A multi-end project collaboration evaluation method and system

Abstract

The application provides a multi-end item collaborative evaluation method and system, and relates to the technical field of multi-end collaboration. The application introduces a trust gradient calculation and dynamic weight adjustment mechanism, so that the risk evaluation process can reflect the actual trust state of each collaboration end in real time, thereby guaranteeing the accuracy of the evaluation result at the data source. When a low-trust collaboration end is detected, isolation and weight recalibration are performed to avoid interference of abnormal data on the global risk level, thereby significantly improving the stability and robustness of the evaluation.

Description

Technical Field

[0001] This invention relates to the field of multi-terminal collaboration technology, specifically to a multi-terminal project collaboration evaluation method and system. Background Technology

[0002] In engineering project management, multi-terminal collaboration has become the norm. On-site, mobile, and cloud terminals jointly participate in data collection and decision-making. It mainly relies on cloud computing platforms to collect progress, cost, and quality data by integrating IoT sensors, mobile terminals, and on-site management systems. It also uses big data analysis or machine learning models to generate risk warnings. Typically, multi-source data is integrated using fixed weights or simple averaging.

[0003] Existing risk assessment systems typically assume that data from all collaborating parties has equal credibility, lacking a dynamic identification and differentiated processing mechanism for data source reliability. When data from a collaborating party is distorted due to equipment failure, network instability, or human error, the system still includes it in the calculation, causing deviations in the overall risk assessment results and even leading to misjudgments.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a multi-terminal project collaboration evaluation method and system to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A multi-terminal project collaboration evaluation method, the specific steps of which include: Step 1: Collect data from multiple collaborating parties, including project execution data and collaborative behavior data, to form the first dataset; Step 2: Calculate the trust gradient value of each collaborating end based on the first dataset, and generate the second dataset; Step 3: Dynamically adjust the weights of each collaborating party in the risk assessment based on the second dataset to generate the third dataset; Step 4: Based on the third dataset, merge the risk scores of each collaborating party to generate the fourth dataset; Step 5: Based on the fourth dataset, determine whether there are any collaborating parties with a trust gradient value lower than a preset threshold. If so, generate the fifth dataset. Step 6: Isolate low-trust collaboration endpoints and recalibrate their weights based on the fifth dataset to generate the final risk assessment results.

[0007] Furthermore, the collaboration terminals include on-site terminals, mobile terminals, and cloud terminals, and the project execution data includes progress data, cost data, and quality data.

[0008] Furthermore, the calculation of the trust gradient value is based on factors including historical data accuracy, device operating status, network stability, and collaborative response time.

[0009] Furthermore, the dynamic adjustment process of the weights includes transferring a portion of the weights from the low-trust collaboration endpoint to the high-trust collaboration endpoint.

[0010] Furthermore, the risk scores are fused using a weighted fusion method, and the fusion result is used to characterize the overall risk level.

[0011] Furthermore, the preset threshold is a dynamic value automatically generated by the system based on historical project data.

[0012] Furthermore, the recalibration of weights is performed after the isolation ends, based on the recovery status of the collaborating end.

[0013] This invention also discloses a multi-terminal project collaboration evaluation system for implementing a multi-terminal project collaboration evaluation method, comprising: The data acquisition module is used to collect data from multiple collaborating parties, including project execution data and collaborative behavior data, to form a first dataset; The trust calculation module is used to calculate the trust gradient values ​​of each collaborating end based on the first dataset and generate the second dataset. The weighting adjustment module is used to dynamically adjust the weights of each collaborating party in the risk assessment based on the second dataset, and generate the third dataset. The risk fusion module is used to fuse the risk scores of each collaborating party based on the third dataset to generate a fourth dataset; The anomaly detection module is used to determine whether there are any collaborating ends whose trust gradient values ​​are lower than a preset threshold based on the fourth dataset. If such ends exist, the fifth dataset is generated. The calibration control module is used to isolate low-trust collaboration endpoints and recalibrate their weights based on the fifth dataset, generating the final risk assessment result.

[0014] Compared with the prior art, the beneficial effects of the present invention are: This invention introduces a trust gradient calculation and dynamic weight adjustment mechanism, enabling the risk assessment process to reflect the actual trust status of each collaborating party in real time. This ensures the accuracy of the assessment results at the data source. By performing isolation and weight recalibration when a low-trust collaborating party is detected, it avoids interference from abnormal data on the global risk level, significantly improving the stability and robustness of the assessment. (See attached figures.) Figure 1 This is a schematic diagram of the overall method flow of the present invention. Detailed Implementation

[0015] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments. Example

[0016] Please see Figure 1 The present invention provides a technical solution: A multi-terminal project collaboration evaluation method, the specific steps of which include: Step 1: Collect data from multiple collaborating parties, including project execution data and collaborative behavior data, to form the first dataset; The collaboration platform includes on-site terminals, mobile terminals, and cloud platforms, and the project execution data includes progress data, cost data, and quality data.

[0017] Multi-terminal data acquisition, based on heterogeneous network interfaces and data synchronization protocols, collects data from on-site sensors, manually entered data from mobile terminals, and historical and real-time data from the cloud in parallel. This achieves cross-terminal data aggregation, ensuring that the input sources for risk assessment cover the project site, mobile work environment, and cloud management platform. Project execution data reflects the event status, while collaborative behavior data reflects the manual execution status. Integrating these two data sources improves the comprehensiveness of risk identification, unifying hard execution indicators such as schedule, cost, and quality with soft behavioral indicators such as response timeliness and task closure status into the evaluation system. Through timestamp alignment, missing value imputation, and format normalization, multi-source data is transformed into a unified first dataset, integrating scattered multi-source data into a standardized structure, facilitating subsequent trust gradient calculation and weight adjustment.

[0018] Step 2: Calculate the trust gradient value of each collaborating end based on the first dataset, and generate the second dataset; The calculation of the trust gradient value is based on factors including historical data accuracy, device operating status, network stability, and collaborative response time.

[0019] Historical data accuracy statistics measure the degree of matching between the data and actual results of past projects on the collaborative end. The higher the accuracy, the higher the trust gradient value. Device operating status is assessed by monitoring device health indicators such as CPU load, storage space, and sensor integrity to determine its data acquisition capability. Network stability is evaluated based on indicators such as network latency, packet loss rate, and connection interruption frequency to assess the reliability of data transmission. Collaboration response time records the response time of the collaborative end after receiving a task or warning. The faster the response, the higher the trust gradient value. All indicators are weighted and fused according to preset weights to obtain a comprehensive trust gradient value. Step 3: Based on the second dataset, dynamically adjust the weights of each collaborative end in the risk assessment to generate a third dataset; The dynamic adjustment process of the weights includes transferring a portion of the weights from the low-trust collaboration end to the high-trust collaboration end.

[0020] When the trust gradient drives the weight adjustment, the collaborating end with a higher trust gradient value is assigned a higher weight. When the trust gradient of a collaborating end is lower than a set threshold, its weight is proportionally distributed to the other collaborating ends with a higher threshold, keeping the sum of the overall weights constant. The weight adjustment is carried out in real time as the second dataset is updated, without the need for manual intervention. Based on the trust gradient values ​​of each collaborating end recorded in the second dataset, their influence in the overall risk assessment is redistributed, and the adjusted weight distribution and corresponding collaborating end identifiers are saved, providing directly usable input for subsequent risk fusion.

[0021] Step 4: Based on the third dataset, merge the risk scores of each collaborating party to generate the fourth dataset; The risk scores are fused using a weighted fusion method, and the fusion result is used to characterize the overall risk level.

[0022] Step 5: Based on the fourth dataset, determine whether there are any collaborating parties with a trust gradient value lower than a preset threshold. If so, generate the fifth dataset. The preset threshold is a dynamic value automatically generated by the system based on historical project data.

[0023] The weighted fusion process multiplies the risk scores from each collaborating party by their corresponding weights and then sums them to obtain the fusion result. The weights in the third dataset have been dynamically adjusted based on the trust gradient in step 3 to ensure that higher-trust parties contribute more. The global risk level characterizes the fusion result and maps it to a preset risk level range to intuitively reflect the overall risk status of the project. Using the adjusted weights in the third dataset, the risk scores from each collaborating party are aggregated and calculated, and the fused global risk level, along with its corresponding timestamp and collaborating party identifier information, is saved.

[0024] Step 6: Isolate low-trust collaboration endpoints and recalibrate their weights based on the fifth dataset to generate the final risk assessment results.

[0025] The recalibration of weights is performed after the isolation ends, based on the recovery status of the collaborating end.

[0026] By comparing the trust gradient values ​​in the fifth dataset with a preset threshold, it is determined whether to temporarily exclude collaborating parties from risk assessment. During the isolation period, the device operating status, network stability, and collaboration response time of the collaborating parties are continuously monitored to determine whether the recovery criteria have been met. Weights are then redistributed based on the recovered trust gradient values ​​to ensure that high-trust collaborating parties remain dominant while appropriately restoring the contribution ratio of low-trust collaborating parties. Collaborating parties with trust gradient values ​​below the preset threshold are identified in the fifth dataset and removed from the current risk assessment input set. After the isolation period ends, their weights are adjusted based on their recovery status, such as restored device operating status, improved network stability, and shortened collaboration response time. The recalibrated weights are applied to the latest risk score data, and the global risk assessment result is output.

[0027] This invention also discloses a multi-terminal project collaboration evaluation system for implementing a multi-terminal project collaboration evaluation method, comprising: The data acquisition module is used to collect data from multiple collaborating parties, including project execution data and collaborative behavior data, to form a first dataset; The trust calculation module is used to calculate the trust gradient values ​​of each collaborating end based on the first dataset and generate the second dataset. The weighting adjustment module is used to dynamically adjust the weights of each collaborating party in the risk assessment based on the second dataset, and generate the third dataset. The risk fusion module is used to fuse the risk scores of each collaborating party based on the third dataset to generate a fourth dataset; The anomaly detection module is used to determine whether there are any collaborating ends whose trust gradient values ​​are lower than a preset threshold based on the fourth dataset. If such ends exist, the fifth dataset is generated. The calibration control module is used to isolate low-trust collaboration endpoints and recalibrate their weights based on the fifth dataset, generating the final risk assessment result.

[0028] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0029] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0030] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A multi-terminal project collaboration evaluation method, characterized in that, The specific steps include: S1: Collect data from multiple collaborating parties, including project execution data and collaborative behavior data, to form the first dataset; S2: Calculate the trust gradient value of each collaborating end based on the first dataset, and generate the second dataset; S3: Based on the second dataset, dynamically adjust the weights of each collaborating party in the risk assessment to generate the third dataset; S4: Based on the third dataset, the risk scores of each collaborating party are merged to generate the fourth dataset; S5: Based on the fourth dataset, determine whether there are any collaborating ends with a trust gradient value lower than a preset threshold. If so, generate the fifth dataset. S6: Based on the fifth dataset, the low-trust collaboration ends are isolated and their weights are recalibrated to generate the final risk assessment results.

2. The multi-terminal project collaboration evaluation method according to claim 1, characterized in that: The collaboration platform includes on-site terminals, mobile terminals, and cloud platforms, and the project execution data includes progress data, cost data, and quality data.

3. The multi-terminal project collaboration evaluation method according to claim 1, characterized in that: The calculation of the trust gradient value is based on factors including historical data accuracy, device operating status, network stability, and collaborative response time.

4. The multi-terminal project collaboration evaluation method according to claim 1, characterized in that: The dynamic adjustment process of the weights includes transferring a portion of the weights from the low-trust collaboration end to the high-trust collaboration end.

5. The multi-terminal project collaboration evaluation method according to claim 1, characterized in that: The risk scores are fused using a weighted fusion method, and the fusion result is used to characterize the overall risk level.

6. The multi-terminal project collaboration evaluation method according to claim 1, characterized in that: The preset threshold is a dynamic value automatically generated by the system based on historical project data.

7. The multi-terminal project collaboration evaluation method according to claim 1, characterized in that: The recalibration of weights is performed after the isolation ends, based on the recovery status of the collaborating end.

8. A multi-terminal project collaboration evaluation system, used to implement the multi-terminal project collaboration evaluation method described in claims 1-7, characterized in that, include: The data acquisition module is used to collect data from multiple collaborating parties, including project execution data and collaborative behavior data, to form a first dataset; The trust calculation module is used to calculate the trust gradient values ​​of each collaborating end based on the first dataset and generate the second dataset. The weighting adjustment module is used to dynamically adjust the weights of each collaborating party in the risk assessment based on the second dataset, and generate the third dataset. The risk fusion module is used to fuse the risk scores of each collaborating party based on the third dataset to generate a fourth dataset; The anomaly detection module is used to determine whether there are any collaborating ends whose trust gradient values ​​are lower than a preset threshold based on the fourth dataset. If such ends exist, the fifth dataset is generated. The calibration control module is used to isolate low-trust collaboration endpoints and recalibrate their weights based on the fifth dataset, generating the final risk assessment result.

Images