(1-P risk ), in, P price This is a price deviation index, which is the percentage deviation between the subcontractor's current price and the benchmark price statistically analyzed in the subcontractor benchmark database; P qual The qualification matching index is based on the degree of matching between the qualification level, local experience and project requirements of the subcontractors in the subcontractor benchmark database. P risk Performance risk indicators: risk assessment values ​​of the subcontractor's historical performance records, cross-border payment risks, and geopolitical impacts, as statistically analyzed in the subcontractor benchmark database; ω1, ω2, and ω3 correspond to P respectively. pric P qual P risk The weights are ω1+ω2+ω3=1, and the values ​​of ω1, ω2, and ω3 are obtained through training with historical project data; S213. Determination of Subcontractor Cost Assessment Results: When S sub A price ≥ 0.8 is considered reasonable; 0.6 ≤ S sub A performance guarantee is required if the value is less than 0.8; S sub If the value is less than 0.6, reject the subcontractor's offer.

[0009] A further improvement of the present invention is that obtaining the supply chain cost analysis results in step S2 specifically includes the following steps: S221. Extract feature vectors from the structured data, time-series data, and unstructured data of the material supply chain in the standardized data, refine the features through the spatial-channel decoupled attention gating module to obtain spatial attention weights and channel attention weights, and use additive fusion to obtain gating features; S222. The gating features are hierarchically grouped by the adaptive hierarchical feature selection module, the importance of the features of each link in the material supply chain is evaluated, weights are dynamically allocated and weighted summation is performed to obtain the fused features of each link. The links include procurement, transportation, customs clearance, warehousing, transportation and warehousing losses. S223. Based on the above-mentioned fusion characteristics, calculate the cost of each corresponding stage and the total cost C. material : C material = C purchase + C transport + C customs + C storage + C loss in: C purchase Costs in the material supply chain procurement process; C transport Costs associated with cross-border transportation in the material supply chain; C customs Costs related to tariffs and customs clearance in the material supply chain; C storage Costs associated with overseas warehousing in the material supply chain; C loss Costs related to cross-border transportation and warehousing losses in the material supply chain; S224. Conduct cost variance analysis, comparing the costs of each stage with the total cost C. material By comparing each value with the corresponding budget baseline, the corresponding deviation rate ΔC is calculated. Based on the cost of each link in the material supply chain, the total cost, and each deviation rate ΔC, the supply chain cost analysis results are obtained.

[0010] A further improvement of the present invention is that obtaining the cross-border cost risk analysis results in step S2 specifically includes the following steps: S231. Construct risk indicators based on the standardized data, including: Exchange rate risk indicator R exchange : Rate of cost change due to exchange rate fluctuations; Tariff Risk Indicator R customs Increased costs resulting from tariff policy adjustments; Supply chain disruption risk indicator R break Disruption probability based on logistics trajectories and geopolitical risks; Subcontract performance risk indicator R contract The probability of subcontractors delaying contract performance or adjusting their prices; S232. Dynamic Threshold Calculation: Using a rolling statistical method, based on risk indicator data for a set number of recent days, the dynamic risk threshold is calculated using the conditional quantile method. T high = Q_p(S risk | project_type, country, time_window=xd) T mid = Q_q(S risk | project_type, country, time_window=xd) Wherein: T high T represents the high-risk threshold. mid S represents the medium-risk threshold. risk (This is a comprehensive risk score, where p is the quantile of the high-risk threshold and q is the quantile of the medium-risk threshold. Both p and q can be adjusted according to actual requirements, where p>q, and Q is the quantile function. Short-term fluctuations are suppressed by exponential smoothing (smoothing coefficient η=0.3) to obtain the dynamic risk threshold. project_type represents the project type and is used to filter historical risk data of the same category as the current project. country represents the country or country of origin of the project and is used to filter historical risk data under the same national environment. time_window represents the time window. time_window=xd means that the time window is the last x days, that is, the dynamic warning threshold is calculated based on the risk indicator data within x days prior to the current time. x is the set number of days and can be adjusted.) S233. The comprehensive risk score is assessed, and the cross-border cost risk analysis result is obtained based on the assessment result: when S risk >T high At that time, it was judged as high risk; T mid ≤ S risk <T high The risk level was assessed as medium at that time; S risk <T mid It was judged to be low risk at the time.

[0011] This invention discloses a management and control system for implementing international engineering cost analysis and control methods, comprising: A data acquisition module for collecting data on international subcontractors, material supply chains, cross-border environments, and project engineering, and for standardizing this data to obtain standardized data; A subcontractor evaluation module is used to analyze and evaluate the standardized data to obtain subcontractor cost evaluation results. A supply chain analysis module for analyzing and evaluating the standardized data to obtain supply chain cost analysis results; A risk warning module used to analyze and evaluate the standardized data to obtain cross-border cost risk analysis results; A blockchain management module is used to package the subcontractor cost assessment results, the supply chain cost analysis results, the cross-border cost risk analysis results, and the standardized data with timestamps into a cost control block and upload it to the blockchain for evidence storage.

[0012] This invention discloses an international engineering cost analysis and control method and system. By collecting data on international subcontractors, material supply chains, cross-border environments, and project engineering, and standardizing the above data, the system analyzes the standardized data to obtain subcontractor cost assessment results, supply chain cost analysis results, and cross-border cost risk analysis results, thereby achieving accurate control of international engineering costs. Attached Figure Description

[0013] Figure 1 This is an overall flowchart of the method of the present invention; Figure 2 This is a diagram illustrating the subcontractor cost assessment model architecture of the present invention. Figure 3 This is a flowchart of the supply chain cost analysis for this invention; Figure 4 This is a flowchart of the cross-border cost risk analysis process for this invention; Figure 5 This is a schematic diagram illustrating the dynamic risk threshold for cross-border costs in this invention. Figure 6 This is a hardware architecture block diagram of the system of the present invention. Detailed Implementation

[0014] like Figure 1 As shown, an international engineering cost analysis and control method includes the following steps: S1. Collect data on international subcontractors, material supply chains, cross-border environments, and project engineering, and standardize the above data to obtain standardized data. S2. Analyze and evaluate the standardized data to obtain subcontractor cost assessment results, supply chain cost analysis results, and cross-border cost risk analysis results; S3. The subcontractor cost assessment results, the supply chain cost analysis results, the cross-border cost risk analysis results, and the standardized data obtained in step S1 are combined with timestamps and packaged into a cost control block, which is then uploaded to the blockchain for evidence storage.

[0015] Preferably, in this embodiment, according to the international engineering cost analysis and control method of claim 1, the standardization process in step S1 specifically includes the following steps: S11. Currency Unification: All monetary data in the aforementioned international subcontractor data, material supply chain data, cross-border environment data, and project engineering data are converted into the project settlement currency using the real-time exchange rate; S12, Field Unification: Define a core field library for international engineering costs to unify all heterogeneous fields in the international subcontractor data, material supply chain data, cross-border environmental data, and project engineering data; S13. Data cleaning: Use quantile rules to remove extreme outliers from the international subcontractor data, material supply chain data, cross-border environmental data, and project engineering data, and fill in the missing values ​​through cross-border data consistency verification; S14. Feature Engineering: Perform feature engineering on the cleaned data to generate derived features, and combine the cleaned data and the derived features to obtain the standardized data.

[0016] Preferably, in this embodiment, the derived features that feature engineering can obtain include subcontractor performance capability index, cross-border transportation cost volatility of materials, and tariff policy sensitivity coefficient, etc., to provide support for subsequent evaluation by generating derived features.

[0017] like Figure 2 As shown, obtaining the subcontractor cost assessment results in step S2 specifically includes the following steps: S211. Establish a subcontractor benchmark database based on the standardized data; S212. Based on the aforementioned subcontractor benchmark database, calculate and weight the three core indicators of the subcontractor to obtain the subcontractor cost reasonableness score S. sub : S sub = ω1×(1-P price ) + ω2×P qual + ω3×(1-P risk ), in P priceQuotation deviation is the percentage deviation between the subcontractor's current quotation and the benchmark quotation compiled from the subcontractor benchmark database. P qual The qualification matching score is based on the degree of matching between the qualification level, local experience, and project requirements of the subcontractors in the subcontractor benchmark database. P risk Performance risk score: The risk assessment value of the subcontractor's historical performance record, cross-border payment risk, and geopolitical impact, as statistically analyzed by the subcontractor benchmark database; ω1, ω2, and ω3 correspond to P respectively. pric P qual P risk The weights are ω1+ω2+ω3=1, and the values ​​of ω1, ω2, and ω3 are obtained through training with historical project data.

[0018] S213. Determination of Subcontractor Cost Assessment Results: When S sub A price ≥ 0.8 is considered reasonable; 0.6 ≤ S sub A performance guarantee is required if the value is less than 0.8; S sub If the value is less than 0.6, reject the subcontractor's offer.

[0019] like Figure 3 As shown, obtaining the supply chain cost analysis results in step S2 specifically includes the following steps: S221. Extract feature vectors from the structured data, time-series data, and unstructured data of the material supply chain in the standardized data, refine the features through the spatial-channel decoupled attention gating module to obtain spatial attention weights and channel attention weights, and use additive fusion to obtain gating features; S222. The gating features are hierarchically grouped by the adaptive hierarchical feature selection module, the importance of the features of each link in the material supply chain is evaluated, weights are dynamically allocated and weighted summation is performed to obtain the fused features of each link. The links include procurement, transportation, customs clearance, warehousing, transportation and warehousing losses. S223. Based on the above-mentioned fusion characteristics, calculate the cost of each corresponding stage and the total cost C. material : C material = C purchase + C transport + C customs + C storage + C loss in: C purchase Costs in the material supply chain procurement process; C transport Costs associated with cross-border transportation in the material supply chain; C customs Costs related to tariffs and customs clearance in the material supply chain; C storage Costs associated with overseas warehousing in the material supply chain; C loss Costs related to cross-border transportation and warehousing losses in the material supply chain; S224. Conduct cost variance analysis, comparing the costs of each stage with the total cost C. material By comparing each value with the corresponding budget baseline, the corresponding deviation rate ΔC is calculated. Based on the cost of each link in the material supply chain, the total cost, and each deviation rate ΔC, the supply chain cost analysis results are obtained.

[0020] like Figure 4 , Figure 5 As shown, obtaining the cross-border cost risk analysis results in step S2 specifically includes the following steps: S231. Construct risk indicators based on the standardized data, including: Exchange rate risk indicator R exchange : Rate of cost change due to exchange rate fluctuations; Tariff Risk Indicator R customs Increased costs resulting from tariff policy adjustments; Supply chain disruption risk indicator R break Disruption probability based on logistics trajectories and geopolitical risks; Subcontract performance risk indicator R contract The probability of subcontractors delaying contract performance or adjusting their prices; S232. Dynamic Threshold Calculation: Using a rolling statistical method, based on risk indicator data for a set number of recent days, the dynamic risk threshold is calculated using the conditional quantile method. T high = Q_p(S risk | project_type, country, time_window=xd) T mid = Q_q(S risk | project_type, country, time_window=xd) Wherein: T high T represents the high-risk threshold. mid S represents the medium-risk threshold. risk(This is a comprehensive risk score, where p is the quantile of the high-risk threshold and q is the quantile of the medium-risk threshold. Both p and q can be adjusted according to actual requirements, where p>q, and Q is the quantile function. Short-term fluctuations are suppressed by exponential smoothing (smoothing coefficient η=0.3) to obtain the dynamic risk threshold. project_type represents the project type and is used to filter historical risk data of the same category as the current project. country represents the country or country of origin of the project and is used to filter historical risk data under the same national environment. time_window represents the time window. time_window=xd means that the time window is the last x days, that is, the dynamic warning threshold is calculated based on the risk indicator data within x days prior to the current time. x is the set number of days and can be adjusted.) S233. The comprehensive risk score is assessed, and the cross-border cost risk analysis result is obtained based on the assessment result: when S risk >T high At that time, it was judged as high risk; T mid ≤ S risk <T high The risk level was assessed as medium at that time; S risk <T mid It was judged to be low risk at the time.

[0021] In this embodiment, a red alert is triggered when the risk analysis result indicates a high risk; a yellow alert is triggered when the result indicates a medium risk; and no alert is triggered when the result indicates a low risk.

[0022] In this embodiment, after judging the comprehensive risk score, risk attribution analysis is also performed, which includes the following steps: using the improved Shapley Value algorithm, the contribution weight of each risk factor to the cost deviation is calculated, and the root cause of the risk is identified (such as "exchange rate fluctuation contributes 60%", "tariff adjustment contributes 30%", "logistics delay contributes 10%").

[0023] Preferably, in this embodiment, establishing a subcontractor benchmark database based on the standardized data specifically includes the following steps: 1. Build a subcontractor benchmark database: Integrate historical international project subcontractor quotation data, performance data, and cross-border cost data, and establish a hierarchical index (categorized by project type, country of origin, and subcontracting specialty). 2. Hierarchical Weighted Retrieval: An improved hierarchical weighted table retrieval enhancement method (HCRAG) is adopted. The current subcontractor's quote is used as the query, and a cross-database retrieval is performed by combining "professional matching weight," "geographical similarity weight," and "qualification level weight." This aggregates statistical ranges of quotes from similar projects and performance risk samples to form a structured evidence package. In this embodiment, step S3, combining the subcontractor cost assessment results, the supply chain cost analysis results, the cross-border cost risk analysis results, and the standardized data obtained in step S1 with timestamps, packages them into a cost control block and uploads it to the blockchain for evidence storage. Specifically, this includes the following steps: In this embodiment, step S3, which involves combining the subcontractor cost assessment results, the supply chain cost analysis results, the cross-border cost risk analysis results, and the standardized data obtained in step S1 with timestamps to package them into a cost control block and uploading it to the blockchain for evidence storage, specifically includes the following steps: 1. Cost data on the blockchain: Key data such as subcontractor cost assessment results, material supply chain cost analysis results, risk analysis results, and cross-border payment vouchers are packaged into cost control blocks with timestamps and uploaded to the consortium blockchain for notarization to ensure that the data cannot be tampered with; 2. End-to-end traceability: Generate a unique traceability identifier for each cost link, establish an associated link of "subcontractor quotation - material procurement - cross-border transportation - customs clearance and payment - project settlement", and support reverse traceability of cost composition and responsible parties; 3. Closed-loop management: Based on supply chain cost analysis results and risk warnings, generate management strategies. For subcontractors: High-risk subcontractors are required to provide additional performance guarantees, and subcontractors with reasonable quotations are included in the preferred list; For the material supply chain: optimize procurement plans (such as locking in prices in advance and changing to low-cost origins) and adjust transportation plans (such as changing from sea to land transportation to avoid geopolitical risks). To mitigate cross-border risks: Establish an exchange rate hedging mechanism and reserve funds for adjustments to tariff policies; 4. Dynamic updates: Cost data, risk indicators, and early warning thresholds are updated every 15 days, and model weight parameters are optimized quarterly based on project execution data to achieve adaptive adjustments to management strategies.

[0024] like Figure 6 As shown, a management and control system for implementing international engineering cost analysis and control methods includes: A data acquisition module used to collect data on international subcontractors, material supply chains, cross-border environments, and project engineering, and to standardize the above data to obtain standardized data; A subcontractor evaluation module is used to analyze and evaluate the standardized data to obtain subcontractor cost evaluation results. A supply chain analysis module for analyzing and evaluating the standardized data to obtain supply chain cost analysis results; A risk warning module used to analyze and evaluate the standardized data to obtain cross-border cost risk analysis results. A blockchain management module is used to package the subcontractor cost assessment results, the supply chain cost analysis results, the cross-border cost risk analysis results, and standardized data with timestamps into a cost control block and upload it to the blockchain for evidence storage.

[0025] Preferably, this embodiment also includes a visualization module for displaying the collected international subcontractor data, material supply chain data, cross-border environmental data, project engineering data, as well as subcontractor cost assessment results, supply chain cost analysis results, and cross-border cost risk analysis results. In this embodiment, a cost panorama dashboard is used as the visualization module.

[0026] like Figure 6 As shown, the control system in this embodiment includes: an application display layer, a core algorithm layer, and a cross-border data acquisition layer. The application display layer includes a cost overview dashboard, a control strategy output interface, and a role-based display interface. The core algorithm layer includes a subcontractor evaluation module, a risk warning module, a cross-modal fusion model, and a blockchain existence proof. The cross-border data acquisition layer includes: a data acquisition agent and a heterogeneous database interface.

[0027] In one embodiment, a project is a new highway construction project in a certain country. The project settlement currency is US dollars. It involves three international subcontractors (roadbed engineering, pavement engineering, and bridge engineering). The material supply chain covers 12 types of core building materials (steel bars, cement, asphalt, etc.). The cross-border transportation mode is "sea transport + land transport". The target country is subject to risks such as tariff policy adjustments and geopolitical instability.

[0028] The implementation process of this embodiment is as follows: 1. Data Acquisition and Standardization: Through the deployment of cross-border data acquisition agents, data such as subcontractor quotations, material procurement contracts, logistics trajectory GPS data, target country tariff policy documents, and BIM bill of quantities are obtained; all currency data are converted into US dollars and normalized according to the core field library to generate derived features such as "subcontractor performance capability index" and "asphalt cross-border transportation cost volatility".

[0029] 2. Subcontractor Cost Assessment: Using the subcontractor's quote for the roadbed engineering ($12 million) as the query, a stratified weighted search was performed in the benchmark database, matching three similar roadbed subcontracting quotes for African highway projects (statistical range: $11 million-$11.8 million); the quote deviation P was calculated. price =1.7%, Qualification matching score P qual =0.92, performance risk score Prisk =0.08, thus obtaining S sub =0.4×(1-0.017)+0.3×0.92+0.3×(1-0.08)=0.95, which is considered a reasonable price.

[0030] 3. Supply Chain Cost Analysis: Cross-modal fusion analysis was performed on steel reinforcement materials to extract feature vectors from procurement contracts (structured), logistics trajectories (time-series), and customs clearance documents (unstructured). These vectors were then fused using the SC-DAG and AFS modules to obtain supply chain features. The total cost was calculated as follows: Procurement cost $8.5 million + Transportation cost $1.2 million + Customs duty cost $850,000 + Warehousing cost $300,000 + Loss cost $150,000 = $11 million. Compared with the budget baseline $10.5 million, the deviation rate ΔC = 4.8%, identifying transportation cost overrun as the main reason (due to rising freight rates).

[0031] 4. Cross-border cost risk analysis: Calculate the comprehensive risk score S risk =0.88, dynamic threshold T calculated based on data from the past 60 days. high =0.85, T mid =0.70, triggering a red alert; attribution using the Shapley Value algorithm shows that exchange rate fluctuations contributed 55%, tariff policy adjustments contributed 30%, and logistics delays contributed 15%; management strategies were generated: activate exchange rate hedging tools, reserve 10% of tariff reserve funds, and change some material transportation routes.

[0032] In this embodiment, as Figure 5 As shown, S risk A graph showing the relationship between the number of days and the number of days, ranging from 1 to 60 days.

[0033] 5. Blockchain Control and Traceability: The above cost data, assessment results, early warning records, and control strategies are packaged and uploaded to the blockchain to generate a unique traceability identifier. When disputes arise in the later stages of the project due to cost overruns of asphalt materials, the traceability link can be used to quickly locate the cause as to the increase in freight rates during cross-border transportation, and the dispute can be resolved quickly based on the on-chain evidence data.

[0034] (III) Implementation Results Through the application of the method and system of this invention, the subcontractor quotation deviation rate of the overseas highway engineering project was reduced to less than 5%, the material supply chain cost overrun rate was reduced from 12% to 4.8%, the cross-border risk warning response time was shortened from 72 hours to 24 hours, the efficiency of cost dispute handling was improved by 60%, and the overall project cost was reduced by 8% (approximately US$2.8 million), which verified the practicality and effectiveness of the invention.

Claims

1. A method for international engineering cost analysis and control, characterized in that, Includes the following steps: S1. Collect data on international subcontractors, material supply chains, cross-border environments, and project engineering, and standardize the above data to obtain standardized data. S2. Analyze and evaluate the standardized data to obtain subcontractor cost assessment results, supply chain cost analysis results, and cross-border cost risk analysis results; S3. The subcontractor cost assessment results, the supply chain cost analysis results, the cross-border cost risk analysis results, and the standardized data obtained in step S1 are combined with timestamps and packaged into a cost control block, which is then uploaded to the blockchain for evidence storage.

2. The international engineering cost analysis and control method according to claim 1, characterized in that, The standardization process in step S1 specifically includes the following steps: S11. Currency Unification: All monetary data in the aforementioned international subcontractor data, material supply chain data, cross-border environment data, and project engineering data are converted into the project settlement currency using the real-time exchange rate; S12, Field Unification: Define a core field library for international engineering costs to unify all heterogeneous fields in the international subcontractor data, material supply chain data, cross-border environmental data, and project engineering data; S13. Data cleaning: Use quantile rules to remove extreme outliers from the international subcontractor data, material supply chain data, cross-border environmental data, and project engineering data, and fill in the missing values ​​through cross-border data consistency verification; S14. Feature Engineering: Perform feature engineering on the cleaned data to generate derived features, and combine the cleaned data and the derived features to obtain the standardized data.

3. The international engineering cost analysis and control method according to claim 1, characterized in that, Step S2, obtaining the subcontractor cost assessment results, specifically includes the following steps: S211. Establish a subcontractor benchmark database based on the standardized data; S212. Based on the aforementioned subcontractor benchmark database, calculate the three core indicators of the subcontractor and weight them to obtain the subcontractor cost reasonableness score S. sub : S sub = ω1×(1-P price ) + ω2×P qual + ω3×(1-P risk ), in, P price This is a price deviation index, which is the percentage deviation between the subcontractor's current price and the benchmark price statistically analyzed in the subcontractor benchmark database; P qual The qualification matching index is based on the degree of matching between the qualification level, local experience and project requirements of the subcontractors in the subcontractor benchmark database. P risk Performance risk indicators: risk assessment values ​​of the subcontractor's historical performance records, cross-border payment risks, and geopolitical impacts, as statistically analyzed in the subcontractor benchmark database; ω1, ω2, and ω3 correspond to P respectively. pric P qual P risk The weights are ω1+ω2+ω3=1, and the values ​​of ω1, ω2, and ω3 are obtained through training with historical project data; S213. Determination of Subcontractor Cost Assessment Results: When S sub A price ≥ 0.8 is considered reasonable; 0.6 ≤ S sub A performance guarantee is required if the value is less than 0.8; S sub If the value is less than 0.6, reject the subcontractor's offer.

4. The international engineering cost analysis and control method according to claim 1, characterized in that, The specific steps for obtaining the supply chain cost analysis results in step S2 include: S221. Extract feature vectors from the structured data, time-series data, and unstructured data of the material supply chain in the standardized data, refine the features through the spatial-channel decoupled attention gating module to obtain spatial attention weights and channel attention weights, and use additive fusion to obtain gating features; S222. The gating features are hierarchically grouped by the adaptive hierarchical feature selection module, the importance of the features of each link in the material supply chain is evaluated, weights are dynamically allocated and weighted summation is performed to obtain the fused features of each link. The links include procurement, transportation, customs clearance, warehousing, transportation and warehousing losses. S223. Based on the above-mentioned fusion characteristics, calculate the cost of each corresponding stage and the total cost C. material : C material = C purchase + C transport + C customs + C storage + C loss in: C purchase Costs in the material supply chain procurement process; C transport Costs associated with cross-border transportation in the material supply chain; C customs Costs related to tariffs and customs clearance in the material supply chain; C storage Costs associated with overseas warehousing in the material supply chain; C loss Costs related to cross-border transportation and warehousing losses in the material supply chain; S224. Conduct cost variance analysis, comparing the costs of each stage with the total cost C. material By comparing each value with the corresponding budget baseline, the corresponding deviation rate ΔC is calculated. Based on the cost of each link in the material supply chain, the total cost, and each deviation rate ΔC, the supply chain cost analysis results are obtained.

5. The international engineering cost analysis and control method as described in claim 1, characterized in that, Step S2, which yields the results of the cross-border cost risk analysis, specifically includes the following steps: S231. Construct risk indicators based on the standardized data, including: Exchange rate risk indicator R exchange : Rate of cost change due to exchange rate fluctuations; Tariff Risk Indicator R customs Increased costs resulting from tariff policy adjustments; Supply chain disruption risk indicator R break Disruption probability based on logistics trajectories and geopolitical risks; Subcontract performance risk indicator R contract The probability of subcontractors delaying contract performance or adjusting their prices; S232. Dynamic Threshold Calculation: Using a rolling statistical method, based on risk indicator data for a set number of recent days, the dynamic risk threshold is calculated using the conditional quantile method. T high = Q_p(S risk | project_type, country, time_window=x d) T mid = Q_q(S risk | project_type, country, time_window=x d) Wherein: T high T represents the high-risk threshold. mid S represents the medium-risk threshold. risk (This is a comprehensive risk score, where p is the quantile of the high-risk threshold and q is the quantile of the medium-risk threshold. Both p and q can be adjusted according to actual requirements, where p>q, and Q is the quantile function. Short-term fluctuations are suppressed by exponential smoothing (smoothing coefficient η=0.3) to obtain the dynamic risk threshold. project_type represents the project type and is used to filter historical risk data of the same category as the current project. country represents the country or country of origin of the project and is used to filter historical risk data under the same national environment. time_window represents the time window. time_window=xd means that the time window is the last x days, that is, the dynamic warning threshold is calculated based on the risk indicator data within x days prior to the current time. x is the set number of days and can be adjusted.) S233. The comprehensive risk score is assessed, and the cross-border cost risk analysis result is obtained based on the assessment result: when S risk > T high At that time, it was judged as high risk; T mid ≤ S risk < T high The risk level was assessed as medium at that time; S risk < T mid It was judged to be low risk at the time.

6. A control system for implementing the international engineering cost analysis and control method as described in claim 1, characterized in that, include: A data acquisition module used to collect data on international subcontractors, material supply chains, cross-border environments, and project engineering, and to standardize the above data to obtain standardized data; of A subcontractor evaluation module is used to analyze and evaluate the standardized data to obtain subcontractor cost evaluation results. A supply chain analysis module for analyzing and evaluating the standardized data to obtain supply chain cost analysis results; A risk warning module used to analyze and evaluate the standardized data to obtain cross-border cost risk analysis results; A blockchain management module is used to package the subcontractor cost assessment results, the supply chain cost analysis results, the cross-border cost risk analysis results, and the standardized data with timestamps into a cost control block and upload it to the blockchain for evidence storage.

Images